Endothelin-l Stimulates Eicosanoid Production in Cultured Human Nasal Mucosa Tong Wu, Joaquim Mullol, R. Dwaine Rieves, Carolea Logun, Jeffrey Hausfield, Michael A. Kaliner, and James H. Shelhamer Department of Critical Care Medicine and Allergic Diseases Section, Laboratory of Clinical Investigation, National Institutes of Health, Bethesda, Maryland, and Department of Facial Plastic and Reconstructive Surgery, Washington Hospital Center, Washington, DC

Endothelin (ET) has been shown to contract both vascular and nonvascular smooth muscle and to stimulate human nasal glandular secretion of serous and mucous cell products. Some effects of ET are thought to be mediated by eicosanoid production. To explore the direct effect of ET on arachidonate metabolism in cultured human nasal mucosal explants, eicosanoids were measured after ET-1 stimulation. After labeling the explants with pH]arachidonic acid (AA), supernatant from control and ET-1-treated explants were fractionated by reverse-phase high-performance liquid chromatography (HPLC). The resulting elution pattern suggested the release of prostaglandin (PG) E, and AA in response to ET-1 stimulation. Radioimmunoassay after HPLC resolution confirmed that ET-1 induced a significantly increased release of PG~ as well as PGD 2, PGF 2a, thromboxane Bh and 15-hydroxyeicosatetraenoic acid (15-HETE). Although significant amounts of 15-HETE were generated, cyclooxygenase product generation was most remarkable. Eicosanoid release after ET-1 exposure (10 to 0.1 ~M) is concentration dependent and occurs within 1 h. Whereas 15-HETE release was maximal at 4 h, prostanoid production was maximal 1 h after exposure to ET-l. Other assayed AA metabolites, including the peptidoleukotrienes, did not significantly change after ET-1 stimulation. We conclude that ET-1 induces the release of predominantly cyclooxygenase products from cultured human nasal mucosal explants.

Endothelin-l (ET-1) is a 21-amino acid peptide that was first isolated from cultured porcine aortic endothelial cells (1). It possesses potent and sustained contractile activity on both vascular and nonvascular smooth muscle, including trachea and bronchus (1-10). In vitro analysis confirms that ET-1 causes vasoconstriction of arteries and veins. After intravenous injection of ET-1 in laboratory animals, a transient vasodilatory phase precedes a sustained pressor response (11-13). The initial vasodilatory response may be mediated by the ET-1-induced release of eicosanoids, endotheliumderived relaxing factor, or atrial natriuretic factor (13). In the pulmonary system, ET-1-induced constriction of guinea pig airways is inhibited by the cyclooxygenase inhibitor indomethacin, by the specific thromboxane (TX) receptor antagonists BM 13177 and BM 13505, and by the specific platelet-

activating factor (PAF) receptor antagonist BN 52021 (4, 7, 14). Although ET induces TXB 2 production from guinea pig airways (10), no study to date has examined the effect of ET on arachidonate metabolism in the respiratory mucosa. Recent studies show that the human nasal mucosa contains ET-1 in the epithelium, submucosal gland serous cells, and endothelial cells (15). Addition of ET-1 and ET-2 to cultured human nasal mucosal explants results in the secretion of both mucous and serous cell products, probably through stimulation ofET receptors localized to submucosal glands (15). We report here that ET-1 releases predominantly cyclooxygenase products from cultured human nasal mucosa explants, suggesting that some of the actions of ET might be secondary to prostaglandin (PG) generation.

Materials and Methods (Received in original form March 29, 1991 and in revised form July 30, 1991) Address correspondence to: James H. Shelhamer, M.D., Department of Critical Care Medicine, National Institutes of Health, Building 10, Room 7043, Bethesda, MD 20892. Abbreviations: arachidonic acid, AA; endothelin, ET; hydroxyeicosatetraenoic acid, HETE; high-performance liquid chromatography, HPLC; leukotriene, LT; platelet-activating factor, PAF; prostaglandin, PG; radioimmunoassay, RIA; thromboxane, TX. Am. J. Respir, Cell Mol. BioI. \;>1. 6. pp. 168-174, 1992

Materials ET-1 was obtained from Peninsula Laboratories (Belmont, CA); pH]arachidonic acid (AA), (3H]leukotriene (LT) B., PH]LTC., PH] LTD. , PH] LTE. , (3H]5-hydroxyeicosatetraenoic acid (HETE), PHl12-HETE, PHl15-HETE, PHlPGD h PH]PGE2, PH] PGF 2a, (3H]6-keto-PGF'a, PH]-TXB h and radioimmunoassays (RIAs) for LTB., LTC./D./E., 15HETE, PGDh PGF 1a, 6-keto-PGF'a, and TXB 2 were from Amersham Co. (Arlington Heights, IL); RIAs for 5-HETE,

Wu, Mullol, Rieves et al.: Endothelin and Arachidonic Acid Metabolism

12-HETE, and PGE 2 were from Advanced Magnetics (Cambridge, MA); aprotinin and unlabeled standards for PGB2, PGD2, PGE 2, PGF 2" 6-keto-PGFIQ, TXB2, LTC., LTD., LTE., 5-HETE, 12-HETE, 15-HETE, and AA were from Sigma Chemical Co. (St. Louis, MO); medium CMRL 1066, medium Ll5, penicillin, streptomycin, and amphotericin B were from GIBCO (Grand Island, NY); Gelfoam was from the Upjohn Co. (Kalamazoo, MI); high-performance liquid chromatography (HPLC) grade methanol, HPLC grade acetonitrile, and HPLC grade water were from 1. T. Baker Chemical Co. (Phillipsburg, NJ); HPLC grade trifluoroacetic acid was from Pierce (Rockford, IL); and phosphoric acid was from Mallinckrodt Inc. (Paris, KY). The PAF antagonist, Ro 19-3704, was a gift from Dr. Peter Sorter of Hoffman-LaRoche (Nutley, NJ). Tissue Handling Human inferior turbinates were obtained from patients undergoing elective surgery for nasal obstructive syndromes. No patient had had a recent infection. At the time of surgery, 2 % tetracaine HCI and 0.25 % phenylephrine HCI were applied topically on nasal packs. The turbinates were injected with 2 to 4 rnl of 1% lidocaine with 1:100,000 epinephrine. An incision was made from the lateral wall of the inferior turbinate at the level of the infundibulum through the inferior conchal bone, and then inferiorly along the medial aspect of that bone. The medial flap of residual turbinate tissue was wrapped superiorly and laterally to close the wound. Within 20 min of surgical excision, the nasal mucosa was dissected from the inferior conchaI bone. Specimens were placed in Ll5 medium supplemented with penicillin (100 U/rnl), streptomycin (100 JLg/rnl), and amphotericin B (0.5 JLg/ml) for transport to the laboratory. Human Nasal Mucosal Explant Culture Fresh nasal mucosa was cut into 3 X 3 mm fragments (16, 17). Pairs of fragments were placed on 5 X 10 mm Gelfoam pads in 35-mm petri dishes. The mucosa cultures (80 to 100 mg wet tissue/dish) were maintained in 2 rnl of CMRL 1066 medium containing penicillin (100 U/ml), streptomycin (100 JLg/rnl), and amphotericin B (0.5 JLg/ml) in a controlled atmosphere chamber gassed with 45% O2, 50% N2, and 5% CO 2, and incubated at 37°C. After 24 h of initial culture, medium was replaced with fresh medium containing 400 KIU/rnl aprotinin to inhibit the proteases capable of degrading ET-I (17). Experiments were performed after an additional 24 h of culture. The tissue from each patient was used for one or two experiments. Experimental Conditions Studies of (3H]AA-labeled explants with and without ET-I stimulation were performed to determine the spectrum of eicosanoids generated at baseline and in response to ET-l. As the optimal incubation period for AA incorporation into respiratory cellular phospholipid fractions remains controversial (18-22), the nasal mucosal explants were incubated with (3H]AA (I JLCi/rnl) for both 2 and 12 h. 82.9% and 88.2% of added (3H]AA were incorporated into the explants at the end of 2- and 12-h incubation periods, respectively. The unincorporated (3H]AA was removed by washing the tissue

169

twice with 2 rnl medium. Cultures were continued for a I-h experimental period. During the experimental period, some explants were treated with ET-I, while others were maintained as controls. Aprotinin was maintained in all culture media throughout the experimental period. Supernatants from the experimental period were harvested. The spent media from two identically treated dishes were pooled and processed as described below. Additional studies involved the RIA of specific eicosanoids generated in control experiments and in response to ET-l. At the beginning of each experiment, each culture was washed with 2 rnl medium and replenished by an equal amount of fresh aprotinin-containing medium. Some cultures were then exposed to various concentrations of ET-I, while others were maintained as controls. The supernatants from four identically treated dishes were pooled for each sample. To assess the effect of a PAF antagonist on eicosanoid generation, some explants were treated with Ro 193704 and/or ET-l. Both ET-I and Ro 19-3704 were dissolved in culture media 30 min before addition to cultures. To establish the time course for eicosanoid generation after ET-I exposure, the cells were exposed to ET-I for I h. The spent medium was harvested from control and ET-I-exposed cultures, and fresh medium was added. The supernatants were again harvested, and medium changed at 2, 4, and 8 h. The samples were extracted on Sep-Pak CIS columns (Waters Associates, Milford, MA), and eicosanoids were measured by RIA after HPLC separation. The results from the control explants were compared with those from the stimulated explants. Extraction of Eicosanoids Polypropylene labware was used throughout the extraction process to avoid binding of the eicosanoids to glass surfaces. The spent media were centrifuged at 1,000 X g for 10 min before extraction on octadecyl-silane (ODS) Cs cartridges (Sep-Pak CIS; Waters Associates). Individual cartridges were prepared with 15 ml of ethanol followed by 5 rnl of 5 mM EDTA and 10 rnl of water. Samples were loaded onto the cartridges, washed with 10 rnl of water, and eluted with 4 rnl of methanol. The methanol fraction was collected, evaporated to dryness under steady flow nitrogen gas, and resuspended in 200 JLI of mobile phase A (see below) for analysis by HPLC. The extraction efficiency was evaluated by using 3H-Iabeled eicosanoid standards. After Sep-Pak Cs extraction, the recovery of 6-keto-PGFIQ, TX~, PGD 2, PG~, PGF2Q, LTC., LTD., LTB., 15-HETE, 12-HETE, and 5-HETE was 90.5,90.6,89.4,92.5,98.1,86.1,97.4,98.3, 92.8, 99.8, and 95.3 %, respectively. HPLC Reverse-phase HPLC was performed using a modification of the method described by Powell (23). A Beckman model 344 liquid chromatography system (Beckman Instruments, Fullerton, CA) was used with dual pumps (model 114; Beckman), an autosampler (model 506; Beckman), and a variable wavelength ultraviolet detector (model 164; Beckman). An Ultrasphere CIS (Beckman) column (4.7 X 250 mm) with 5-JLm particle size was used. A gradient program was used with mobile phase A: water/acetonitrile/phosphoric acid

170

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992

(75:25:0.025) and mobile phase B: methanol/acetonitrile/trifluoroacetic acid (60:40:0.0016) at a flow rate of 1.5 ml/min. HPLC grade reagents were used for all experiments. When samples from (3H]AA-prelabeled mucosal explants were applied, the radioactivity in each fraction was quantitated in an LS-8100 liquid scintillation counter (Beckman). Based upon the retention times of the eicosanoid standards, l-rnin fractions were pooled for RIA. Fractions of LTC., LTD., and LTE. were pooled for the measurement of peptidoleukotrienes by RIA. The pooled fractions were diluted in mobile phase A and passed over Sep-Pak C I8 cartridges before RIA. Recovery of eicosanoids after HPLC fractionation, measured by using radiolabeled standards, was 42.8, 55.2, and 60.7% for 15-HETE, 12-HETE, and PGF 20 , respectively. PGB2 was added as an internal standard. Because the baseline UV absorption was highly variable at 205 nM, where many PGs display maximal absorption, PGs were monitored only as peaks of radioactivity corresponding to elution times determined by radiolabeled standards. At 280 and 240 nM UV absorption, authentic standards of LTs and HETEs yielded distinct peaks corresponding to the radioactive peaks of pH]-labeled standards (data not shown). Therefore, LTs and HETEs were monitored as both UV absorption peaks and radioactive peaks. The lower limit of UV detection for LTs and HETEs is approximately 10 ng. RIA of Eicosanoids LTC./D./E., LTB., 5-HETE, 12-HETE, 15-HETE, PGD 2 , PG~, PGF 20 , 6-keto-PGF i o , and TXB2 were measured by RIA. The Sep-Pak C I8 eluates were evaporated and dissolved in assay buffer. The samples were assayed in duplicate following exactly the instructions supplied with the assay kits. The antisera were highly specific and the crossreactivity (50% B/BO displacement) with other eicosanoids is low, as listed in the description of the assay systems. The antiserum for LTC./D.IE. is specific for the peptidoleukotrienes, with a cross reactivity at 50% B/BO being 55.0, 100,51.0, and 66.7% for LTC., LTD., LTE., and LTF•. The total immunologic recovery of eicosanoids by HPLC purification was 79.3, 90.4, 63.1, and 49.3% for PGF 20 , 6-ketoPGF l o , LTB., and 5-HETE, respectively. The results were corrected for wet tissue weight and then expressed either as absolute values or as percent increase from control derived from the following equation: [(sample - control)/control]

x

100.

Statistics Results are expressed as the mean ± SE. The Student's paired t test was used to compare eicosanoid release from control and ET-l-treated explants. Competitive RIAs were analyzed by logit transformation and linear regression analysis. The dose-response effect of ET-l on eicosanoid production was analyzed by regression analysis.

Results Effect of ET-l on [3H]AA Metabolism Both 2- and 12-h prelabeling periods were used for (3H]AA incorporation into cellular membrane phospholipids. Figure 1 shows the profiles of AA and its metabolites under con-

A.

~ t

600

,

AA

1

PGDliE2iF2a 6·kPGF1 TXB2

800

lTC4

LT~

12HETE

~

15HEr 6r

TE

Control

ET·l

400

200

i' D-

e.

...

0 10

:~

.

;:; .~ ~

a:

30

50

70

B. 3000

2000

1000

, -o

Figure I. The high-performance liquid chromatography (HPLC) profiles of eicosanoids released by cultured nasal mucosa at baseline and in response to endothelin-l (ET-l) stimulation. The mucosal explants (approximately 100 mg of wet tissue/dish) were incubated with ['Hlarachidonic acid (AA) (l ILCi/rnl) in 2 rnl medium for 2 h (A) and 12 h (B). After washing twice with medium, 2 ml fresh medium containing aprotinin (400 KIU/rnl) were added to each dish. Subsequently, some explants were exposed to ET-l (10 ILM), while others were maintained as controls. After a l-h incubation with ET-l, the supernatants were harvested. The samples were chromatographed by HPLC after Sep-Pak C I8 extraction, and l-rnin fractions collected and counted for radioactivity. The profiles show the baseline release of radiolabeled peaks coeluting with AA and prostaglandin (PG) D2/E2/F20 ' as well as an unidentified peak eluted at 56 min. The addition of ET-l increased the release of PGD2/E2/F20 and AA, as well as the unknown AA metabolite. AA metabolism patterns were similar after 2 and 12 h of [3H]AA prelabeling. The retention times of 3H_ labeled eicosanoid standards are indicated by arrows along the top of the figure. The elution patterns represent the average of four to six individual experiments.

trol conditions or after ET-l stimulation. Although the radioactive peaks released after 2 h of (3H]AA labeling were less than after 12 h, similar chromatographic patterns were demonstrated for both 2- and 12-h periods of PH]AA labeling. Under basal conditions, chromatography of media with l-min fractions collected over a subsequent 70-min period revealed two peaks of radioactivity coeluting with PGD21 ~/F20 and AA, as well as an unidentified peak eluting at 56 min. The addition of ET-l (10 /LM) augmented the PGD2/E2/F20 and AA peak, as well as the unidentified peak. The separation of PGD2/~/F20 was achieved by a program with slow gradient increase of mobile phase Band the collection of 15-s fractions over 70 min. Figure 2 shows this extended profile of AA metabolism in response to ET-l stimulation after a 12-h [3H]AA prelabeling period. Under

Wu, Mullol, Rieves et al.: Endothelin and Arachidonic Acid Metabolism

M

~

1500

PG~

1000

~

- - ET·' Contro l

500

70 Retention Time (min)

Figure 2. The extended HPLC profile of AA metabolism in control dishes and in response to ET-l stimulation. After the mucosal explants were prelabeled with PH]AA for 12 h, the experiments were performed as described. The samples were chromatographed by HPLC with a slow gradient increase of mobile phase B, and 15-s fractions were collected and counted for radioactivity over 70 min. The individual peaks for PGDz, PGE z, and PGFza were well separated. At basal culture conditions, there was spontaneous release of PGE z and AA, with smaller amounts of PGFza and PGDz . The addition of ET-l augmented the PGE z and AA peaks. The change in the peaks coeluting with PGFza and PGDz was less remarkable. The chromatogram represents the average of six individual experiments.

basal conditions, the release of AA and PGE z dominate AA metabolism, with less generation of PGF za and PGD z• The addition of ET-l (10 JtM) clearly raised the PGE z and AA peaks. The relative change in radiolabeled peaks of PGF za and PGD z was less remarkable. Chromatograms of supernatants from ET-l-stimulated mucosal explants, including both (3R]AA-labeled or unlabeled specimens, revealed no detectable UV absorption peaks at the wavelengths of 280 and 240 nM (data not shown). Effect of ET-l on Eicosanoid Release Analyzed by RIA Because (3R]AA may not label all the potential sources of AA metabolites in cultured mucosa and this method is not quantitative, the production of AA metabolites was also assessed by RIA. Unlabeled mucosal explants were stimulated with ET-l, the eicosanoids isolated by RPLC, and the fractions corresponding to various authentic eicosanoid standards analyzed by RIA after Sep-Pak Cs extraction. The amounts of eicosanoids detected were corrected for wet tissue weight and the results for ET-l-stimulated explants compared with unstimulated tissues. Table I shows the eicosanoids generated from control and ET-l-stimulated explants. Nasal turbinate tissue produced a wide spectrum of eicosanoids, with the exception of peptidoleukotrienes. Under basal culture conditions, the eicosanoids released in highest concentration were PGE z (10,380 ± 2,130 pgl100 mg wet weight of tissue; n = 7) and PGF za (1,730 ± 300 pgl100 mg wet weight of tissue; n = 7), with lipoxygenase products in lower amounts. After a l-h incubation, ET-I significantly increased the generation of nearly all cyclooxygenase prod-

171

ucts (except for 6-keto-PGF ,a), with the relative increases (percent increase from control) being PGD z > PGE z > PGF za > TXB z• Although the percent increase from control for PGD z appeared to be greater than for PGEz, the absolute change for PGEz was more than 100 times higher than for PGD z• IS-RETE also increased significantly, but the percent increase was relatively small. None of the other lipoxygenase products increased, and no L1C4/D4/E4 was detected. Effect of Different ET-l Concentrations on Eicosanoid Release Mucosal explants were incubated with 0.1 to 10 JtM of ET-l for I h. Figure 3 shows the effects of ET-l on the release of PGD z, PGEz, PGF za, TXB z , and IS-RETE. Regression analysis demonstrated that the ET-l-induced release of PGDz, PGEz, PGF za, TXBz, and IS-RETE was concentration dependent (P < 0.01). Time Course Effect of ET-l on Eicosanoid Generation Figure 4 shows the ET-l-stimulated production and release of PGDz, PGEz, PGF za, TXBz, and IS-RETE at various time intervals. Maximal release of PGDz, PGEz, PGF za, TXB z was detected within I h (P < O.OS) after exposure to ET-l, with a progressive decline over the subsequent hours. Although a significant increase in IS-RETE release was noted at I h (P < O.OS), a greater response was detected at 4 h (P < O.OS). These data possibly suggest a different time course for the production and release of is-RETE. Effect of PAP Receptor Antagonist It has been reported that the airway smooth muscle response to ET-l is mediated through the generation of PAF (14). To investigate the role of PAF in ET-stimulated eicosanoid generation, Ro 19-3704, a PAF receptor antagonist, was used in this study. Figure S shows the effect of 1 JtM Ro 19-3704 on 10 JtM ET-l-stimulated release of PGD z, PGEz, PGF za, TXB z, and IS-RETE. ET-l alone consistently stimulated the release of the eicosanoids (P < O.oS); Ro 19-3704 alone had no effect (P = NS) and when added to ET-l exhibited no inhibitory effect on the stimulation of eicosanoid release (P = NS).

Discussion In addition to its potent vasoconstrictor effects, ET has recently been shown to have extravascular effects, including contractile activity on bronchial and intestinal smooth muscle, neurotransmitter action in the central nervous system, and cellular mitogenic properties (3-10, 13). It has been suggested that ET-l-induced bronchoconstriction and the initial vasodilatory response may be mediated by the secondary generation of bronchoconstrictive or vasodilatory eicosanoids (13). The present study investigates the effects of ET-l upon eicosanoid generation and release in cultured human nasal mucosa explants. Under baseline culture conditions, human nasal mucosa explants spontaneously generated cyclooxygenase products with a predominance of PGEz, whereas smaller amounts oflipoxygenase products were produced. Specifically, no peptidoleukotriene production was found. ET-l induced the release of several eicosanoids, with the greatest effect (percent increase from control) upon the

172

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992

TABLE 1

The effect of ET-l on eicosanoid production from cultured human nasal mucosa * LTB.

LTC.ID./E.

5-HETE

12-HETE

15-HETE

PGD,

ET-I

21.7 (4.0)

NA

15.6 (8.0)

85.6 (29.2)

104.2 (46.1)

321.0 (76.6)

38,690 (8,180)

5,460 (870)

784.0 (86.4)

278.2 (37.8)

Control

22.9 (4.4)

NA

14.5 (4.7)

83.8 (13.9)

71.0 (43.2)

45.4 (9.1)

10,380 (2,130)

1,730 (300)

830.9 (188.7)

128.5 (20.5)

n

3

P

NS

3

PGE,

6-keto-PGF 'Q

PGF'Q

TXB,

3

3

8

6

7

7

3

7

NS

NS

< 0.05

< 0.05

< 0.01

< 0.01

NS

< 0.01

Definition a/abbreviations: ET-1 = endothelin-1; LT = leukotriene; HETE = hydroxyeicosatetraenoic acid; PG = prostaglandin; TXB, = thromboxane B,; = numbers of individual experiments; NA = not assayed; NS = not statistically significant. * Eicosanoid production in human nasal mucosal explants at baseline and in response to ET-I stimulation was measured by radioimmunoassay after Sep-Pak CIS cartridge extraction and high-performance chromatography separation. Some explants were exposed to 10 /LM ET-l for I h, while others were maintained as controls. Nasal mucosal explants spontaneously release eicosanoids, including predominantly PGE,. After l-h incubation with ET-l (10 /LM), there was a significant increase in the release of PGD" PGE" PGF'Q, TXB" and l5-HETE, with no significant change for other eicosanoids. There was no detectable peptidoleukotriene production at baseline or in response to ET-l stimulation. Values represent mean (± SE) of eicosanoids measured in pg/lOO mg wet weight of tissue.

n

release of PGD 2 followed by PGE 2 > PGF 2Q > TXB 2 > l5-HETE. No significant increase of other eicosanoids was found in response to ET-l stimulation. Cyclooxygenase product generation from human nasal mucosa dominated the AA metabolism cascade under baseline culture conditions and in response to ET stimulation. Eicosanoids are a series of biologically active compounds implicated in multiple physiologic and pathologic processes. In the present study, a significant production of eicosanoids, predominantly cyclooxygenase pathway products, was found in response to ET-l stimulation. As the cultured human nasal mucosa explants contain a variety of cell types, including epithelial cells (ciliated, nonciliated, and goblet cells), vascular cells, submucosal glands, and connective tissue, any of

these cells may be responsible for the ET-l-stimulated eicosanoid generation. The cyclooxygenase pathway products released may have important inflammatory and immunoregulatory actions. PG~, the predominant AA product in the explants, exhibits regulatory effects on multiple cell types (24). For example, PG~ has been reported to suppress immune function by downregulation of interleukin-2 receptors on lymphocytes, inhibition of fibroblast proliferation and hyaluronate synthesis, inhibition of macrophage production of interleukin-l and tumor necrosis factor, and inhibition of cell proliferation and differentiation (25). PGD 2 caused peripheral vasodilation, pulmonary vasoconstriction, broncho-

•r:a

• 800

"2

.

C u

PC02

TXIl2

0

'SHElE

600

400

= c

PGf2' TXIl2 '5HETE

.

400

C u

E

,g

300

~

200

= c

.

....

~

~

....

"2

...

,g

..:

0

~

fa

E

S

0 0

500

PGE'

PC02 PGE'

100

200

·,00 4 --

Figure 3. The dose-dependent effects of ET-l on eicosanoid production by human nasal mucosal explants. The explants were incubated with 0.1 to 10 p.Mof ET-l for 1 h. Eicosanoids were measured by radioimmunoassay (RIA) after Sep-Pak C I8 cartridge extraction and HPLC separation. The results are expressed as percent increase from control. The values represent mean ± SE for seven to eight separate experiments. Regression analysis revealed that the ET-l stimulated release of PGD2 , PGE2, PGF 2Q, and thromboxane B2 (TXB2) , and that 15-HETE was positively correlated to the concentrations of ET-l (P < 0.01).

-

-

,.

-

_

-

-

-

2.

-

_

-

~

8.

Figure 4. The time course of ET-I effect on eicosanoid generation. Mucosal explants were exposed to 10 p.M ET-I for different time intervals, while others were maintained as controls. After harvesting of the medium after the initial l-h incubation, 2 ml fresh medium containing aprotinin was added to each dish and the culture continued. The supernatants were subsequently harvested at 2, 4, and 8 h. The samples were analyzed for eicosanoid generation by RIA after HPLC separation. The values represent mean ± SE for n = 4. The results were expressed as percent increase from control.

Wu, Mullol, Rieves et al.: Endothelin and Arachidonic Acid Metabolism

.00

ec

•

PG02 I'Gf2

fJ

PGf20 TXB2

0

300

' !>HEl E

0

u

E

~

200

:.

.

1;

.=

100

.

c

..... ~

.100 +.-- - - - - - - - - - - - - - - ETlo..M

Figure 5. The effect of Ro 19-3704 on ET-1-induced eicosanoid production. The exp1ants were treated with 10 JLM ET-1 and/or 1 JLM Ro 19-3704.Ro 19-3704 wasdissolved in medium andadded

to the cultures immediately before the addition of ET-l. After a 1-h incubation, the supernatants were harvested and analyzed for eicosanoid production by RIA after HPLC separation. The values representmean ± SE for n = 5. The resultsare expressed as percent increase from control. constriction, enhanced airway reactivity, and inhibition of p~atelet aggregation (24). PGF2a is capable of contracting airway smooth muscle and stimulating airway sensory nerve endings (26, 27). TXA 2 , the unstable parent compound of TXB2 , is a potent constrictor of vascular and airway smooth muscle (28). 15-HETE may cause mucus secretion, modify cellular immune functions, act as a mitogen, and modulate other oxygenation enzymes (29-32). Therefore, the production of these multiple eicosanoids may mediate some ET-l actions, such as vascular and bronchial responses. However, the physiologic or pathologic effects of ET-l in human nasal airway in vivo need further investigation because the concentrations of ET-l that were required for eicosanoid release were relatively high. In addition to its effect on AA metabolism, ET-l has recently been demonstrated to stimulate lactoferrin and mucous glycoprotein secretion from cultured human nasal mucosa (15). Immunohistochemical staining of human nasal mucosa demonstrated immunoreactive ET-l in epithelial and basal cells, submucosal glands, and arterial vessels (15). ET1 binding sites were present in submucosal glands, vascular walls, and perhaps in the respiratory epithelium (15). These findings suggested that ET-l may regulate local tissue blood flow and stimulate nasal submucosal gland secretion and eicosanoid production via an ET receptor or receptor subtypes. The finding that ET-l is localized mainly in vessels and glands contrasts with other peptides, such as calcitonin gene-related peptide, gastrin-releasing peptide, vasoactive intestin~ peptide, and neuropeptide Y, which are mainly present In nerve fibers (33-36). Elucidating the possible interactions of ET family with other peptides in nasal mucosa and other tissues may contribute to the understanding of the physiologic and pathophysiologic actions of this newly discovered peptide. The family of ET peptides consists of three distinct 21amino acid chains, which are coded by three separate genes

173

in the human, rat, and porcine genomes (37). Endothelial cells cultured from aortas have been shown to express the ET gene and represent sources of its synthesis (1). Additionally, ET mRNA IS expressed by renal tubular epithelial cells (38). ET production has also been identified from cultured tracheal epithelial cells (39, 40). By in situ hybridization, ET gene expression and synthesis have been localized to the kid?ey, lung,. cerebellum, and eye (41). The released ET may Interact WIth cells as a local autacoid or as a circulating hormone (13). ET binding sites have been identified not only in the cardiovascular system and human nasal mucosa, but also in the lung, kidney, adrenal gland, brain, spinal cord, gastr.ointe~tinal tract, liver, and spleen (42-45). ET may elicit biologic responses by various signal transduction mechanis~s, including the G protein-coupled activation of phosphohpase C and the activation of voltage-dependent calcium Ion channels (9). The diverse functions ofET may be mediated by interaction with different receptors. Recently, it was rep~rted that ET induced the activation of phospholipase A2 In cultured smooth muscle (46). Pretreatment with phorbol ester inhibited ET-induced inositol phosphate formation but potentiated ET-stimulated AA release (47). The~e findings suggest that ET elicits activation of phosphohpase A2 and phospholipase C through independent pathways. The free AA, released by phospholipase A2 from membrane phospholipids, may be subsequently metabolized by cyclooxygenase and lipoxygenase pathway enzymes. However, it is not clear whether ET-l increases eicosanoid production only by activating phospholipase A2 and/or phospholipase C or by stimulating cyclooxygenase and lipoxygenase enzymes as well. A recent study has reported that ET-l-induced constriction of guinea pig airways was partially attenuated by BN 52021, a PAF antagonist (14). This finding suggested that PAF may mediate a portion of ET-induced airway constriction. In our study, the effect of Ro 19-3704, a potent structurally related PAF antagonist, on ET-induced eicosanoid production in cultured human nasal mucosa was examined and no effect was found. 15-HETE has been reported to be the major AA metabolite from human nasal and bronchial epithelia (48). This present study documents thy release of several eicosanoids from human nasal mucosa explants following exposure to ET-l. Although significant amounts of 15-HETE were generated, cyclooxygenase product generation was most remarkable. Eicosanoid release after ET-l exposure (10 to 0.1 J.'M) is concentration dependent, and occurs within 1 h. Whereas prostanoid production was maximal 1 h after exposure to E.T-l, 15-~ETE release was maximal at 4 h. Thus, ET-l may directly stimulate some responses and amplify these direct actions by the simultaneous generation of a variety of potent eicosanoids. References Yanagisawa~

M., H. Kurihara, S. Kimura et al. 1988. A novel potent vasoconstnctor peptide produced by vascular endothelial cells. Nature 332:411-415. 2. Shigeno, T., T. Mirna, K. Takakuraetal. 1989. Endothelin-I acts in cerebral arteries from the adventitial but not from the luminal side. J. Cardiovasc. Pharmacol. 13(Suppl. 5):SI74-S176. 3. Uchida, Y., H. Ninomiya, M. Saotome et al. 1988. Endothelin, a novel vasoconstrictor peptide, as potent bronchoconstrictor. Eur. J. Pharma1.

174

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992

col. 154:227-228. 4. Payne, A., and B. Whittle. 1988. Potent cyclo-oxygenase-mediated bronchoconstrictor effects of endothelin in the guinea-pig in vivo. Eur. J. Pharmacol. 158:303-304. 5. Maggi, C. A., R. Patacchini, S. Giuliano, and A. Meli. 1989. Potent contractile effect of endothelin in isolated guinea pig airways. Eur. J. Pharmacol. 160:179-182. 6. Macquin-Mavier, I., M. Levarne, N. Istin, and A. Harf. 1989. Mechanisms of endothelin-mediated bronchoconstriction in the guinea pig. J. Pharmacol. Exp. 1her. 250:740-745. 7. Battistini, B., J. Filep, and P. Sirois. 1990. Potent thromboxane-mediated in vitro bronchoconstrictor effect of endothelin in the guinea pig. Eur. J. Pharmacol. 178:141-142. 8. Secrest, R., and M. Cohen. 1989. Endothelin: differential effects in vascular and nonvascular smooth muscle. Life Sci. 45:1365-1372. 9. Masaki, T., M. Yanagisawa, Y. Takuwa, K. Yoshitoshi, S. Kimura, and K. Goto. 1990. Cellular mechanism of vasoconstriction induced by endothelin. In The Biology and Medicine of Signal Transduction. Y. Nishizuka, M. Endo, and C. Tanaka, editors. Raven Press, New York. 425-428. 10. Filep, J., B. Battistini, and P. Siroris. 1990. Endothelin induces thromboxane release and contraction of isolated guinea-pig airways. life Sci. 47: 1845-1850. II. Hoffman, A., E. Grossman, K. Ohman, E. Marks, and H. Keiser. 1989. Endothelin induces an initial increase in cardiac output associated with selective vasodilation in rats. life Sci. 45:249-255. 12. Wright, C., and J. Fozard. 1988. Regional vasodilation is a prominent feature of the haemodynamic response to endothelin in anaesthetized, spontaneously hypertensive rats. Eur. J. Pharmacol. 155:201-203. 13. Lerman, A., F. Hilderbrand, K. Margulies et al. 1990. Endothelin: a new cardiovascular regulatory peptide. Mayo Clin. Proc. 65:1441-1455. 14. Battistini, B., P. Sirois, P. Braquet, andJ. Filep. 1990. Endothelin-induced constriction of guinea pig airways: role of platelet-activating factor. Eur. J. Pharmacol. 186:307-310. 15. Mullol, J., K. Ohkubo, D. Rieves et al. 1991. Endothelin in human nasal mucosa. J. Allergy Clin. Immunol. 87:217. (Abstr.) 16. Patow, C., J. Shelhamer, Z. Marom, C. Logun, and M. Kaliner. 1984. Analysis of human nasal mucous glycoproteins. Am. J. Otolaryngol. 5:334-343. 17. Lundgren, J., C. Wiederrnann, C. Logun, J. Plutchok, M. Kaliner, and J. Shelhamer. 1989. Substance P receptor mediated secretion of respiratory glycoconjugate from feline airways in vitro. Exp. Lung Res. 15: 17-29. 18. Holtzman, M., J. Hansbrough, G. Rosen, and J. Turk. 1988. Uptake, release and novel species-dependent oxygenation of arachidonic acid in human and animal airway epithelial cells. Biochim. Biophys. Acta 963: 401-413. 19. Eling, T., R. Danilowicz, D. Henke, K. Sivarajah, J. Yankaskas, and R. Boucher. 1986. Arachidonic acid metabolism by canine tracheal epithelial cells. J. Bioi. Chem. 261:12841-12849. 20. Chilton, F., 1. Westcott, L. Zapp, J. Henson, and N. Voelkel. 1989. Incorporation of arachidonic acid into lipids of the isolated perfused rat lung. J. Appl. Physiol. 66:2763-2771. 21. Doupnik, C., and G. Leikauf. 1990. Acrolein stimulates eicosanoid release from bovine airway epithelial cells. Am. J. Physiol. 259:L222-L229. 22. Churchill, L., F. Chilton, J. Resau, R. Bascom, W. Hubbard, and D. Proud. 1989. Cyclooxygenase metabolism of endogenous arachidonic acid by cultured human tracheal epithelial cells. Am. Rev. Respir. Dis. 140:449-459. 23. Powell, W. 1985. Reversed-phase high-pressure liquid chromatography of arachidonic acid metabolites formed by cyclooxygenase and lipoxygenases. Anal. Biochem. 148:59-69. 24. Holtzman, M. 1991. Arachidonic acid metabolism: implications ofbiological chemistry for lung function and disease. Am. Rev. Respir. Dis. 143: 188-203. 25. Goodwin, J. 1989. Immunomodulation by eicosanoids and anti-inflammatory drugs. Curro Opin. Immunol. 2:264-268. 26. O'Byrne, P., H. Aizawa, R. Bethel, K. Chung, J. Nadel, and M. Holtzman. 1984. Prostaglandin F,. increases responsiveness of pulmonary airways in dogs. Prostaglandins 28:537-543.

27. Coleridge, H., J. Coleridge, K. Ginzel, D. Baker, R. Banzett, and M. Morrison. 1976. Stimulation of irritant receptors and afferent C-fibers in the lungs by prostaglandins. Nature 264:451-453. 28. Seeger, W., H. Walter, N. Suttorp., M. Muhly, and S. Bhakdi. 1989. Thromboxane-mediated hypertension and vascular leakage evoked by low doses of Escherichia coli hemolysin in rabbit lungs. J. Clin. Invest. 84: 220-227. 29. Marom, Z., J. Shelhamer, F. Sun, and M. Kaliner. 1983. Human airway monohydroxyeicosatetraenoic acid generation and mucus release. J. Clin. Invest. 72:122-127. 30. Setty, B., J. Graeber, and M. Stuart. 1987. The mitogenic effect of 15- and 12-hydroxyeicosatetraenoic acid on endothelial cells may be mediated by diacylglycerol kinase inhibition. J. BioI. Chem. 262:17613-17622. 31. Vanderhoek, J., N. Tare, J. Bailey, A. Goldstein, and D. Pluznik. 1982. New role for 15-hydroxyeicosatetraenoic acid. Activator of leukotriene biosynthesis in PT-18 mast/basophil cells. J. BioI. Chem. 257:1219112195. 32. Setty, B., and M. Stuart. 1986. 15-hydroxy-5,8,ll,13-eicosatetraenoic acid inhibits human vascular cyclooxygenase. J. Clin. Invest. 77:202211. 33. Baraniuk, J., J. Lundgren, J. Goff et al. 1990. Calcitonin gene-related peptide in human nasal mucosa. Am. J. Physiol. 258:L81-L88. 34. Baraniuk, J., J. Lundgren, J. Goff et al. 1990. Gastrin-releasing peptide in human nasal mucosa. J. Clin. Invest. 85:998-1005. 35. Baraniuk, J., J. Lundgren, M. Okayama et al. 1990. Vasoactive intestinal peptide in human nasal mucosa. J. Clin. Invest. 86:825-831. 36. Baraniuk, J., S. Castellino, J. Lundgren et al. 1990. Neuropeptide Y (NPY) in human nasal mucosa. Am. J. Respir. Cell Mol. BioI. 3: 165-173. 37. Inoue, A., M. Yanagisawa, S. Kimuraetal. 1989. The human endothelin family: three structurally and pharmacologically distinct isopeptides predicted by three separate genes. Proc. Natl. Acad. Sci. USA 86:28632867. 38. Uchida, S., M. Naruse, E. Ogata, M. Yanagisawa, T. Masaki, and K. Kurokawa. 1990. Gene expression and synthesis of endothelin in renal tubular epithelial cells in culture. Kidney Int. 37:380. (Abstr.) 39. Black, P., M. Ghatei, K. Takahashi et al. 1989. Formation of endothelin by cultured airway epithelial cells. FEBS Lett. 255: 129-132. 40. Mattoli, S., M. Mezzetti, G. Riva, L. Allegra, and A. Fasoli. 1990. Specific binding of endothelin on human bronchial smooth muscle cells in culture and secretion of endothelin-like material from bronchial epithelial cells. Am. J. Respir. Cell Mol. Bioi. 3:145-151. 41. MacCumber, M., C. Ross, B. Glaser, and S. Synder. 1989. Endothelin: visualization of mRNAs by in situ hybridization provides evidence for local action. Proc. Nail. Acad. Sci. USA 86:7285-7289. 42. Power, R., J. Wharton, Y. Zhao, S. Bloom, andJ. Polak. 1989. Autoradiographic localization of endothelin-l binding sites in the cardiovascular and respiratory systems. J. Cardiovasc. Pharmacol. 13(Suppl. 5):S50S56. 43. Hoyer, D., C. Waeber, and J. Placios. 1989. p"I)Endothelin-l binding sites: autoradiographic studies in the brain and periphery of various species including humans. J. Cardiovasc. Pharmacol. 13(Suppl. 5):SI62S165. 44. Davenport, A., D. Nunez, J. Hall, A. Kaumann, and M. Brown. 1989. Autoradiographical localization of binding sites for porcine p"I)endothelin-I in humans, pigs, and rats: functional relevance in humans. J. Cardiovasc. Pharmacol. 13(Suppl. 5):SI66-S170. 45. Martin, E., P. Marsden, B. Brenner, and B. Ballermann. 1989. Identification and characterization of endothelin binding sites in rat renal papillary and glomerular membranes. Biochem. Biophys. Res. Commun. 162: 130-137. 46. Resink, T., T. Scott-Burden, and F. Buhler. 1989. Activation of phospho lipase A, by endothelin in cultured vascular smooth muscle cells. Biochem. Biophys. Res. Commun. 158:279-286. 47. Reynolds, E., L. Mok, and S. Kurosawa. 1989. Phorbol ester dissociates endothelin-stimulated phosphoinositide hydrolysis and arachidonic acid release in vascular smooth muscle cells. Biochem. Biophys. Res. Commun. 160:868-873. 48. Henke, D., R. Danilowicz, J. Curtis, R. Boucher, and T. Eling. 1988. Metabolism of arachidonic acid by human nasal and bronchial epithelial cells. Arch. Biochem. Biophys. 267:426-436.