Furosemide Differentially Relaxes Airway and Vascular Smooth Muscle in Fetal, Newborn, and Adult Guinea Pigs 1 - 4 EDWARD L. STEVENS, CATHERINE F. T. UYEHARA, W. MICHAEL SOUTHGATE, and KENNETH T. NAKAMURA

Introduction

Although bronchopulmonary dysplasia (BPD) is characterized by lung parenchymal damage, there is substantial evidence to suggest that large airway dysfunction, often with symptoms similar to those of asthma, plays a significant role (1,2). Furosemide is the most widely used diuretic in newborns, in part because of its beneficial effect in babies with BPD (2, 3). In patients with BPD, furosemide was shown by Kao and colleagues (2) to acutely decrease airway resistance. Najak and coworkers (3) found that improvement in pulmonary function with furosemide occurred before the onset of diuresis, suggesting the presence of a direct pulmonary effect. Recently, studies in adults reported benefits employing nebulized furosemide in reducing airway hyperresponsiveness to exercise (4), ultrasonically nebulized distilled water (5, 6), and metabisulfite-, methacholine- (7), and allergen-induced bronchospasm (8). Such studies suggest a potential future mode of therapy in neonates. However, there is at present no information regarding the effect of age on furosemide-mediated relaxation in airway hyperresponsiveness. Furthermore, furosemide has been shown to inhibit aortic (9) and pulmonary arterial (10) smooth muscle contractility, suggesting that vascular tone may also respond to this therapy. Little is known about the direct effect of furosemide on smooth muscle of immature animals. Therefore, this developmental study was designed to answer the following questions: (1) Does furosemide relax airway smooth muscle preconstricted with physiologicconstrictors? (2)Does furosemide relax vascular smooth muscle? (3) Will ontogenetic changes in responsiveness of airway smooth muscle to furosemide differ from those of vascular smooth muscle? Methods Extrathoracic trachea, mainstem bronchi, 1192

SUMMARY Furosemide, an Inhibitor of CI-dependent Na..,K+ cotranspart, Is the most frequently used diuretic in newborns. Recently, furosemide was also demonstrated to decrease bronchial hyperresponsiveness In adUlts, although little Is known about the direct effect of furosemide on smooth muscle of Immature animals. This in vitro study was designed to determine the action of furosemide on airway and vascular smooth muscle during ontogeny. Extrathoraclc trachea (ET), main stem bronchi, main pUlmonary artery, and thoracic aorta ring segments from fetal, newborn, and adult Hartley albino guinea pigs were suspended In HEPES solution for measurement of Isometric tension. Furosemide (30 or 300 J,l.M) was administered after preconstrlctlon with an ED35 - 7oconcentration of histamine or acetylcholine for airway and ED~o-loo concentration of norepinephrine for vessels. Furosemide (30 J,l.M) caused significant relaxation of airway smooth muscle at all ages. After histamine-Induced preconstrlctlon, fetal airway segments exhibited greatest relaxation (183 ± 28%), with newborn airway demonstrating 123 ± 15% relaxation and modest relaxation seen In adults (40 ± 4%). This pattern was similar for both ET and bronchus and appeared greater for histamine compared with ACh preconstrlctlon. Epithelial removal slightly enhanced relaxation. Furosemide also relaxed pulmonary artery segments, but at a 10-fold higher concentration. In striking contrast to the pattern seen in airway, adult pulmonary artery relaxed more than newborn and newborn, more than fetus. Cyclooxygenase blockade and endothelium removal did not change pUlmonary artery relaxation. Furosemide did not significantly relax aorta after NE preconstrlction. Taken together, these results suggest that furosemide may be more effective In relaxing airway compared with vascular smooth muscle, and the ontogeny of these responses indicates a greater efficacy AM REV RESPIR DIS 1992; 146:1192-1197 and selectivity In airways of Immature animals.

main pulmonary artery (PA), and thoracic aorta of Hartley albino guinea pigs from three age groups werestudied: fetuses of either sex at 55 to 60 days of gestation (82 ± 4 g), term being 68 days; 2- to 5-day-old newborns of either sex (108 ± 5 g); and 6-wk-old adult males (457 ± 9 g). Newborns and adults were sedated with 25 mg/kg of ketamine hydrochloride (Vetalar'"; Parke-Davis) and killed with 1 g/kg and 500 mg/kg, respectively, of sodium pentobarbital (Wyeth, Philadelphia, PA). Pregnant animals weresedated with ketamine hydrochloride (25 mg/kg) and killed with 1 mg/kg of sodium pentobarbital. Fetuses were immediately delivered and killed with 1 g/kg of sodium pentobarbital. Tissues wereremoved,dissected free of connective tissue. cut into 3- to 4-mm ring segments, and placed in a cold physiologic buffer solution (pH = 7.4): 140 mM NaCl, 4.5 mM xci, 10 mM D-glucose, 5 mM HEPES, 1.5 mM CaCI 2·2H2 0 , and 1 mM MgC12·6H2 o. HEPES solution was chosen because Deth and coworkers suggested greater inhibition of Na,K cotransport by furosemide in HEPES compared with Krebs bicarbonate buffer (9). Our experiments confirm this finding (figure 1). In some tissue, epithelium or endothelium was removed by gentle rubbing with a wooden probe. Ring segments were mounted

on two fine stainlesssteelwirespassed through the lumen and connected to a Grass FT.03 force transducer (Grass, Quincy, MA) coupled to a Gould 2600S pen recorder (Gould, Cleveland. OH) for continuous measurement

(Received in original form September 3, 1991 and in revised form March 12, 1992) 1 From the Departments of Clinical Investigation and Pediatrics, Tripler Army Medical Center, TAMC, and the Department of Pediatrics, Kapiolani Medical Center for Women and Children, John A. Burns School of Medicine, Honolulu, Hawaii. 2 The opinions or assertions contained herein are the private views of the authors and are not to be considered as official or as reflecting the view of the Department of the Army or the Department of Defense. 3 Supported by the U.S. Army Health Services Command, the Research Corporation of the University of Hawaii Leahi Trust, the Research Board, Kapiolani Medical Center, and Grant No. HL-45220 from the National Heart, Lung, and Blood Institute. 4 Correspondence and requests for reprints should be addressed to KennethT. Nakamura, M.D., Department of Pediatrics, Kapiolani Medical Center for Women and Children, 1319 Punahou Street, Room 731, Honolulu, HI 96826.

1193

EFFECT OF FUROSEMIDE ON AIRWAY AND VASCULAR SMOOTH MUSCLE

TABLE 1

dure (13). Maximum responses between furosemide doses were compared with t tests. p values < 0.05 were considered significant. Values are expressed as mean ± standard error of the mean (SEM).

OPTIMAL RESTING TENSION AND ACTIVE TENSION GENERATED ACCORDING TO AGE, TISSUE, AND PRECONSTRICTING DRUG*

Age Fetus

Newborn

Adult

Active Tension (g)

Tissue (n)

Drug

ET (17) ET (18) BR (15) PA (8) AO(5}

ACh HIST HIST NE NE

1.14 1.13 1.09 1.26 2.52

± ± ± ± ±

0.07 0.05 0.06 0.08 0.05

0.76 0.71 0.68 1.29 0.92

± ± ± ± ±

ET (18) ET (21) BR (18) PA (11) AO(6}

ACh HIST HIST NE NE

1.46 1.40 1.39 2.01 1.90

± 0.04 ± 0.07 ± 0.05 ± 0.11 ± 0.04

1.51 1.09 0.78 1.04 2.08

± ± ± ±

ET (18) ET (21) BR (20) PA (9) AO(5}

ACh HIST HIST NE NE

2.88 2.90 3.06 3.41 3.24

± ± ± ±

2.99 2.90 1.45 0.88 2.08

± ± ± ±

ORT (g)

± 0.09 0.07 0.09 0.09 0.14

Results

0.09 0.08 0.09 0.14 0.18

± 0.22 0.11 0.06 0.12 0.12

± 0.27 0.20 0.12 0.12 0.20

Definition of abbreviations: OAT = optimal resting tension; ET = extrathoracic trachea; SA = bronchus; PA = pulmonary artery; AO = aorta; ACh = acetylcholine (3 x 10-1 M); HIST = histamine (3 x 10-8 M); NE = norepinephrine (10-8 M newborns and adults, 3 x 10-1M fetal PA, and 10-5 M fetal AO); (n) = number of ring segments. • Values are mean ± SEM.

of isometric force (11). Ring segments were suspended in a 25 ml organ bath (370 C), bubbled with 100070 O 2 , and equilibrated for 60 min with a bath change every 15 min. Ring segments were stretched to their optimal resting tension as determined by maximal force developed to 3 x 10-6 M acetylcholine (ACh) for airway and 10-6 M norepinephrine (NE) for vasculature. Preparations were washed and optimal resting tension reattained and stabilized before and between procedures. Extrathoracic trachea and bronchial ring segments werepreconstricted with a 35to 70% effective dose (ED 35 - 7o ) of either ACh (3 x 10-6 M) or histamine (3 x 10-6 M). Experiments with histamine were conducted in the presence of 10-8 M atropine to prevent the possible neuronal releaseof ACh by histamine (12).Vessel segments were preconstricted with an ED 4 o- 1o o of NE (3 x 10-6 M for fetal PA, 10-5 M for fetal aorta, and 10-6 M in newborns and adults), the concentration required to maintain equilibrium tension over time. Tissues were exposed to furosemide doses when equilibrium tension was stabilized after preconstriction. Normal saline with pH adjusted to equal that of the furosemide solution was employed in control experiments. Optimal resting tension and active equilibrium tension for each tissue type and constricting agent can be seen in table 1. This study was approved by the Animal Care and Use Committee, Tripler Army Medical Center. Procedures on the guinea pigs were in accordance with National Institutes of Health policies and the Guidefor the Care and UseofLaboratory Animals (NIH Publication No. 85-23, revised 1985).

Drugs The following drugs were used: NE hydrochloride, ACh chloride, atropine, histamine, acetylsalicylic acid (Sigma Chemical Company, St. Louis, MO), and furosemide (Lasixs; Hoechst-Roussel Pharmaceuticals, Inc., Somerville, NJ). Solutions for NE, ACh, atropine, and histamine were prepared in distilled water on the day of the experiment and kept on ice. Doses were administered in 1oo-1l1 aliquots, and drug concentrations are expressed as final molarity in the bath. Furosemide (2 mg/lO ml) was administered as a 25-1l1 aliquot to achieve a 30 IlM bath concentration and a 248-1l1 aliquot for a final 300 J.1M bath concentration. Statistical Analysis Time-dependent changes in tension after administration of furosemide or control solution for different age groups were compared by two-way analysis of variance (ANOVA) with Duncan's multiple-comparison proce-

Effect of Furosemide on Extrathoracic Trachea Preconstricted with Histamine After histamine-induced preconstrietion, statistically significant furosemide- mediated (30 IlM) relaxation was seen in the fetus (153 ± 10%) and newborn (73 ± 11%) by 10 min and in the adult by 15 min (9 ± 3070) (figure 2). By 35 minutes, furosemide produced dramatic relaxation in fetal (183 ± 28070) (figure 3) and newborn (123 ± 15070) ring segments and modest relaxation in adults (40 ± 4070). The effect of a 10-fold increase in furosemide concentration is demonstrated in figure 4. Increased relaxation was seen in the adult 20 min after addition of 300 IlM furosemide (74 ± 10070) versus 30 IlM (18 ± 2%), indicating a dosedependent effect (P < 0.05).Although not statistically significant, dose dependence was also suggested in the newborn (138 ± 13% versus 119 ± 16%) and fetus (201 ± 26% versus 183 ± 28%). To determine the effect of epithelial removal on furosemide-mediated relaxation, additional paired experiments were performed on fetal extrathoracic trachea to compare relaxation after 3, 10,30, and 100 IlM furosemide administered to ring segments with and without epithelium. Optimal resting tension and active tension generated were 1.3 ± 0.1 and 0.9 ± 1.8 with intact epithelium and 1.4 ± 0.1 and 0.7 ± 0.1 g without epithelium, respectively. Shown in figure 5 is the concentration-response effect with and without epithelium. Although not statistically significant, a trend toward greater relaxation was observed when the epithelium was removed. TIME FROM ADMINISTRATION OFFUROSEMIDE (min)

-5 o 5 10 15 20 25 30 J5 25+---+---+----+--1---+---+---+---1

z

a

o

~

-2S

Vi

Fig. 1. Comparison of response in Krebs versus HEPES solution to furosemide (30 J.1M) in newborn guinea pig extrathoracic trachea after preconstriction with histamine. n = number of animals; values are mean ± SEM.

~

-50

~

-75

::IE

a

-100

..... C)

-125

fE

~

u

~

,

-.

,T

....

.... T

·~~~.T

-lSO

-175 -200

0- newborn > adult) demonstrates that furosemide has a direct effect on airway that diminishes during development. Although the exact mechanism of furosemide-mediated relaxation in airway smooth muscle is not known (14), one postulate is by inhibition of the electrically neutral CIdependent Na+,K+ cotransport (9). A 30 J..1M concentration as employed in this study could inhibit this mechanism (15), suggesting a relation to decreased contractility. Inhibition of Cl-dependent Na+,K+ cotransport could reduce intracellular sodium, consequently increasing extracellular Na' and intracellular Ca' exchange and thereby reducing contractility (16, 17). Inhibition of Na" influx by furosemide also blocks 45Ca uptake (9), although the ion-exchange mechanism is not known. Recently, Elwood and coworkers demonstrated in guinea pig airway smooth muscle that furosemide inhibits cholinergic and noncholinergic contraction induced by electrical field stimulation in vitro in a dose-dependent manner (18, 19). In contrast, Knox and Ajao (20) found no airway relaxation with 10-5 M furosemide in bovine tracheal smooth muscle strips without epithelium after preconstriction with histamine, potassium chloride, or hypertonic saline. Furosemide was also without effect on hypertonic saline-induced contractions of human bronchial rings. In addition, Elwood and coworkers (19) found no inhibition by furosemide to contraction induced with exogenous acetylcholine in guinea pigs. We speculate that although their

STEVENS, UYEHARA, SOUTHGATE, AND NAKAMURA

1196 TIME FROM ADMINISTRATION OF FUROSEMIDE (min) -5

o

5

10

15

20

25

30

35

25

z

0

enz

~

0 -25

~

-50

~

-75

~

::J:

-

.~i;,;s;~~:~:::::~:::·····9········9········g:·.:·.:.·~·.o

- --·8. - - - .8~1~':::=.-r~~1 --_ .. ~::::::,

0···· 0 Fetal, pH-control (n=2)

UJ -125

u..

O' - 0 Newborn, pH-control (n=2)

o

0 - 0 Adult, pH-control (n-2)

l:t:

~

U

M

-150 -175

--'1'---'

NE preconstriction:

-100

0

.I.

•....• Fetus, Furosemide 300 JJJtA (n=3) . - - .• Newborn, Furosemide 300 JJJtA (n=4) • - . Adult, Furosemide 300 JJJtA (n=3)

-200

Fig. 9. Effect of furosemide on guinea pig aorta after preconstriction with the ED4o-10o of norepinephrine (NE). n = number of animals; values are mean ± SEM.

methods (19, 20) were somewhat similar to ours, differences in age, species (20), and, in particular, the physiologic buffer solution employed may be responsible for the contrasting findings. For example, our experiments demonstrated a dramatically decreased relaxation with 30 IlM furosemide using Krebs buffer solution compared with HEPES (figure 1). The mechanism by which HEPES may allow furosemide to have a greater effect on smooth muscle was discussed by Deth and coworkers (9), who postulated that the absence of bicarbonate in HEPES is the basis for this observation. With Krebs buffer solution, increased CO 2 enters the cell, causing intracellular bicarbonate formation with H+ generation, which is extruded via increased Na"/H+ exchange. In addition, Na+,K+ countertransport activity is increased (increased ouabain-sensitive 86Rb uptake), with a net reduction in the furosemide sensitive Cl-dependent Na+,K+ cotransport system (decreased furosemide-sensitive 86Rb uptake). Another mechanism by which furosemide may exert its effects could be by alterations in epithelial function since the role of respiratory epithelium in the modulation of smooth muscle activity may involve several mechanisms, which include, among others, release of relaxing factors, acting as a simple barrier, and release of contractile factors (21). Removal of the epithelium slightly increased the furosemide-mediated relaxation of histamine-preconstricted trachea, although this effect was not statistically significant. Thus although the epithelium may modulate furosemide-mediated airway smooth muscle relaxation, significant relaxation still occurred in airway tissue without epithelium, suggesting that the epithelial

cell is not essential for furosemide's actions. We chose the 30 IlM concentration of furosemide because it approximates the plasma concentration expected in a newborn after intravenous administration of 1 mg/kg of furosemide (22, 23). It is not known what concentrations are achieved in airway smooth muscle with different routes of administration, but one can speculate that airway concentrations could be higher with nebulized compared with intravenous therapy. On the other hand, the minimal concentration of furosemide required to cause relaxation of immature airways may be several logarithmic units less than the 30 IlM concentration employed. The similarity in relaxation between 30versus 300 IlM concentrations in the fetus and newborn, together with the marked relaxation observed with 30 IlM, suggests these concentrations are at the upper end of a dose-response curve. The effect of furosemide on vascular smooth muscle differed from that on airway smooth muscle. Pulmonary artery was less sensitive to the relaxant effect of furosemide than airway, where relaxation of the pulmonary artery seen at a 10-fold higher concentration (300 IlM) was less than the response to 30 IlM seen in airways. An ontogenetic pattern opposite that seen for airway became apparent, with a trend for adult relaxation greater than newborns and both greater than the fetus. Lundergan and coworkers demonstrated a reduction in mean pulmonary perfusion pressure in the isolated canine lung lobe after 10-5 M furosemide (10). This response was eliminated by pretreatment with 5 mg/kg of indomethacin, suggesting the vasodila-

tory effect is due to increased production of an endogenous cyclooxygenase product. However, the mechanisms and even the presence of a direct effect of furosemide on vasodilator activity remain controversial (24). The results here support the notion that furosemide has a direct vasodilator activity, but the mechanism of action does not require the presence of endothelium or cyclooxygenase products. Furosemide may be somewhat selective for the pulmonary bed, since no significant relaxation was seen with 300 IlM furosemide in adult thoracic aorta and only minimal relaxation was observed in fetal and newborn aorta. Species differences may also affect responses: Deth and coworkers (9) noted dose-dependent relaxation of rat aorta after furosemide (10-5 , 10-\ and 10-3 M) but no significant relaxation in rabbit aorta after furosemide (10-4 M), suggesting that guinea pig and rabbit aorta may respond similarly. In summary, our study demonstrates that (1) furosemide directly relaxes airway smooth muscle with an ontogenetic pattern of fetal> newborn> adult; (2) epithelium is not required for airway smooth muscle relaxation; (3) furosemide also relaxes pulmonary artery, but a higher concentration is required; (4) cyclooxygenase products and endothelium do not affect this relaxation; (5) in contrast to airways, the ontogenetic pattern of pulmonary artery relaxation is adult > newborn> fetus; and (6) furosemide does not significantly relax aorta. Taken together, these results suggest that furosemide may be more effective in relaxing airway than vascular smooth muscle, and the ontogeny of these responses indicates a greater efficacy and selectivity in airways of immature animals. Acknowledgment The authors appreciate the expert technical assistance of Naomi Fujiwara and the advice of Dr. John R. Claybaugh. References 1. MotoyamaEK, Brody JS, Colten HR, Warshaw JB. Postnatal lung development in health and disease. Am Rev Respir Dis 1988; 137:742-6. 2. Kao LC, Warburton D, Sargent CW, Platzker ACG, Keens TO. Furosemide acutely decreases airways resistance in chronic bronchopulmonary dysplasia. J Pediatr 1983; 103:624-9. 3. Najak ZD, Harris EM, Lazzara A Jr, Pruitt AW. Pulmonary effects of furosemide in preterm infants with lung disease. J Pediatr 1983; 102: 758-63. 4. Bianco S, Robuschi M, Vaghi A, Pasargiklian M. Prevention of exercise-inducedbronchoconstric-

EFFECT OF FUROSEMIDE ON AIRWAY AND VASCULAR SMOOTH MUSCLE

tion by inhaled furosemide. Lancet 1988;2:252-5. 5. Robuschi M, VaghiA, Gambaro G, Spagnotto S, Bianco S. Inhaled furosemide (F) is highly effective in preventing ultrasonically nebulized water (UNH 2 0 ) bronchoconstriction. Am Rev Respir Dis 1988; 137(4 part 2:412). 6. Moscato G, Dellabianca A, Falagiani P, Mistrello G, Rossi G, Rampulla C. Inhaled furosemide prevents both the bronchoconstriction and the increase in neutrophil chemotactic activity induced by ultrasonic "fog" of distilled water in asthmatics. Am Rev Respir Dis 1991; 143:561-6. 7. Nichol GM, Alton EW, Nix A, Geddes OM, Chung KF, Barnes PJ. Effect of inhaled furosemide on metabisulfite and methacholine induced bronchoconstriction and nasal potential difference in asthmatic subjects. Am Rev Respir Dis 1990; 142:576-80. 8. Bianco S, Pieroni MG, Refini RM, Rottoli L, Sestini P. Protective effect of inhaled furosemide on allergen-induced early and late asthmatic reactions. N Engl J Med 1989; 321:1069-73. 9. Deth RC, Payne RA, Peecher DM. Influence of furosemide on rubidium-86 uptake and alphaadrenergic responsiveness of arterial smooth mus-

cleo Blood Vessels 1987; 24:321-33. 10. Lundergan CF, Fitzpatrick TM, RoseJC, Ramwell PW, Kot PA. Effect of cyclooxygenase inhibition on the pulmonary vasodilator response to furosemide. J Pharmacol Exp Ther 1988; 246:102-6. 11. Balaraman V,Kullama LK, Easa 0, Robillard JE, Hashiro GM, Nakamura KT. Developmental changes in sodium nitroprusside and atrial natriuretic factor mediated relaxation in the guinea pig aorta. Pediatr Res 1990; 27:392-5. 12. Shore S, Irvin CG, Shenkier T, Martin JG. Mechanisms of histamine-induced contraction of canine airway smooth muscle. J Appl Physiol1983; 55:22-6. 13. Winer BJ. Statistical principles in experimental design, 2nd ed. New York: McGraw-Hill. 1971; 196-7. 14. Inhaled frusemide and asthma (editorial). Lancet 1990; 335:944-6. 15. Chipperfield AR. The (Na+-K+-CI-) cotransport system. Clin Sci 1986; 71:465-76. 16. Scheid CR, Honeyman TW, Fay FS. Mechanism of 13-adrenergic relaxation of smooth muscle. Nature 1979; 277:32-6. 17. Van Breemen C, Aaronson P, Loutzenhiser R.

1197

Sodium-calciuminteractions in mammalian smooth muscle. Pharmacol Rev 1979; 30:167-208. 18. ChungKF, BarnesPJ. Inhaledfrusemideand asthma (letter). Lancet 1990; 335:1539. 19. Elwood W, Lotvall JO, BarnesPJ, ChungKF. Loop diuretics inhibit cholinergic and noncholinergic nerves in guinea pig airways. Am Rev Respir Dis 1991; 143:1340-4. 20. Knox AJ, Ajao P. Effect of frusemide on airway smooth muscle contractility in vitro. Thorax 1990; 45:856-9. 21. Vanhoutte PM. Epithelium-derived relaxing factor(s) and bronchial reactivity. Am Rev Respir Dis 1988; 138:S24-30. 22. Aranda JV, Perez J, Sitar DS, et al. Pharmacokinetic disposition and protein binding of furosemide in newborn infants. J Pediatr 1978; 93:507-11. 23. Roberts RJ. Drug therapy in infants: pharmacologic principles and clinical experience. Philadelphia: W. B. Saunders, 1984; 226-49. 24. Gerkens JF. Does furosemide have vasodilator activity? Trends Pharmacol Sci 1987;8:254-7.