Hydroxy-alkenals from the peroxidation of n-3 and n-6 fatty acids and urinary metabolites.

4-Hydroxy-2E-hexenal (4-HHE) and 4-hydroxy-2E-nonenal (4-HNE) have been characterized as prominent by-products of n-3 and n-6 hydroperoxy derivatives of n-3 and n-6 fatty acids, respectively. We also have characterized the homolog 4-hydroxy-2E,6Z-dodecadienal (4-HDDE) as a specific by-product of the 12-lipoxygenase product of arachidonic acid 12-hydroperoxy-eicosatetraenoate (12-HpETE). The three hydroxy-alkenals have been found in human plasma with 4-HHE being the most prominent followed by 4-HNE. They were found increased in tissues submitted to oxidative stress, according to the fatty acid characteristic of those tissues, e.g., 4-HNE and 4-HDDE in blood platelets and 4-HHE in the retina. We have shown they covalently bind to the primary amine moiety of ethanolamine phospholipids (PE), especially the plasmalogen subclass, with the highest hydrophobic alkenal (4-HDDE) being the most reactive. Their carboxylic acid metabolites, 4-hydroxy-2E-hexenoic acid (4-HHA), 4-hydroxy-2E-nonenoic acid (4-HNA) and 4-hydroxy-2E,6Z-dodecadienoic acid (4-HDDA), respectively, were found in human urine and measured in higher amounts in situations in which oxidative stress has been reported such as aging and diabetes. As reported above with their hydroxy-alkenals precursors, 4-HHA proved to be the most prominent followed by 4-HNA. Altogether, the three hydroxy-alkenals, either in their free form or bound to membrane PE, may be considered as specific markers of lipid peroxidation able to discriminate between n-3 and n-6 fatty acids. This is corroborated by the measurement of their urinary carboxylic acid metabolites.

Fatty acid composition of the rat pineal gland. Dietary modifications.

When compared to brain, the fatty acid composition of the rat pineal gland revealed that the total proportion of n-6 polyunsaturated fatty acids (PUFA) was 2.3-fold higher, whereas the proportion of n-3 fatty acids was similar. Specifically, 20:4(n-6) and 18:2(n-6) were respectively 1.56- and 11.80-fold higher in the pineal than in the brain, while the proportions of 22:6(n-3) were similar in both tissues. In addition, 18:1(n-9) was found 2.15-fold lower in the pineal. Feeding adult rats with fish oil concentrates induced a significant alteration of the polyunsaturated fatty acid composition of the pineal. There was a reciprocal replacement of the n-6 by the n-3 fatty acids. Conversely, in rats fed a n-3 fatty acid-deficient diet (sunflower oil or coconut oil diet), the pineal gland contained reduced proportions of n-3 fatty acids. We conclude that the pineal gland (i) differs from the brain in containing much higher proportions of 18:2(n-6) and from the other tissues for its high proportions of 22:6(n-3) and (ii) is highly sensitive to the n-3 fatty acid diet in contrast to what is known for the brain. These findings are discussed in the context of melatonin biosynthesis, the major hormone of the pineal gland.

Distribution and metabolism of arachidonic and docosahexaenoic acids in rat pineal cells. Effect of norepinephrine.

The time-course incorporation of 10 microM [14C]arachidonic (AA) and docosahexaenoic (DHA) acids into glycerolipids was studied in rat pineal cells. The incorporation of both labeled fatty acids into total lipids was approximately equal, but their distribution profiles among the various cell lipids showed marked differences. The esterification of [14C]DHA in the neutral lipids, triacylglycerols (TAG) and cholesterol esters (CE), was 2-fold higher than that of [14C]AA whereas the opposite could be observed in total phospholipids (PL). The order of incorporation into PL was phosphatidylcholine (PC) > phosphatidylinositol (PI) = phosphatidylethanolamine (PE) for [14C]AA and PC = PE for [14C]DHA, the incorporation of both fatty acids being not detected in phosphatidylserine (PS) and that of DHA not in PI. When using 0.5 microM [3H] fatty acids, the respective distribution patterns resembled that of fatty acids at 10 microM, except for a lower proportion in TAG. The stimulation of 3H-labeled cells by 100 microM norepinephrine induced a 170% increase of basal release of [3H]AA into the medium, while [3H]DHA was virtually not released. However, the analysis of cell labeling revealed that both [3H] fatty acid levels were decreased in PL and increased in TAG. These findings suggest different involvement for AA and DHA in the pineal function. The preferential incorporation of DHA in TAG suggests that TAG might play an important role in the pineal enrichment with DHA. The absence of DHA release after NE stimulation, which however cannot be ascertained, may raise the question of the role of DHA in NE transduction.

Leukotriene B4 level in neutrophils from allergic and healthy subjects stimulated by low concentration of calcium ionophore A23187. Effect of exogenous arachidonic acid and possible endogenous source.

Peripheral blood neutrophils from patients with allergic rhinitis and from normal subjects were incubated for 5 min at 37 degrees C with 0.15 microM calcium ionophore A23187 in the absence or presence of exogenous arachidonic acid (2.5 to 10 microM). In neutrophils from allergic patients, the leukotriene B4 (LTB4) level was significantly increased by exogenous arachidonic acid in a concentration-dependent manner (16.2 +/- 4.2 and 38.1 +/- 6.8 pmol/5 min per 2 X 10(6) cells in the absence and presence of 10 microM arachidonic acid, respectively; P less than 0.005; n = 8). The LTB4 level in neutrophils from healthy subjects was only 0.97 +/- 0.17 pmol/5 min per 2 x 10(6) cells (n = 5) and was not enhanced by exogenous arachidonate. When cells from allergic patients were challenged in the presence of exogenous [1-14C]arachidonic acid, released LTB4 was radiolabeled and the incorporated radioactivity increased with the labeled arachidonate concentration. Labeled LTB4 was never detectable after incubating neutrophils from normal donors with exogenous labeled arachidonate. When neutrophils were incubated with [1-14C]arachidonate for 1 h, the different lipid pools of the two cell populations were labeled but both types of neutrophils produced unlabeled LTB4 in response to ionophore stimulation. The hydrolysis of choline and ethanolamine phospholipids into diacyl-, alkenylacyl- and alkylacyl-species revealed that solely the alkylacyl-subclass of phosphatidylcholine was unlabeled. We conclude (i) that neutrophils from allergic patients stimulated by low ionophore concentration produce more LTB4 than neutrophils from healthy subjects and incorporate exogenous arachidonate, (ii) that endogenous arachidonate converted to LTB4 by the 5-lipoxygenase pathway may provide only from 1-O-alkyl-2-arachidonoyl-glycero-3-phosphocholine.

Correlation between arachidonic acid oxygenation and luminol-induced chemiluminescence in neutrophils: inhibition by diethyldithiocarbamate.

Neutrophils from allergic subjects were hypersensitive to stimulation by low calcium ionophore concentration (0.15 microM), resulting in an increased formation of leukotriene B4 (LTB4), 5S-hydroxy-6,8,11,14-(E,Z,Z,Z)-eicosatetraenoic acid (5-HETE), and other arachidonic acid metabolites through the 5-lipoxygenase pathway. In parallel, luminol-dependent chemiluminescence was also higher in neutrophils from allergic patients at the basal state and after stimulation by calcium ionophore, revealing an enhancement of radical oxygen species and peroxide production. The activity of glutathione peroxidase, the main enzyme responsible for hydroperoxide reduction, was lowered in these cells. Diethyl-dithiocarbamate (DTC) induced a concentration-dependent decrease in chemiluminescence and arachidonic acid metabolism after neutrophil stimulation. These data show that the elevation of arachidonic acid metabolism in neutrophils from allergic patients is strongly correlated with oxidative status. This elevation may be the consequence of an increased cellular hydroperoxide known to activate 5-lipoxygenase (5-LOX) activity and/or an increased arachidonic acid availability, due either to phospholipase A2 (PLA2) activation or inhibition of arachidonate reesterification into phospholipids. Lowering this oxidative status was associated with a concomitant decrease of this metabolism. Our results suggest that the effect of DTC may be the consequence of an inhibition of peroxyl radical and cellular lipid hydroperoxide production. Thus, DTC may modulate arachidonic acid metabolism in neutrophils by modulating the cellular hydroperoxide level.

Increased plasma levels of atrial natriuretic factor, renin activity, and leukotriene C4 in chronic obstructive pulmonary disease.

We studied atrial natriuretic factor (ANF), plasma renin activity (PRA), and plasma levels of leukotrienes (LTs) B4 and C4 in 23 patients with COPD undergoing right cardiac catheterization for suspected pulmonary hypertension. Hemodynamic measurements together with concomitant ANF levels (both in venous and pulmonary artery blood and right atrial and pulmonary artery plasma levels of LTC4 and LTB4, were determined at rest (T0), after 30 min of breathing oxygen (3 L/min) (T1), and after 30 min recovering and breathing air (T2). Patients with effective exacerbation or definitive evidence of left ventricular disease, hypertension, arrhythmias, or vasodilator or diuretic therapy were excluded. Increased levels of ANF, both in peripheral venous blood (117 +/- 65 pg/ml) and the pulmonary artery (153 +/- 75 pg/ml), were found in patients with COPD, with or without pulmonary hypertension. Levels of LTC4 were also significantly increased (366 +/- 406 pg/ml) when compared with our control values. No correlations among ANF, LTC4 values, functional tests, and hemodynamic measurements were found. Brief increased levels of oxygen did not modify ANF or LTC4 plasma levels, either in patients with or without pulmonary hypertension.

Modulation of norepinephrine-stimulated cyclic AMP accumulation in rat pinealocytes by n-3 fatty acids.

This work showed that docosahexaenoic (22:6n-3) and eicosapentaenoic (20:5n-3) acid supplementation for 48 h have opposite effects on the norepinephrine-stimulated cyclic AMP accumulation in rat pinealocytes. We found that 22:6n-3 supplementation of pineal cells, done by increasing specifically 22:6n-3 in phospholipid and triacylglycerol pools, led to inhibition of norepinephrine-stimulated cyclic AMP production whereas 20:5n-3 supplementation, by increasing 20:5n-3, and 22:5n-3 and 22:6n-3 in the same pools, stimulated it. In contrast, direct treatment of pinealocytes with each fatty acid (50 microM) did not affect cyclic AMP production in the presence of (0.1-10 microM) norepinephrine. The results indicate that, using pharmacological agents such as forskolin or prazosin: (a) neither basal nor forskolin-stimulated cyclic AMP levels were modified in fatty acid-supplemented cells compared to control cells; (b) in the presence of 1 microM prazosin, the activation by 20:5n-3 was still effective whereas no additional inhibition of norepinephrine stimulation was observed in 22:6n-3-supplemented cells. Taken together our results suggest that 22:6n-3 or 20:5n-3 supplementation modulates specifically the alpha 1- or beta-adrenoceptors in the rat pineal gland.

Docosahexaenoic acid (cervonic acid) incorporation into different brain regions in the awake rat.

A quantitative method is presented to examine the localization, in individual brain regions of awake rats, of docosahexaenoic acid (22:6 n-3 or cervonic acid), the main polyunsaturated fatty acid of the nervous system together with arachidonic acid. Following the intravenous injection of 10 microCi [14C]22:6 n-3 (around 0.2 mumol/rat). 0.11-0.28% of the initial radioactivity was located in specific brain areas after detection from 10 to 240 min. Brain regional radioactivity determined by quantitative autoradiography indicated that 60 min after injection, [14C]22:6 n-3 concentrations ranged from 13.75 nCi/g of tissue in inferior olive to 5.59 nCi/g in frontal cortex. The results indicate a higher incorporation into the auditory system: inferior colliculus, central cochlear nucleus, lateral lemniscus, into neuroendocrine structures: paraventricular and supraoptic nuclei, and into certain circumventricular organs such as the pineal gland and neurohypophysis. Analysis of the Bligh and Dyer lipid extracts of rat brain revealed that 60 min after injection, 80-85% of the radioactivity was in choline and ethanolamine phosphoglycerides. These observations suggest that intravenous injection of [14C]22:6 n-3 may be used to study the brain lipid compartmental metabolism in vivo in order to visualize alterations of structural lipid components.

Decreased arachidonic acid metabolism in human platelets by autologous neutrophils: possible role of cell adhesion.

The amount of the 12-lipoxygenase and cyclo-oxygenase products, 12(S)-hydroxy-(Z,Z,E,Z)-5,8,10,14-eicosatetraenoic acid (12-HETE) and 12(S)-hydroxy-(E,E,Z)-5,8,10-heptadecatrienoic acid (HHT), in human platelets stimulated by thrombin (0.1 and 2.5 units/ml), was studied in the presence of autologous neutrophils. A decreased formation of both products was induced by unstimulated neutrophils or neutrophils challenged with N-formylmethionyl- leucyl-phenylalanine (0.1 microM) or Ca2+ ionophore A23187 (0.15 microM). The effect of neutrophils was observed only in the presence of Ca2+. 12-HETE and HHT were also produced in platelets stimulated with thrombin in the absence of Ca2+ and/or Mg2+, but their level was not altered by neutrophils. 12(S),20-Dihydroxy-(Z,Z,E,Z)-5,8,10,14-eicosatetraenoic acid (12,20-DHETE), the cytochrome P-450 product from 12-HETE in neutrophils, was hardly detected, and its level did not compensate for the decrease in 12-HETE observed after platelet and neutrophil co-incubation. 5(S),12(S)-Dihydroxy-(E,Z,E,Z)- 6,8,10,14-eicosatetraenoic acid (5(S),12(S)-DHETE), the 5-lipoxygenase product of 12-HETE in neutrophils, was never detectable. In addition, the inhibition of 12-HETE and HHT formations appeared not to be due to degradation or thrombin uptake by neutrophils, nor was the decrease observed when the two cell populations were physically separated. A monoclonal antibody against the human platelet glycoprotein GMP140 (CD62), mediating Ca(2+)-dependent platelet-neutrophil adhesion, mimicked the inhibitory effect of neutrophils in a dose-dependent fashion. Furthermore, the 12-HETE and HHT productions were not affected when platelets were stimulated in the presence of neutrophils previously incubated with sialidase, which removes the sialic acid from a sialyl Lewis(x) structure assumed to be the neutrophil receptor for platelet GMP140. We conclude that the decrease in thrombin-stimulated 12-HETE and HHT formation observed when platelets were co-incubated with autologous neutrophils might be the consequence of platelet-neutrophil adherence, presumably through platelet GMP140.

Specific markers of lipid peroxidation issued from n-3 and n-6 fatty acids.

Several markers of lipid peroxidation are available with different degrees of specificity, from malondialdehyde as a global marker, to F(2)-isoprostane, which is specifically produced from arachidonic acid. Among these, 4-hydroxynonenal is recognized as a breakdown product of fatty acid hydroperoxides, such as 15-hydroperoxy-eicosatetraenoic acid and 13-hydroperoxy-octade cadienoic acid from the n -6 fatty acids. Furthermore, 4-hydroxyhexenal (4-HHE) derives from n -3 fatty acid hydroperoxides. We have recently described the occurrence of 4-hydroxydodecadienal (4-HDDE) from the 12-lipoxygenase product of arachidonic acid 12-hydroperoxy-eicosatetraenoic acid. These three hydroxy-alkenals may be measured in human plasma by GC-MS, but they may partly be generated in the course of sampling, and the relative volatility of 4-HHE makes its measurement quite unreliable. We have successfully characterized and measured the stable oxidized carboxylic acid products from the hydroxy-alkenals 4-HNA, 4-HHA and 4-HDDA in urine. The ratio between 4-HHA and 4-HNA found in the same urinary sample might provide useful information on the location of lipid peroxidation, accounting for the high enrichment of the cerebrovascular system with docosahexaenoic acid, the main n -3 fatty acid in humans.

Preferential incorporation of sn-2 lysoPC DHA over unesterified DHA in the young rat brain.

The uptake and metabolism of [3H]docosahexaenoic acid (DHA) esterified at the sn-2 position of lysophosphatidylcholine (lysoPC DHA) and in the unesterified form, both bound to albumin, was studied in 20-day-old rats. LysoPC DHA was preferentially recovered in the brain (4-5% of the injected radioactivity) over the unesterified form of DHA (0.3-0.4%). Conversely, the lysoPC form was taken up less than or at the same extent as the unesterified form by the liver, heart, and kidney. In the brain, DHA was mainly recovered in phosphatidylethanolamine whether the esterified or the unesterified form was used, although DHA from lysoPC was esterified at the same extent in phosphatidylcholine and phosphatidylethanolamine after 2.5 min. The uptake of labeled palmitic, oleic, linoleic, and arachidonic acids, esterified or not in lysophosphatidylcholine, was also studied in brain, liver, heart, and kidney. Only the brain preferentially incorporated unsaturated (but not saturated) lysoPC, with the uptake increasing with the degree of unsaturation of the fatty acid moiety. These results strongly suggest that the young rat brain specifically utilizes albumin-lysoPC-containing polyunsaturated fatty acids.

Uptake of 12-HETE by human bronchial epithelial cells (HBEC): effects on HBEC cytokine production.

12-HETE, the major lipoxygenase end-product of platelets and macrophages, may be released in contact of bronchial epithelium in inflammatory diseases of the lung. We have studied the outcome of 12-HETE in presence of human bronchial epithelial cells (HBEC). When HBEC were incubated with [3H]12-HETE for 30 minutes, 27.5% of total radioactivity was found in HBEC and 72.5% in supernatants. Unesterified 12-HETE accounted for 22.4% of total radioactivity, 4.5% being recovered in phospholipids, preferentially in phosphatidylcholine and phosphatidylethanolamine. No incorporation in neutral lipids was detected. 72.9% of the incubated radioactivity was recovered in un identified metabolites. As 12-HETE has been shown to modulate the expression and production of various proteins, the consequence of the 12-HETE uptake on the release of GM-CSF and IL8 by HBEC was assessed. HBEC from control subjects were cultured for 24 hours with 12-HETE (10(-9) to 10(-7)M) in the presence or absence of TNF alpha. Detectable amounts of both cytokines were released in the supernatant in basal conditions at 24hr, and TNF alpha increased significantly the release of GM-CSF. 12-HETE at 10(-7)M weakly but significantly decreased the TNF-induced release of GM-CSF from HBEC. Thus the uptake of 12-HETE could affect the epithelial cell function in some situations.

[Antitumor activity of several new pentopyranosyl-chloroethylnitrosoureas]. / Activité antitumorale de quelques nouvelles chloroéthyl-nitrosourées pentopyranosiques.

Six N-(2-chloroethyl)-N-nitrosoureido derivatives of 4-deoxy-DL-pentopyranose have been synthesized and their antitumor activities were evaluated against melanoma MB16 in mice. The nitrosourea 38 was highly active in comparison with the clinically useful CCNU.

Leukotriene B4 levels from stimulated neutrophils from healthy and allergic subjects: effect of platelets and exogenous arachidonic acid.

Leukotriene B4 (LTB4) levels were measured in peripheral blood neutrophils from allergic and healthy donors after stimulation by calcium ionophore A 23187. This level was higher in neutrophils from allergic subjects than in neutrophils from healthy subjects in the presence as well as in the absence of exogenous arachidonic acid. Platelets from allergics increased LTB4 levels from neutrophils from allergics but not levels in those from healthy donors. Moreover, platelets from healthy subjects reduced LTB4 in neutrophils from both groups. These results suggest that biochemical differences exist in neutrophils and platelets from allergics which contribute to changes in arachidonic acid metabolism via the 5-lipoxygenase pathway. In addition, they support the concept that platelets may play an important role in the regulation of neutrophil LTB4 levels, possibly by affecting the 5-lipoxygenase activity during the course of allergic inflammatory reactions.

Preferential transfer of 2-docosahexaenoyl-1-lysophosphatidylcholine through an in vitro blood-brain barrier over unesterified docosahexaenoic acid.

The passage of either unesterified docosahexaenoic acid (DHA) or lysophosphatidylcholine-containing DHA (lysoPC-DHA) through an in vitro model of the blood-brain barrier was investigated. The model was constituted by a brain capillary endothelial cell monolayer set over the medium of an astrocyte culture. Cells were incubated for 4 h with a medium devoid of serum, then the endothelial cell medium was replaced by the same medium containing labeled DHA or lysoPC-DHA and incubations were performed for 2 h. DHA uptake by cells and its transfer to the lower medium (astrocyte medium when they were present) were measured. When the lower medium from preincubation and astrocytes were maintained during incubation, the passage of lysoPC-DHA was higher than that of unesterified DHA. The passage of both forms decreased when astrocytes were removed. The preference for lysoPC-DHA was not seen when the lower medium from preincubation was replaced by fresh medium, and was reversed when albumin was added to the lower medium. A preferential lysoPC-DHA passage also occurred after 2 h with brain endothelial cells cultured without astrocytes but not with aortic endothelial cells cultured and incubated under the same conditions. Altogether, these results suggest that the blood-brain barrier cells released components favoring the DHA transfer and exhibit a preference for lysoPC-DHA.

Effect of an n-3 fatty acid-deficient diet on the adenosine-dependent melatonin release in cultured rat pineal.

We studied the effect of a diet deficient in n-3 fatty acids on the adenosine-dependent melatonin release from cultured rat pineal gland after stimulation by 5'-N-ethylcarboxamidoadenosine (NECA), an A2 adenosine agonist. Experiments were conducted with 2-month-old rats raised on semipurified diets containing either peanut oil (n-3 deficients) or peanut plus rapeseed oil (controls). The proportion of docosahexaenoic acid (22:6 n-3) in the pineal total lipid fraction and in phosphatidylcholine and phosphatidylethanolamine was significantly decreased in n-3-deficient rats. This was compensated for partially by an increase in 22:4 n-6 and 22:5 n-6 levels. The activity of the cultured rat pineal, in terms of cyclic AMP content and N-acetylserotonin and melatonin release in the medium, was lower after stimulation by 10(-5) mol/L NECA in the group fed peanut oil than in the group fed peanut plus rapeseed oil. The increased ratio of n-6/n-3 fatty acids in pineal total lipids and the major glycerophospholipids (phosphatidylcholine and phosphatidylethanolamine) may have an important influence on the rat pineal responses. The results are discussed in the context of changes in membrane-bound proteins, including enzymes and/or receptors involved in the rat pineal gland function.

Astrocytes are mainly responsible for the polyunsaturated fatty acid enrichment in blood-brain barrier endothelial cells in vitro.

To determine the respective roles of endothelial cells from brain capillaries and astrocytes in the conversion of circulating 18:2n-6 and 18:3n-3 into 20:4n-6 and 22:6n-3, respectively, a coculture of the two cell types mimicking the in vivo blood-brain barrier was used. During the culture period, endothelial cells cultured on an insert were set above the medium of a Petri dish containing or not a stabilized culture of astrocytes. Five days after confluence, labeled 18:2n-6 and 18:3n-3 (10 microM each) were added to the endothelial cells and incubated for 48 h. Analogous experiments were also performed by using each cell type cultured alone in the culture device. The distribution of radioactivity in lipids and fatty acids was studied in all the compartments of the culture device. Endothelial cells cultured alone weakly converted the precursor fatty acids into 20:4n-6 and 22:6n-3. When endothelial cells were cocultured with astrocytes, their content of polyunsaturated fatty acids increased dramatically. This effect was associated with the uptake of polyunsaturated fatty acids from the lower medium (astrocyte medium). These fatty acids were released by astrocytes after they were synthesized from the precursor fatty acids that passed through the endothelial cell monolayer into the lower medium. Polyunsaturated fatty acids were released by astrocytes as unesterified fatty acids and as phospholipids (mainly phosphatidylcholine and lysophosphatidylcholine) even when the medium was devoid of serum. These results suggest that astrocytes could play a major role in the delivery of essential polyunsaturated fatty acids to the barrier itself and to the brain.

Leukotriene B4 level in stimulated blood neutrophils and alveolar macrophages from healthy and asthmatic subjects. Effect of beta-2 agonist therapy.

Leukotriene B4 levels were measured after stimulation by calcium ionophore A23187: (i) in peripheral, neutrophils (PMN) from allergic asthmatics, rhinitis and healthy subjects; (ii) in macrophages collected by bronchoalveolar lavage. LTB4 levels in PMNs were significantly higher in non-treated allergic asthmatics and non-treated subjects with rhinitis compared to controls. Beta-2 agonist-treated asthmatics showed a significantly decreased LTB4 production which was not different from those of controls. In vitro, LTB4 production decreased significantly after PMN incubation with Salbutamol (10(-6) mol l-1). LTB4 produced by AM collected by BAL was measured in non-treated (n = 5) and treated (n = 11) asthmatics with inhaled beta-2 agonist. AM collected from all controls and non-treated asthmatics produced LTB4. By contrast, no production of LTB4 was observed in the treated group. LTB4 production decreased when normal AM were incubated in vitro with Salbutamol (10(-8) mol l-1). These results suggest that biochemical differences occur in PMN and macrophages from subjects treated with beta-2 agonist, presumably in changing the 5-lipoxygenase pathway.