Endogenous Arachidonic Acid Metabolism by Calcium Ionophore A23187-stimulated Lamb Lungs: Effect of Hypoxia Basil O. The, Warren B. Isenberg, and J. Usha Raj Department of Pediatrics, Harbor-UCLA Medical Center, Torrance, California

We have determined eicosanoid production from endogenous arachidonic acid by neonatal lamb lungs stimulated with calcium ionophore A23187 during normoxia and hypoxia. Lungs of lambs 19 to 25 d of age were isolated and perfused with cell-free Krebs' bicarbonate buffer at a flow rate of 15 to 20 ml/kglmin. After 30 min of equilibration in a recirculating system, A23187 was added to the perfusate in.a 5-~M concentration and perfusion continued for 15 min more. Eicosanoids were measured in perfusate and lung homogenate supernatant. Cyclooxygenase metabolites prostaglandin (PG) ~, thromboxane A2 , and PGI 2 were measured by radioimmunoassay, and 5-lipoxygenase metabolites leukotrienes (LT) B4 , C4 , D4 , and E4 by high performance liquid chromatography. During normoxia, all three cyclooxygenase metabolites were present in perfusate, but only PGI 2 and thromboxane A2 were present in lung homogenate supernatant. Prostacyc1inconstituted 50 % of all the cyclooxygenase products measured. LTC4 and LTD4 were detected in both perfusate and lung homogenate supernatant with little production of LTE4 and LTB4 • During hypoxia, the profile of cyclooxygenase products was unchanged and prostacyc1in production was not increased. However, the profile of leukotriene metabolites was altered. LTC4 synthesis was markedly reduced. The synthesis of LTE4 and LTB4 was increased 10-fold, with most of the leukotrienes being retained in lung tissue. We conclude that hypoxia significantly alters leukotriene metabolism of endogenous arachidonic acid by calcium ionophore-stimulated lungs. The increased production by stimulated lungs during hypoxia of LTE4 , a substance that may increase lung capillary permeability, and that of LTB4 , a powerful chemoattractant, may be important contributing factors to lung injury.

The lung is a major site of biosynthesis and action of leukotrienes (LT) (1, 2) and prostaglandins (PG) (3, 4). Arachidonic acid and some of its metabolites regulate vascular tone in the lung by maintaining a local balance of vasodilator and vasoconstrictor effects (5). For instance, prostacyclin, PGI2 , is synthesized and released continuously so that basal vasomotor tone is maintained low in the normal pulmonary circulation (6). Prostacyc1in is also involved in modulating vasomotor tone during active vasoconstriction (7). Other metabolites, such as thromboxane (TX) A2 (8) and leukotrienes (9), have been implicated as mediators of pulmonary vasoconstriction in disease states. Vasodilator prostaglandins, such as prostacyc1in, may play an important role in decreasing pulmonary vascular resistance during initiation (Received in originalform June 29, 1990and in revisedform September26, 1990) Address correspondence to: Basil 0. The, Ph.D., A-17 Annex, HarborUCLA Medical Center, Torrance, CA 90509.

Abbreviations: dimethyl sulfoxide, DMSO; high performance liquid chromatography, HPLC; leukotriene, LT; prostaglandin, PG; radioimmunoassay, RIA; reverse-phase HPLC; RP-HPLC; thromboxane, TX; -y-glutamyl transpeptidase, -y-GTP. Am. J. Respir. Cell Mol. BioI. Vol. 4. pp. 379-385, 1991

of ventilation at birth (10, 11). There is considerable uncertainty as to the exact role of arachidonic acid metabolites in the pulmonary hypoxic vasopressor response (12-14), though the evidence seems.to point to a modulating role (7, 15) for eicosanoids duringhypoxia. The role of these metabolites in modulation of'vasomotor tone in lung disease states associated with hypoxia is not known. Most studies investigating the role of these inflammatory mediators in maintenance of pulmonary vasomotor tone utilize agents that block either their synthesis or the receptor interaction of the various metabolites. It is probably equally important to determine the production and release of eicosanoids by the whole lung to investigate the potential role of these metabolites in regulation of vasomotor tone. In this study, we wished to determine the metabolism of endogenous arachidonic acid by lungs of neonatal lambs in response to a stimulus, calcium ionophore. We have also investigated the effect of hypoxia on lung metabolism of endogenous arachidonic acid under stimulated conditions. Our study was designed to mimic the pathophysiologic situation of an infected lung or a lung stimulated by allergens and/or antigens that is also exposed to hypoxic conditions. We chose calcium ionophore as the stimulus as it is an effective pharmacologic agent that has been widely used to stimulate ara-

380

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 4 1991

chidonic acid metabolism in lung and other tissues (16, 17). Calcium ionophore was preferred over other stimuli such as an antigen because it is more effective in stimulating the release of a wider range of arachidonic acid metabolites in a consistent manner (18).

Materials and Methods Neonatal lambs (19 to 25 d of age) were supplied by Nebekar Farms, Scientific Associates (Santa Monica, CA). Calcium ionophore A23187 and all chemicals used to compound the physiologic buffers were obtained from Sigma Chemical Co. (St. Louis, MO). PGE2, 6-keto-PGF l c" PGB2, TXB2 standards, and LTB4 , LTC4 , LTD4 , and LTE4 standards were purchased from Biomol Research Laboratories (Plymouth Meeting, PA). [3H]TXB2 (145 Ci/mmol), (3H]6-keto-PGF l a (153 Ci/mmol), (3H]PG~ (185 Ci/mmol), and (3H]LTC4 (38.4 Ci/mmol) were obtained from New England Nuclear (Boston, MA). Antisera for prostaglandins and TXB2 were gifts of Dr. William B. Campbell (University of Texas Southwestern Medical Center at Dallas, Dallas, TX). Bakerbond octadecyl (CIS) column cartridges were purchased from 1. T. Baker (Phillipsburg, NJ). The high performance liquid chromatography (HPLC) solvents (Optima grade) were purchased from Fisher Scientific (Santa Clara, CA). Methods Isolated lung preparation. Lungs of eight neonatal lambs 19 to 25 d of age were isolated and perfused using a modification of methods previously described by us (15). Briefly, animals were anesthetized with intramuscular ketamine (5 mg/kg) and subcutaneous 2 % lidocaine, after which heparin (1,000 IU/kg) was administered to each animal intravenously. An endotracheal tube was tied into the trachea. Euthanasia was accomplished with an intravenous overdose of pentobarbital (100 mg/kg). After sternotomy, the superior and inferior venae cavae, ascending aorta, and ductus arteriosus were ligated, and a suture was tied around the atrioventricular groove to occlude the ventricular lumen. Lungs were then excised from the thoracic cavity, weighed, and cannulae filled with saline were tied into the pulmonary artery and left atrium. The lungs were placed on their dorsal surface on a heated pad and covered with a humidified nonevacuated Plexiglas chamber. The perfusion buffer and the lungs were maintained at body temperature (38 to 39° C). The perfusion circuit included a bubble trap placed before the pulmonary artery to prevent air bubbles from entering the lungs. The vascular cannulae were connected to the perfusion circuit, and perfusion was initiated using a pulsatile flow pump (Harvard Instruments, Boston, MA). Lungs were ventilated via the tracheal cannula using an anesthesia bag (INTEC, Englewood, CO). Lungs were washed in a nonrecirculating system with normal saline using a pulsatile pump at a flow rate of 30 to 40 ml/kg/min. During this time, the lungs were ventilated with a gas mixture containing 30 % O2, 6 % CO 2, and balance N2. The lungs were washed until the effluent was clear and devoid of red and white blood cells. We confirmed this in some cases by examining the effluent under a light microscope. After washing, the lungs were connected to the perfusion circuit, which now recirculated the perfusion

buffer. Perfusate consisted of Krebs' bicarbonate buffer containing (in mM): NaCI, 119; KCI, 4.7; NaHC03 , 22.6; MgS04 , 1.17; KH 2P0 4 , 1.18; csct, 1.6; and glucose, 5.5. Dextran (mol wt, 72,000) was added at 2 g/100 ml of buffer, and buffer pH was 7.4. Experimental protocol. The lungs were divided into two groups: group I, n = 4, was studied in normoxia, and group II, n = 4, was studied in hypoxia. All lungs were ventilated with peak airway pressure ranging between 15 and 20 cmH 20 until they were uniformly inflated and thereafter were kept distended at a constant airway pressure of 7 cmH 2o. Lungs were inflated manually every 5 min up to 15 cmH 20 airway pressure throughout the duration of the experiment. To achieve normoxic conditions, lungs were ventilated with a gas mixture comprised of 30% O2, 6% CO 2, and 64% N2. For hypoxia, the gas mixture contained 8 % CO 2, 0 % O2, and 92 % N2. It was necessary to use an anoxic gas mixture to lower perfusate p02 < 50 torr. In both groups, perfusate pC02 was kept below 50 torr. The perfusate was kept in an airtight chamber that was placed in a water bath so that perfusate temperature was 38 to 39° C. The normoxic or anoxic gas was bubbled through the perfusate during the experiment. Lungs were perfused at a flow rate of 15 to 20 ml/kg/min with the pulsatile pump. The lungs were perfused for a period of 30 min so that perfusate temperature, pH, and gas tensions could be adjusted (BMS 3MK2/PHM73; Radiometer, Copenhagen, Denmark). Lungs were perfused under zone 3 conditions, i.e., the outflow pressure was kept greater than the airway pressure throughout the height of the lungs. At the end of the 30min equilibration period, in all lungs, calcium ionophore A23187, solubilized in dimethyl sulfoxide (DMSO), was added to the perfusate reservoir (250 ml) to give a final concentration of 5 J.'M. Westcott and associates (17) reported that a 5-J.'M concentration of A23187 was sufficient to give a maximal stimulation of eicosanoid synthesis in isolated perfused lung. The concentration of DMSO in the perfusate was less than 0.1%. The perfusion was allowed to continue for another 15 min after the addition of the ionophore, after which the perfusate was drained and collected from the lungs and frozen immediately in a dry ice bath. The lungs were then drained of all perfusate from the vasculature by gravity, weighed, and immediately placed on an ice bath. The weight gain due to perfusion was determined from the difference between the initial lung weight and the weight after perfusion. Each lung was homogenized in 90 to 110 ml of ice-cold Krebs' bicarbonate buffer. The lung homogenate was centrifuged at 5,000 x g for 15 min in a Sorvall RC-5B centrifuge maintained at 4°C. The supernatant was decanted and frozen at once in a dry-ice bath while the lung tissue sediment was quantitatively transferred to preweighed receptacles. Recovery of the lung wet weight by this method was usually between 90 and 95 %. Lung tissue was dried in an oven at 80° C until three consecutive weights were unchanged. All arachidonic acid metabolites were expressed as ng/g dry lung tissue. Extraction ofleukotrienes. Samples of perfusate and lung homogenate supernatant were extracted for HPLC analysis as previously described (19, 20). Briefly, [3H]LTC4 (3,000 cpm) was added to each sample to monitor the recovery of leukotriene metabolites. To aid in quantitation of metabo-

The, Isenberg, and Raj: Arachidonic Acid Metabolism in Stimulated Lamb Lungs

lites, PGBz (250 ng) was added to each sample as the internal standard before extraction. The lung homogenate supernatant was further centrifuged at 3,000 X g for 10 min to facilitate extraction. The perfusate samples did not need further centrifugation. The recovery of radioactivity after centrifugation was between 80 and 86 %. The prepared samples were then extracted using Bakerbond C I 8 cartridges previously equilibrated with 5 ml methanol and washed sequentially with 10 ml each of methanol, water, and 0.1% EDTA. The sample was then loaded to the column, after which it was washed with 10 ml water. Leukotriene metabolites were eluted from the column with 5 ml methanol. The recovery of radioactivity from the extraction step was 85 % for the perfusate and 80 % for lung homogenate supernatant samples. Sample extracts were dried under a stream of N, and taken up in 70 % methanol for HPLC analysis. HPLC of leukotrienes. The HPLC equipment was an ISCO model 2360 pump with a Valco injector, model 2350 ternary gradient programmer, and a V4 variable wavelength detector (Instrument Specialists Co., Lincoln, NE). The detector was connected to an HP3394 integrator (HewlettPackard, Palo Alto, CA). A reversed-phase HPLC (RPHPLC) column was used in the analysis. The RP-HPLC column was a Spherisorb C I8 , 5 J-tm, 4.6 X 250 mm (Instrument Specialists) with a 10-J-tm C I8 guard column (Phenomenex, Rancho Palos Verdes, CA). The elution solvent was a mixture of methanol-water-glacial acetic acid (50:50:0.1) buffered at pH 5.6 with ammonium hydroxide. Elution of metabolites was done at a flow rate of 1 ml/min beginning with 50 % methanol for 5 min, then changed to 58 % methanol isocratically for 45 min, ending with a linear gradient to 100% methanol in 10 min. This is a modification of the method described by Westcott and associates (17). Using this solvent system, the retention times of the leukotrienes were: LTC4 , 17 min; PGBz, 21 min; LTD4 , 32 min; LTB4 , 39 min; and LTE4 , 41 min. The HPLC assay was tested for linearity of recovery. The calibration curves were linear at Absz80 for each metabolite tested. The linearity of the assay was assessed on the basis of the correlation coefficient of the slope of the calibration curves of the added metabolite versus its standard. The slope of each metabolite was as follows (slope, correlation coefficient): LTC4 , 1.08 ± 0.04,0.98; LTD4 , 0.84 ± 0.03,0.85; LTE4 , 0.94 ± 0.01, 0.88; LTB4 , 1.01 ± 0.02, 0.95. The concentration of leukotriene metabolites in the sample is obtained from a calibration curve using authentic leukotrienes that is run each day of analysis. Radioimmunoassay (RIA) of eicosanoids. The metabolites quantitated by RIA were PGE z, PGIz, assayed as its stable metabolite 6-keto-PGF!'H and TXA z, assayed as its stable metabolite TXBz• The metabolites were extracted before RIA. The procedure for the extraction and RIA was a modification of a previous study (21). Briefly, each sample of perfusate or lung homogenate supernatant was acidified to pH 3.0 with glacial acetic acid. (3H]6-keto-PGF Ia , [3H]PGEz, or [3H]TXBz (3,000 cpm) was added to the samples to monitor recovery of eicosanoids. The precipitate that formed after acidification was removed by centrifugation at 3,000 X g for 10 min. The supernatant was decanted, and the sediment discarded. A 100-J-tl aliquot of supernatant was used to determine the recovery of metabolites. There-was no

381

significant difference in percent recovery among the three eicosanoids. Therefore, in subsequent assays, [3H]6-ketoPGF Ia was the prostanoid of choice to monitor recovery. Recovery of eicosanoids was > 95 % from perfusate samples and> 90 % from lung tissue. Final extraction of metabolites was done on Bakerbond C I8 extraction columns. The columns were prewashed with 10 ml of 95% ethanol, followed by 10 ml HPLC-grade water. The sample was loaded onto the column and sequentially washed with 5 ml each of 15% ethanol and toluene. The metabolites were eluted from the column with 6 ml ethyl acetate and dried under Ns. Recovery of the eicosanoids from this extraction process was > 95 %. Dried samples were dissolved in 1 ml RIA buffer, which was phosphate-buffered saline (pH 7.4) containing 0.1% polyvinylpyrrolidone and 0.1% sodium azide. We tested the antisera for cross reactivity with standards of all the metabolites we wished to measure (6-keto-PGF Ia , PGEz, TXBz, LTC4 , LTD4 , LTE4 , and LTB4 ) with some other arachidonic acid metabolites, i.e., PGF za, 5-HETE and 12-HETE, arachidonic acid, and A23187. The cross reactivity of each antiserum with these other compounds was < 0.01 %. RIA measurements were based on standard curves using an authentic standard of each metabolite in a concentration range of 5 to 200 pgfO.3 ml for both 6-keto-PGF Ia and PGE z but a range of 2.5 to 100 pgfO.3 ml for TXBz. The 6-keto-PGF Ia and PGEz antisera produced 50% displacement of bound standard at 25 pg. The corresponding value for TXBz antisera was 20 pg. Curve-fitting of standard and calculation of amount of metabolite were done with a weighted nonlinear RIA program. Final results were corrected for recovery and expressed as ng/g dry lung weight. Data Analyses Data were analyzed statistically using a two-tailed Student's t test (22). To detect differences in eicosanoids between perfusate and lung tissue, a paired t test was used. An unpaired t test was used to detect differences between normoxia and hypoxia. Differences were considered significant at P < 0.05.

Results All lungs gained weight by the end of the experiment: 67.1 ± 9.8% and 55.6 ± 22.8 % of initial wet weight during normoxia and hypoxia, respectively. None of the lungs developed air space edema. There was no significant difference in the increase in lung weight after perfusion between the two groups of lungs. Synthesis of Cyclooxygenase Metabolites by Lamb Lungs Figure 1 shows the cyclooxygenase metabolites measured from the stimulated lamb lungs. During normoxia, there was no significant difference in the amount of 6-keto-PGF!a, PGE z, and TXBz measured in perfusate. TXBz and 6-ketoPGF l o were also found in lung homogenate supernatant. The amount of 6-keto-PGF Ia in lung homogenate supernatant was significantly higher than that in perfusate, but there was no statistically significant difference in the amounts of TXBz measured in perfusate and lung homogenate supernatant in this group of lungs. PGEz was not detected in lung homogenate supernatant in any of the lungs during normoxia. 6-keto-PGF Ia constituted more than 50% of the

382

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 4 1991

NORMOXIA

100

o

90

IOSSI

*

80 70

.os,

HYPOXIA

NORMOXIA

Perfusate Lung Homogenate Supernatant p(O.05 different from Perfusate

*

E c

*

60

HYPOXIA

o

1IO

~.02

50

III

.Q

40

ce

30 20 10

*

O+----J~lo.I.--....L....J,r-'-----L..Iylo_:lL.-_+-......L..~---J...J,_'""""----....L....J,~-

6-Keto PG~

TXB2

6-Keto PGE2

4~

3D TIME (MIN)

TXB2

Figure 1. Profile of cyclooxygenase metabolites synthesized by cal-

Figure2. Separation of leukotrienes from the perfusate of calcium

cium" ionophore A23187-stimulated neonatal lamb lungs during normoxia and'hypoxia. The profile of the three prostanoids 6-ketoPGFIa, PGE2, and TXB2 was similar during normoxia and hypoxia. Data are shown as mean ± SEM.

ionophore A23187-stimulated neonatal lamb lungs during normoxia and hypoxia. Leukotrienes were extracted from perfusate and analyzed by HPLC on a Spherisorb CIScolumn. The retention times of known standards are shown in the chromatogram.

measured cyclooxygenase metabolites in lung homogenate supernatant and perfusate during normoxia. During hypoxia, 6-keto-PGF 1a and TXB2 were detected in similar amounts in perfusate and lung homogenate supernatant as in normoxia. During hypoxia, the amount of TXB 2 in lung homogenate supernatant tended to decrease, but this was not statistically significant (P = 0.247). Of note, we found that the synthesis of prostacyclin was unchanged during hypoxia. However, 6-keto-PGF 1a remained the predominant prostanoid, accounting for more than 50% of measured metabolites during hypoxia.

unidentified metabolites such as peak I, which was present during normoxia, and in addition peak II, which also absorbed at 280 nm, were present. Figure 4 is a quantitative representation of leukotrienes measured in both perfusate and lung homogenate supernatant during normoxia and hypoxia. During normoxia, L1C4 was the predominant leukotriene measured, accounting for more than 75 % of the measured metabolites. The amount of L1C4 in perfusate tended to be greater than that in lung homogenate supernatant, but this difference did not reach statistical significance (P = 0.15). LTD4 was measured in a much smaller amount in the perfusate than in lung homogenate supernatant, and LTE4 and LTB4 were barely detectable in perfusate. In lung homogenate supernatant, L1C4 , LTD4, and LTB4 were detected in decreasing amounts; however, LTE4 was not detected at all. Hypoxia dramatically altered the pattern of leukotriene synthesis. LTC4 was present in significantly lower amounts

Synthesis of Leukotrienes Calcium ionophore A23187 stimulated the synthesis of leukotrienes by lungs during normoxia and hypoxia. These metabolites were separated effectively by HPLC and are qualitatively represented in Figures 2 and 3. Figure 2 shows the HPLC profile of leukotrienes detected from perfusate from two experiments, one during normoxia and the other during hypoxia. During normoxia, LTC4 was detected in perfusate in the highest amount. LTD4, LTB4, and a very small amount of LTE4 were also detected in perfusate. Hypoxia altered the profile of leukotrienes measured in perfusate. LTC4 was now present in a much smaller amount. LTD4 was measured in perfusate in similar amounts as during normoxia. LTE4 and LTB4 were not detected in perfusate during hypoxia. Some other peaks (e.g., peak I), which we did not identify, were present in the perfusate. Figure 3 shows the HPLC profile for the lung homogenate supernatant, both during normoxia and hypoxia. During normoxia, the peaks corresponding to LTC4, LTD4, and LTB4 were evident on the chromatogram. LTE4 peak was not detected. Peak I, which was present in perfusate, was also evident. During hypoxia, the peak corresponding to LTC4 was reduced while the peak corresponding to LTD4 was the same as during normoxia. The peak for LTB4 was much higher during hypoxia than during normoxia. Also, during hypoxia, a peak for LTE4 was detected in lung homogenate supernatant for the first time. During hypoxia,

·OS

NORMOXIA

HYPOXIA LTB

Ec .02 III

.Q

ce

7

T

0 ;

LTC4

LTC

1

1 4~

6

1~

TIME (MIN)

Figure 3. Separation of leukotrienes from lung homogenate supernatant of calcium ionophore A23187-stimulated neonatal lamb lungs during normoxia and hypoxia. Leukotrienes were extracted from the lung homogenate supernatant and analyzed by HPLC on a Spherisorb CIS column. The retention times of known standards are shown in the chromatogram.

The, Isenberg, and Raj: Arachidonic Acid Metabolism in Stimulated Lamb Lungs

NORMOXIA

3000

o

~

2500 Cl

Z

:::>

HYPOXIA

** *

Perfusate Lung Homogenate Supernatant different from perfusate different from normoxia

* p(O.05 ** p(O.05

2000

....J

>0::: 0 Cl

Endogenous arachidonic acid metabolism by calcium ionophore A23187-stimulated lamb lungs: effect of hypoxia.

We have determined eicosanoid production from endogenous arachidonic acid by neonatal lamb lungs stimulated with calcium ionophore A23187 during normo...
868KB Sizes 0 Downloads 0 Views