UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Liquid movement and electrical potential difference across in vitro lungs from fetal guinea pigs : responses… Kojwang, David Ogweno 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-345688.pdf [ 7.42MB ]
Metadata
JSON: 831-1.0088977.json
JSON-LD: 831-1.0088977-ld.json
RDF/XML (Pretty): 831-1.0088977-rdf.xml
RDF/JSON: 831-1.0088977-rdf.json
Turtle: 831-1.0088977-turtle.txt
N-Triples: 831-1.0088977-rdf-ntriples.txt
Original Record: 831-1.0088977-source.json
Full Text
831-1.0088977-fulltext.txt
Citation
831-1.0088977.ris

Full Text

LIQUID M O V E M E N T A N D E L E C T R I C A L P O T E N T I A L D I F F E R E N C E A C R O S S IN VITRO L U N G S F R O M F E T A L G U I N E A PIGS: R E S P O N S E S T O E X P A N S I O N A N D R E L A T E D F A C T O R S by D A V I D O G W E N O K O J W A N G B . E d (Sc.), Kenyatta University, 1984 M . S c , The University of British Columbia, 1991 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E STUDIES (Department of Zoology) We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y OF BRITISH C O L U M B I A May 1998 © David Ogweno Kojwang, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Z Q Q U 6 OcV The University of British Columbia Vancouver, Canada Date J U L " / 23, » W DE-6 (2/88) Abstract Lungs from near-term fetal guinea pigs were supported in vitro; lung liquid production was measured by a dye dilution method using Blue Dextran 2000; electrical potential difference (PD) across fetal airways was measured using KCl -Agar salt bridges. Untreated (control) preparations produced fluid for three hours with no significant change in rate; unstimulated P D was -7.5 ± 0.4 m V (lumen -ve). 70% expansion with Krebs-Henseleit (K-H) saline decreased fluid production (53.5 ± 6.6%) in preparations from small fetuses (<100 g) and caused fluid reabsorption (124.8 ± 12.1%) in preparations from large fetuses (>100 g). Expansion did not cause any detectable change in bronchial PD. Glucose concentration in lung liquid (0.19 ± 0.03 mM) was lower than that of fetal plasma (1.35 ± 0.11 mM). The use of 0.9% N a C l as expansion fluid, to minimize possible effects of intraluminal glucose, had no effect on the response to expansion, and intraluminal instillation of amiloride (10"6 M or 10"5 M ) did not reduce expansion-induced reabsorption significantly. Neither the (3-adrenergic antagonist propranolol (10"7 M ) nor the a antagonist phentolamine (10"5 M ) inhibited fluid reabsorption following expansion. Untreated control preparations, incubated in pairs, showed no significant changes in fluid production throughout incubation. When one preparation was expanded by 70% with K - H saline, the expanded preparation reabsorbed fluid (125.4 ± 42.6%), and the unexpanded companion lungs reduced fluid production (88.2 ± 10.7%). This suggested the release of an active agent by expanded Ill lungs. Amiloride (10"6 M ) or propranolol (10'7 M ) reduced the effects in unexpanded companion lungs but not in expanded preparations. Lung liquid norepinephrine concentration was 97.0 ± 12.0 n M (n = 6) and increased to 190.0 ± 17.0 n M after 70% expansion; the concentration in the bathing saline was 1.5 ± 0.2 n M and did not increase significantly. Exogenous epinephrine (2.2 x 10"6 M ) and norepinephrine (10"5 M ) caused strong fluid reabsorptions (146.2 ± 23.7% and 137 ± 21.7% respectively). Intraluminal amiloride (10'5 M ) or 5-nitro-2-(3-phenylpropylamino) benzoic acid (NPPB, CI" channel blocker, 10"5 M ) did not reduce effects of norepinephrine. Epinephrine (10"7 M ) reduced bronchial P D by 27.4 ± 2.4% whereas norepinephrine (10"6 M ) reduced it by 31.4 ± 3.7%. Propranolol (10~7 M ) abolished these changes in P D while phentolamine (10"5 M ) , amiloride (10"4 M ) or N-phenylanthrilic acid (DPC, 10"4 M ) did not. A low level of somatostatin-like immunoreactivity was detected in the bathing saline; it did not increase with expansion. Although P G E 2 was not detected in the bathing saline, indomethacin (1.5 x 10"5 M ) reduced responses of unexpanded companion lungs. Exogenous prostaglandins had the following effects: P G F 2 a (5 x 10"7 M ) caused fluid reabsorption (130.8 ± 48.5%); P G E 2 (10"6 M ) halted fluid production (95.2 ± 26.2%) and reduced bronchial P D (33.6 + 7.5%); P G D 2 (10 - 6 M ) reduced production (51.4 ± 13.3%), and P G I 2 (10"6 M ) was without effect. The calcium ionophore, A23187 (10"5 M ) halted fluid production in the hour of application (95.0 ± 8.3% fall) and caused reabsorption in the final hour (140.5 ± 15.8%); Calcium-mediated secretagogues, histamine (10"6 M ) and bradykinin (10"8 M and 10"7 M ) had lesser but equally long-lasting effects. The histamine antagonist, mianserin, did not reduce effects of the unexpanded companion lungs significantly. iv These results suggest that: (1) Expanded lungs decrease fluid production, a response which increases with fetal maturation and turns to reabsorption. Expansion appears not to have any direct effects on airways, therefore, major influences of expansion are probably in alveolar cells. However, lesser influences due to factors including p-adrenergic receptors, prostaglandins and amiloride-sensitive N a + transport can operate, and are detected by companion lungs, possibly in their airways. Perhaps these factors are an insurance for fluid reabsorption, a vital process in the transition from intrauterine to extrauterine life. (2) In guinea pigs, catecholamines cause fluid reabsorption through a seemingly amiloride-insensitive pathway, perhaps the inhibition of CI" secretion. V Table of Contents Abstract ii Table of Contents v List of Figures vii List of Tables ix Acknowledgements x Dedication xi Chapter I General Introduction 1 Chapter II General Methods 8 2.1 Animals 9 2.2 Experimental Procedures 9 2.2 Quantification of results and statistical methods 12 Chapter III Effects of lung expansion on lung liquid production, and the possible involvement of somatostatin 14 3.1 Introduction 15 3.2 Materials and Methods 18 3.3 Results 23 3.5 Discussion 35 Chapter IV Effects of catecholamines on lung liquid production 40 4.1 Background 41 4.2 Materials and Methods 42 4.3 Results 46 4.4 Discussion 63 Chapter V Effects of prostaglandins on lung liquid production 70 5. / Introduction 71 5.2 Materials and Methods 73 5.3 Results 75 5.4 Discussion 86 Chapter VI Effects of calcium and its secretagogues on lung liquid production 94 6.1 Background 95 6.2 Materials and Methods 98 6.3 Results 99 6.4 Discussion 107 Chapter VII Transport mechanisms related to catecholamine stimulation and lung expansion: Effects of ion transport blockers and intraluminal glucose concentration 112 7.1 Introduction 113 7.2 Materials and Methods 115 7.3 Results 119 7.4 Discussion 138 Chapter VIII Transepithelial electrical potentials across fetal guinea pig lungs: Responses to catecholamines, expansion and other factors 146 8.1 Introduction 147 8.2 Materials and Methods 149 8.3 Results 156 8.4 Discussion 172 Chapter IX General Discussion 183 References 190 Appendices 208 Vll List of Figures Fig 1: Effect of expansion with Krebs-Henseleit saline, on lung liquid production by in vitro lungs from near-term fetal guinea pigs 27 Fig 2: The effect offetal weight on responses of in vitro lungs from fetal guinea pigs to saline expansion 29 Fig 3: Influence of expanded lungs on lung liquid production by unexpanded companion lungs supported in the same saline in vitro 31 Fig 4: Levels of somatostatin-like immunoreactivity (SLI) in the outer saline before and after expansion of in vitro lungs from fetal guinea pigs 33 Fig 5: Effect of epinephrine and norepinephrine on lung liquid production in vitro by lungs from near-term fetal guinea pigs 50 Fig 6: The relationship between norepinephrine concentration and the fall in fluid production by in vitro lungs from fetal guinea pigs 52 Fig 7: Influence of phentolamine and propranolol on responses to lung expansion in vitro 54 Fig 8: Influence of expanded lungs on lung liquid production by unexpanded companion lungs supported in the same saline in vitro; studies in the presence of phentolamine and propranolol 56 Fig 9. Changes in concentrations of norepinephrine in the outer saline (a), and in lung liquid (b) in response to 71% lung expansion with Krebs-Henseleit saline 61 Fig 10; Effect of various prostaglandins on lung liquid production in in vitro lungs from fetal guinea pigs 80 Fig 11: Lung liquid production in in vitro lungs from fetal guinea pigs 82 Fig 12: Influence of expanded lungs on lung liquid production by unexpanded companion lungs supported in the same saline in vitro; studies in the presence of indomethacin 84 Fig 13: Effects of various calcium secretogogues on lung liquid production in vitro by lungs from near-term fetal guinea pigs 100 vni Fig 14: Influence of expanded lungs on lung liquid production by unexpanded companion lungs supported in the same saline in vitro; studies in the presence of mianserin 105 Fig 15: Influence of intraluminal amiloride on lung liquid production by expanded lungs and unexpanded companion lungs supported in the same saline in vitro 122 Fig 16: Comparison of effects of expansion with Krebs-Henseleit saline and 0.9% NaCl on lung liquid production by in vitro lungs from near-term fetal guinea pigs 126 Fig 17: Effects of amiloride on lung liquid production by in vitro lungs expanded with 0.9% NaCl 128 Fig 18: Effect of amiloride on lung liquid production by in vitro lungs treated with epinephrine and norepinephrine 134 Fig 19: Effects of intraluminal application of the chloride channel blocker, 5-nitro-2-(3-phenylpropylamino) benzoic acid (NPPB) on lung liquid production by in vitro lungs treated with norepinephrine 136 Fig 20: Schematic diagram of the apparatus for the measurement of electrical potential difference (PD) across in vitro lungs from fetal guinea pigs 154 Fig 21: Typical recording of electrical potential difference (lumen -ve, mV) across in vitro lungs from fetal guinea pigs 161 Fig 22: Potential difference (lumen -ve, mV) across in vitro lungs from fetal guinea pigs 163 Fig 23: Relationship between catecholamine concentration and change in potential difference (PD) across in vitro lungs from fetal guinea pigs 165 Fig 24: Comparison of norepinephrine-induced change in transmural PD (lumen -ve, mV) in the absence (A) and in the presence (B) of intraluminal amiloride 169 IX List of Tables Table I: Blocking effects of adrenergic antagonists, propranolol and phentolamine, on epinephrine- and norepinephrine-induced inhibition of transepithelial PD across in vitro lungs from fetal guinea pigs 167 Table II: Effects of the sodium channel blocker, amiloride, and the chloride transport blocker, DPC, on norepinephrine-induced inhibition of transepithelial PD across in vitro lungs from fetal guinea pig 168 Table III: Effects of lung expansion, and of PGEj on transepithelial PD across in vitro lungs from fetal guinea pigs 171 Acknowledgements First and foremost, I would like to thank my supervisor, Dr. Anthony M . Perks, for his support and guidance throughout my doctoral program. Many thanks to Beverly Chua and Sam Doe for their companionship in the lab, their contributions to our discussions, and for letting me use some of their control data to augment my own. I would also like to thank the following people for their various contributions: Sumi Aota and Joelle Harris for help with H P L C for catecholamines; Herminia Sy and Dr. Kenny K w o k for help with radioimmunoassay for somatostatin; Grant McClel land for help with the glucose assay; Joan Martin for help with measurements of electrical potential difference; Dr. David J. Randall for giving me working space and unlimited use of his computers; Dr. Maryanne R. Hughes for lots of advice and support. I would like to give special thanks to my wife, Juliane Kojwang, and our son, Koju, for their love, support and inspiration. xi Dedication To my parents, Rhoda and Walter Ojwang, and to all my brothers and sisters: Because of this work, most of you have not seen me in nearly a decade. To my wife, Julie, and our son, Koju: Y o u were as involved in this as I was. I love you all . Chapter I General Introduction 2 Mammalian lungs are filled with fluid throughout their intrauterine development. In sheep, fetal lung fluid is evident as early as 74 days gestation (term, 147 days); at this time the lungs contain approximately 1 ml of fluid, 30% of it in the trachea (Olver, Ramsden and Strang, 1981). Liquid production increases through most of gestation (Perks and Cassin, 1985a). Early studies in sheep suggested that the volume of lung liquid begins to drop at approximately 135 days (Kitterman, Ballard, Clements, Mescher and Tooley, 1979; Dickson, Maloney and Berger, 1986). However, more recent studies indicate that fluid production may continue right up to the onset of labour (Lines, Hooper and Harding, 1997). The liquid acts as a space filler and is essential for proper prenatal growth and development of pulmonary tissue (Alcorn, Adamson, Lambert, Maloney, Ritchie and Robinson, 1977; Fewell, Hislop, Kitterman and Johnson, 1983; Moessinger, Harding, Adamson, Singh and K i m , 1990). The chloride ion (CI") concentration in fetal lung fluid is high relative to plasma (Adams, Fujiwara and Rowshan, 1963; Adamson, Boyd, Piatt and Strang, 1969; Adamson, Brodecky, Lambert, Maloney, Ritchie and Walker, 1975; Perks, Dore, Dyer, Thorn, Marshall, Woods, Vanderhorst and Ziabaksh, 1990), and intra-alveolar instillation of potassium cyanide blocks secretion of fetal lung fluid (Olver and Strang, 1974). Studies on in vitro guinea pig lungs using sodium iodoacetate and ouabain provided further evidence that the secretory mechanism in fetal lungs utilizes metabolic energy, and is probably linked to a N a 7 K + ATPase pump (Perks, Thorn, Ruiz, Vanderhorst and Woods, 1989). Additional evidence of the mechanism of fluid production came from studies with loop diuretics. Furosemide and bumetanide inhibited lung fluid production in fetal sheep in vivo 3 (Cassin, Gause and Perks, 1986), and provided evidence for the involvement of N a + - K + - 2 G " cotransport in the secretory process. Further support for these observations have come from studies of in vitro lungs from fetal guinea pigs (Thom and Perks, 1990) and of developing fetal rat alveolar epithelial cells in organ culture (McCray and Welch, 1991; McCray, Bettencourt and Bastacky, 1992a). The evidence is conflicting over whether CI" secretion and Na+reabsorption are simultaneous events, or that secretion must be halted before reabsorption can occur. In fetal lambs late in gestation, intratracheal instillation of the N a + channel transport inhibitor amiloride ( lO^M) could increase the resting rate of lung liquid secretion (Olver, Ramsden, Strang and Walters, 1986). A similar tendency was seen when in vitro fetal guinea pig lungs, with low unstimulated rates of liquid production, were treated with amiloride (Vonder Muhl l and Perks, unpublished observations). These studies suggest ongoing N a + reabsorption during fluid production. However, in other studies of pre-term fetal lambs, amiloride did not affect fluid production when placed within the alveolar space (Olver and Robinson, 1986; Carlton, Cummings, Chapman, Poulain and Bland, 1992), suggesting that resting fluid production depends only on CI" secretion, and is not the net result of simultaneous CI" secretion and N a + reabsorption. This problem has not been resolved. In studies of in vitro lungs from fetal guinea pigs, 2,4-dinitrophenol (DNP) turned fluid production to reabsorption (Perks, Ruiz, Chua, Vonder Muhl l , Kindler and Blair, 1993); this led to the suggestion that whereas fluid production depended on energy from oxidative phosphorylation, the reabsorptive process could proceed with energy from glycolysis alone. 4 A model which may explain the secretory process has been suggested (Walters, Strang and Geubelle, 1985). In this model, a basolaterally placed N a + / K + ATPase pump generates a gradient for Na + , which enters the intracellular space across the basolateral membrane linked to CI". The build-up of CI" ions in the epithelium provides an electrochemical gradient for CI" to exit into the alveolar space across the apical membrane perhaps through specialized CI" channels. Late in gestation lung liquid secretion rates fall to approximately 50% of previous values (Lawson, Brown, Torday, Mandansky and Tauesch, 1978), and alveolar fluid volume can decrease by almost 30%o in fetal lambs (Dickson et al, 1986); spontaneous reabsorption of alveolar fluid begins some days before the onset of labour (Kitterman et al., 1979; Bland, Hansen, Haberken, Bressack, Hazinski, Raj and Golberg, 1982). This fall in fluid production and subsequent reabsorption has been attributed to the actions of epinephrine (Walters and Olver 1978, Brown, Olver, Ramsden, Strang and Walters, 1983). However, opinion is divided over whether circulating epinephrine concentrations increase at all before the onset of labour. Several studies suggest that fetal catecholamine concentrations increase with advancing gestational age (Comline and Silver, 1961; Lagercrantz and Bistoletti, 1973; Jones, 1980; Cohen etal, 1982; Lewis, Evans and Sischo, 1982). Other studies suggest the opposite (Brown et al, 1983; Palmer, Oakes, Lam, Oddie, Hobel and Fischer, 1984). Nevertheless, pulmonary epithelial sensitivity to epinephrine increases with gestational age, and fetal plasma concentration of catecholamines, especially epinephrine, rises dramatically during labour (Brown et al, 1983). In fetal sheep, epinephrine but not norepinephrine, 5 stimulates powerful reabsorption of fetal lung fluid (Walters and Olver, 1978; Brown et al, 1983). The P adrenergic agonist, isoproterenol, produces similar responses, and P receptor blockade with propranolol cancels the effect in sheep (Walters and Olver, 1978; Brown et al, 1983). This suggests that P receptors modulate fluid reabsorption in fetal sheep lungs. On the other hand, norepinephrine could only decrease lung liquid flow rate by 45%, but could not induce liquid reabsorption; propranolol could not cancel this effect (Higuchi, Murata, Miyake, Hessler, Tyner, Keegan and Porto, 1987). In studies of in vitro lungs from fetal guinea pigs, effects of epinephrine could be blocked with the a receptor blocker, phentolamine, but not with propranolol (Woods, Doe and Perks, 1997); the use of specific cc-adrenergic blockers suggested that epinephrine activates lung fluid reabsorption through a2-adrenoreceptors (Doe and Perks, 1998). Like epinephrine, norepinephrine produced lung liquid reabsorption in these lungs (Woods, Doe and Perks, unpublished observations). Thus the difference between studies in sheep and guinea pigs may reflect a species difference in fetal lung responses to a- and p-adrenergic stimulation. In sheep, epinephrine is thought to stimulate amiloride-sensitive N a + channels (Olver et al, 1986). This activation of N a + channels occurs as a result of a rise in the intracellular concentration of adenosine 3',5'-cyclic monophosphate (cAMP) (Walters, Ramsden and Olver, 1990). Increasing intracellular c A M P concentration decreases fluid production and causes reabsorption by in vitro lungs from fetal guinea pigs (Kindler, Ziabaksh and Perks, 1992). These events are thought to increase with gestational age (Barker, Brown, Ramsden, Strang and Walters, 1988, Walters et al, 1990, Rao and Cott, 1991). 6 Like catecholamines, lung expansion can cause liquid reabsorption. Instillation of saline into lungs of fetal goats caused liquid reabsorption provided lung liquid volume increased by more than 47% (Perks and Cassin, 1982; 1985a). These observations were confirmed in guinea pigs; in addition, fluid reabsorption by expanded lungs could be inhibited with amiloride (Garrad Nelson and Perks, 1996a; 1996b). In those studies, 70% lung expansion with saline produced only small and transient increases in intratracheal pressure, and did not produce any discernible tissue damage. In fact, inflating near-term sheep fetal lungs with saline by approximately 90% increased pore radius only slightly (6 to 9 A ) , but slowed down secretion or caused reabsorption of lung liquid (Egan, Olver and Strang, 1975). These observations suggest that expansion most probably moves fluid by activating epithelial ion transport, and not by increasing non-electrolyte permeability. In fact Perks and Cassin (1985a) observed that K + ions move into lungs of fetal goats following lung expansion, suggesting that expansion may activate the basolateral N a 7 K + ATPase pump. Furthermore, intratracheal instillation of amiloride decreases post-natal lung water clearance in spontaneously breathing or artificially ventilated neonates (O'Brodovich, Rafii and Post, 1990, Song, Sun, Curstedt, Grossmann and Robertson, 1992). Amiloride also reduces perivascular cuffs, which are formed around large pulmonary blood vessels and bronchi during the rapid phase of neonatal liquid clearance (Song et ah, 1992). This suggests that epithelial N a + transport could be a significant component of normal lung clearance following the first few breaths. Could epithelial ion transport play an important role in lung fluid reabsorption following saline expansion? 7 Recently Garrad Nelson and Perks (1996b) made the interesting observation that when two lung preparations were incubated together and one was expanded, both preparations reduced fluid production or reabsorbed fluid. This suggested strongly that expanded lungs liberate an agent that can influence another preparation. This raised the question of the identity of the substance as well as its mechanism of action. Various substances including somatostatin ((Perks, Kwok , Mcintosh, Ruiz and Kindler, 1992) and prostaglandins (Berry, Edmonds and Wyll ie , 1969; Said, Kitamura and Vreim, 1972; Leffler, Hessler and Terragno, 1980; Leffler, Hessler and Green, 1984) may be released by expanded lungs. In this thesis, I attempt to identify the mechanisms involved in fluid reabsorption by expanded lungs, and how expanded lungs influence unexpanded ones. I also investigate the mechanisms by which catecholamines reduce fluid production and cause reabsorption in guinea pig lungs. Chapter II General Methods 9 2.1 Animals Pregnant albino guinea pigs of an inbred departmental stock were given food and water ad libitum (guinea pig chow, Ralston-Purina, supplemented with fresh vegetables and vitamin C). Studies were performed on 240 fetuses of 62 ± 2 days of gestation (term = 67 days) and average body weight 93.5 ± 22.6 g (SD). 2.2 Experimental Procedures The rate of lung liquid production was measured by an impermeant tracer technique, using Blue Dextran 2000 (Pharmacia, Dorval, Que; MW 2,000,000, Stokes radius, 270 A , radius of gyration, 380 A ) , based on the methods of Normand, Olver, Reynolds, and Strang (1971), Martins, Ramirez and Doyle (1975), and Liu and Chiou (1981). This method has been used previously, and its validity confirmed by simultaneous use with 125I-albumin (Cassin and Perks, 1982; Perks and Cassin, 1985a; 1985b; 1989; Perks, Dore, Dyer, Thorn, Marshall, Ruiz, Woods, Vanderhorst and Ziabakhsh, 1990) Pregnant guinea pigs were anaesthetized with halothane (Fluothane; Ayerst, Montreal, Que) until full inhibition of the corneal reflex; final euthanasia was accomplished by severing the carotid arteries. The fetuses were removed by Cesarean section and transferred to Krebs-10 Henseleit saline (see Burton, 1975) at 37°C. No fetal breathing movements were seen. The fetal thorax was opened by a midline incision to expose the lungs and trachea. The trachea was ligated rostrally and cannulated caudally with 1.5 - 2.0 cm of polyethylene tubing (PE 50, Intramedic, Clay Adams, Parsippany, NJ) filled with saline. The cannula was attached to a 1.0 ml tuberculin syringe via an 18-gauge hypodermic needle and a 3-way stopcock (K75, Pharmaseal, Puerto Rico). The cannula was tied in place just above the bifurcation of the bronchi, with double ligatures; this eliminated effects from the trachea itself. The trachea was then severed rostral to the cannula. The esophagus and vascular attachments to the lung were cut. During these procedures care was taken to keep the lungs warm and moist with frequent washes with Krebs-Henseleit saline at 37°C. The heart was removed, and the preparation was then rinsed with fresh saline, suspended in a 100 ml bath of Krebs-Henseleit saline at 37°C, oxygenated and maintained at pH 7.4 with 95% 0 2 and 5% C 0 2 . The preparations were set up within 3-4 minutes. Approximately 0.35 ml of lung liquid were withdrawn into the reservoir syringe, and a 10 pl sample was taken from the upper cup of the stopcock with a 1701 N C H gas-tight fixed volume syringe (Hamilton Co., Reno, NV); this was a blank for spectrophotometry. Subsequently 100 pl of Blue Dextran 2000 (50 mg.ml"1 in 0.9% NaCl) were added to the fluid remaining in the reservoir syringe, and the mixture was passed into the lungs. The preparations were left to equilibrate for 30 min, and in this period, lung fluid was withdrawn and returned every 5 min to 11 ensure an even distribution of dye throughout the lungs; the outer saline was replaced every 15 min. Experiments were carried out for 3 h after equilibration. During this time 10 u.1 aliquots of lung liquid were sampled every 10 min. The fluid was withdrawn and returned to the lungs midway between samplings, to ensure proper mixing within the lungs; mixing was also aided by the gentle but continuous stream of bubbles oxygenating the saline. Samples were placed in polyethylene micro test-tubes (250 u.1 Eppendorf C3515-7, Brinkmann Instruments (Canada) Ltd., Rexdale, ON), diluted 1:20 with distilled water, sealed, and vortexed (Vortex-Genie, Fisher Scientific, New York). Samples were centrifuged at 250 x g for 10 min (clinical centrifuge, Model CL, International Equipment Co., Needham Heights, MA). The concentrations of Blue Dextran in the supernatants were estimated by visible light spectrophotometer (Beckman DU-8 spectrophotometer, Beckman Instruments (Canada) Inc., Mississauga, ON, fitted with 250 u.1 quartz microcells, type 10972, NSG Precision Cells Inc., Farmington, NY) at 620 nm wavelength. The experiments followed an A B A design (saline/treatment/saline) in which samples obtained during the first 60 min after equilibration gave the basic rate of lung liquid production. In the treatment hour, lungs were either expanded or exposed to test substances placed in the bathing saline and/or the lung lumen; in the final hour, the lungs were again incubated in fresh saline. 12 To monitor the effects of lung expansion on lung liquid production, the first 4 samples obtained in the first hour of sampling were vortexed, centrifuged and their absorbances measured by visible light spectrophotometry (k = 620 nm). The slope of absorbance over the first 40 min was used to project the absorbance of lung liquid after 60 min. The lung volume at 60 min was estimated, and based on this, a volume of saline suitable for expansion was prepared. This saline was matched for the content of Blue Dextran in the sample immediately before expansion, estimated from the samples previously assayed. Intraluminal liquid was withdrawn into the reservoir syringe, and the expansion saline (approximately 70% total lung volume; based on earlier studies by Garrad Nelson and Perks, 1996a) was introduced in a second tuberculin syringe through the 3-way stopcock. The contents of the two syringes were mixed thoroughly before injection into the lungs. These procedures reduced mixing artifacts. 10 pl aliquots were sampled every 10 min as previously described. The second of the paired lungs was transferred into fresh saline at the beginning of every hour but were left unexpanded. 2.2 Quantification of results and statistical methods The rates of lung liquid production or reabsorption were calculated from changes in concentration of Blue Dextran, as described previously (Cassin and Perks, 1982; Perks and Cassin, 1989). The rates were estimated from plots of the total fluid volume against time, with readings recorded every 10 min; the total fluid volume was the sum of that within the lungs and that removed for spectrophotometry. Appropriate sequential adjustments were made every 10 13 min for the volumes removed through sampling. The rates of fluid production over 1 h intervals were calculated from the volume plots, using slopes of their regressions, fitted by the method of least squares (Steel and Torrie, 1970, by Apple II plus computer). In groups of similar experiments, comparisons of fluid production rates in successive hours were done by analysis of variance(ANOVA) and Scheffe's post-hoc test. Differences between groups were tested by two-way A N O V A . All statistical analyses were done by computer using SYSTAT for Windows (Version 5, SYSTAT, Inc., Evanston IL). Individual points on each volume slope were expressed as a percentage of the volume present in the lungs at the end of the first hour (assumed 100%). When plots from similar experiments were combined, these percentages were averaged. All mean values are given with their standard errors, unless otherwise stated. Statistical significance was accepted at P < 0.05. Chapter III Effects of lung expansion on lung liquid production, and the possible involvement of somatostatin 14 15 3.1 Introduction The removal of fluid from the lung lumen during birth, and soon after, is a critical event in the transition from fetal to extrauterine life. The spontaneous fall in fluid production late in gestation is thought to be the result of a rise in fetal plasma catecholamines (Walters and Olver, 1978, Brown, Olver, Ramsden, Strang and Walters, 1981; 1983). However, concentrations of epinephrine rise 100-fold only during labour (Brown et al, 1983) and fall rapidly in the postnatal period (Lagercrantz and Slotkin, 1986). Liquid clearance persists for hours after the concentration of epinephrine has decreased to low levels in the postnatal period (Jones, 1980). Brown et al. (1983) suggested that this could be the result of increased sensitivity of the pulmonary epithelium to catecholamines in late gestation, an increase which persists postnatally. Lungs of neonates drain fluid rapidly following the first breath. This rapid reabsorption increases the pulmonary interstitial space (Aherne and Dawkins, 1964) and lymph flow (Humphreys, Normand, Reynolds and Strang, 1967; Bland, Hansen, Haberken, Bressack, Hazinski, Raj, and Goldberg 1982). As with the first breath, expansion of fetal or adult lungs with saline results in liquid reabsorption (Perks and Cassin, 1985; Ramsden, Markiewicz, Walters, Gabella, Parker, Barker and Neil, 1992; Vejlstrup, Boyd and Dorrington, 1993; Garrad Nelson and Perks, 1996a; 1996b). Although a tempting proposition, there is no evidence that expansion of lungs with fluid that is comparable in volume to the first breath, increases non-electrolyte permeability of the lung epithelium to 16 cause loss of lung liquid, or that this fluid loss is a result of tissue damage due to inflation. Egan, Olver and Strang (1975) determined that the critical inflation pressure at which the pulmonary epithelium becomes very leaky to solutes was 35 cm of H 2 0. This represents a level of expansion several fold greater than those achieved by the first breath and those simulated in all the other studies. In fact 70% expansion of fetal guinea pig lungs in vitro only caused a transient increase in intratracheal pressure of 0.5-1.0 cm H 2 0 (Garrad Nelson and Perks, 1996a); overinflating near-term fetal lamb lungs with saline approximating 90% only caused a peak inflation pressure of 7.7 cm H 2 0 and a slight (6 to 9 A) increase in pore radius (Egan et al., 1975). The intratracheal instillation of amiloride before the first breath decreases postnatal lung water clearance in spontaneously breathing or artificially ventilated neonates (O'Brodovich, Hannam, Seear and Mullen, 1990; Song, Sun, Curstedt, Grossmann and Robertson, 1992), implying a more subtle mechanism of fluid reabsorption than a simple expansion of aqueous pores. Recently, Garrad Nelson and Perks (1996b) made the interesting observation that when lung preparations isolated from fetal guinea pigs were incubated in pairs, the expansion of one preparation not only caused fluid reabsorption in the expanded preparation, but arrested fluid production in the companion preparation; amiloride reduced fluid reabsorption by expanded lung preparations, but was not tested on the unexpanded companion lungs. These observations suggested that expansion generates some unknown factor that can influence lung liquid production. Thus the evidence points to the activation of a mechanism involving at least in part, 17 stimulated epithelial ion transport as opposed to bulk flow of fluid caused by a widening of the pores. Somatostatin, a gastric peptide, is liberated by perinatal guinea pig lungs following the first breath; this release is maintained for at least 30 min after the onset of breathing (Perks, Kwok, Mcintosh, Ruiz and Kindler, 1992). Like catecholamines, levels of immunoreactive somatostatin in the human feto-placental circulation are known to rise at delivery (Saito, Saito, Sano, Hosoi and Saito, 1983, Marchini, Lagercrantz and Uvnas-Moberg, 1990). In fetal sheep, cells immunoreactive to somatostatin were identified in airway epithelium, but this immunoreactivity was not evident in airways of 3 month-old lambs (Balaguer, Romano and Ruiz-Pesini, 1992). These observations suggest that this peptide (1) could have a role in the functional maturation of the newborn, and (2) has no role in mature air-breathing lungs. As a Na+-K+-2C1" cotransport inhibitor (Foskett and Hubbard, 1981, Foskett, Bern, Machen and Conner, 1983), somatostatin may be expected to reduce lung liquid production by inhibiting the basolateral Na+-K+-2C1" cotransporter which is believed to be involved in fetal lung liquid secretion (Cassin, Gause and Perks, 1986; Thorn and Perks, 1990). Perks et al. (1992) in fact demonstrated that somatostatin could reduce fetal lung liquid production by in vitro lungs from fetal guinea pigs. In these experiments I confirm that isolated lungs reduce liquid production or reabsorb fluid following expansion with saline equivalent in volume to the first breath. I also confirm the report 18 by Garrad Nelson and Perks (1996b) that expanded lungs release a factor that can inhibit liquid production in a second preparation, and extend these studies to test whether the agent liberated by expanded lungs into the bathing medium could be somatostatin. 3.2 Materials and Methods 3.2 1 Animals Pregnant albino guinea pigs of an inbred departmental stock were given food and water ad libitum (guinea pig chow, Ralston-Purina, supplemented with fresh vegetables and vitamin C). Studies were performed on 69 fetuses of 62 ± 2 days of gestation (term = 67 days) and 90.6 ± 24.4 g (SD) body weight. 3.2.2 Experimental Procedures I Studies of lung liquid production Fetal lungs were prepared as previously described and incubated in vitro either singly, or in pairs. Untreated controls were monitored through 3 successive hours. The remaining lungs were used in studies to monitor the effects of lung expansion on lung liquid production. In each of the single preparations, and in one of the paired lungs, the first 4 samples obtained in the first hour of 19 sampling were vortexed, centrifuged and their absorbances measured as previously described. The slope of absorbance over the first 40 min was used to project the absorbance of lung liquid after 60 min. Based on this, a volume of saline suitable for expansion was prepared. This saline was matched for the content of Blue Dextran in the sample immediately before expansion, estimated from the samples previously assayed. Intraluminal liquid was withdrawn into the reservoir syringe and the expansion saline, warmed to 37°C, was introduced in a second tuberculin syringe through the 3-way stopcock. The contents of the two syringes were mixed thoroughly before injection into the lungs. These procedures reduced mixing artifacts. Mixing was done after 5 min and midway between samples; 10 u.1 aliquots were sampled as before through the test and recovery hours. The second of the paired lung preparations was transferred into fresh saline at the beginning of every hour but were left unexpanded. The experiments were divided into two groups. (1) Single incubations: 12 lung preparations, placed in two subgroups; 6 preparations from fetuses over 100 g (large), 6 preparations from fetuses under 100 g (small) were incubated for three successive hours without treatment (controls); 12 similarly grouped lung preparations were expanded with Krebs-Henseleit saline at the end of the first hour. (2) Joint incubations: 12 lung preparations were incubated in pairs, but without treatment (joint controls); 12 others were incubated in pairs, and one preparation from each pair was expanded with Krebs-Henseleit saline at the end of the first hour. All expansions were close to 70% (71.1 ± 3.3% (SD)). 20 II Estimation of somatostatin-like immunoreactivity Fetal lungs were prepared and individually incubated in vitro as previously described. Incubation was done in 50 ml capacity plastic cups to reduce the dilution of any unknown substance released by the lungs. The saline contained 0.2% radioimmunoassay (RIA) grade bovine serum albumin (BSA) and 10% protease inhibitor (aprotinin) from bovine lungs (Sigma Chemicals, St. Louis, MO), to prevent loss of the unknown to adsorption by the cup. Immediately after transferring the lung into the second hour saline, the first hour saline was transferred into a 100 ml capacity plastic bottle (Nalge Nunc International, Rochester, NY), tightly capped and placed in an ice bucket until extracted for lyophilization. At the beginning of the second hour, lung preparations were expanded by approximately 70% (71.4 ± 0.5%) as previously described. The lungs were incubated for 20 min after expansion before the experiment was stopped. The bathing saline was transferred into a 100 ml plastic bottle, capped and placed in an ice bucket until extracted for lyophilization. After each experiment the lung wet weight was taken. a) Sample extraction Samples were applied to Waters Sep-Pak C , g cartridges (Millipore, Mississauga, ON), previously washed with 10 ml acetonitrile containing 0.1% triflouroacetic acid (TFA), followed by 20 ml distilled water, and 10 ml of 1% BSA. After applying the sample, the cartridge was washed with 21 10 ml distilled water, followed by 1 ml 10% acetonitrile with 0.1 % TFA (Pierce, Rockford II). Somatostatin was eluted from the Sep-Pak with 2 ml (1 ml x 2) 65% acetonitrile with 0.1% TFA. The eluate was lyophilized and stored at -20°C until reconstituted for RIA. b) RIA for somatostatin Somatostatin-like immunoreactivity (SLI) was measured by a specific RIA using monoclonal antibody SOMA 03 (Mcintosh, Arnold, Bothe, Becker, Koberling and Creutzfeldt, 1978; Mcintosh, Kwok, Tang and Brown, 1987). A final antibody titre of 1:4,000,000 was used for RIA. The samples were reconstituted with 100 pl distilled water and 400 pl assay buffer. The assay buffer consisted of 23.77 mM barbital buffer (pH 7.4) containing sodium acetate (3.9 mM), sodium chloride (43.60 mM), Ethylmercurithiosylicyclic acid sodium salt (merthiolate) (0.247 mM), aprotinin (500 KlU/ml; Trasylol®, Miles Scientific) and BSA (0.5 g/100 ml, Pentex®, Miles Labs). The 125I-somatostatin label was purified on the day of the assay. It was dissolved in 0.002 N ammonium acetate buffer (pH 4.6) and applied to a column of CM-cellulose (CM 52) previously equilibrated with the same buffer. The column was then washed with 50 ml of 0.002 M ammonium acetate buffer and the label eluted with 0.2 M ammonium acetate (pH 4.6). The peak fraction was neutralized with 2 N NaOH and further diluted with assay buffer to 3,500 ± 300 cpm/100 pl for RIA. 100 pl aliquots of reconstituted samples or standards (synthetic SS-14, Peninsula Laboratories) were mixed with 100 pl antibody, 100 pl ,25I-somatostatin label (approximately 0.5 fmol), and 100 pl assay buffer. The mixture was vortexed and incubated in 22 glass tubes for 36 hours at 4°C. Incubations were done in duplicates for unknowns and in triplicates for standards. Separation of bound and free antigen was done with dextran-coated charcoal. After decantation of the supernatant, both bound and free fractions were counted in a gamma counter. Results are reported as pg/ml (± SEM). 3.2.3 Statistical methods The differences in fluid production rates in successive hours were tested by two-way analysis of variance (ANOVA) followed by Scheffe's post hoc test; and by regression analysis. Difference between effects of different treatments were tested by one-way A N O V A . Any changes in somatostatin-like immunoreactivity in the saline bathing lungs after expansion were assessed by one-way A N O V A . Statistical significance was accepted at P < 0.05. 23 3.3 Results 3.3.1 L u n g l iqu id product ion I Single incubations Untreated control lung preparations from small fetuses of 61 ± 1 days of gestation and 74.8 ± 10.4 g (SD) body weight, produced fluid for three successive hours, and showed no significant change in rate. The fluid production rates were: 1.37 ± 0.20, 1.45 ± 0.16, and 1.53 + 0.26 ml/kg body weight per h (Fig la). Similarly, untreated lungs from large fetuses of 63 ± 1 days gestation and 112.4 ± 13.5 g (SD) body weight produced fluid for three hours without any significant change. The fluid production rates were: 1.03 ± 0.11, 1.18 ± 0.15, and 1.36 ± 0.21 ml/kg body weight per h (Fig lb). Six small lungs, obtained from fetuses below 100 g body weight (83.5 ± 9.0 g, SD), and 61 ± 1 days of gestation, were expanded with saline by about 74% (73.6 ± 3.3% SD). All lungs decreased fluid production significantly in the middle hour. The overall effect was a 53.5 ± 6.6%) fall (P < 0.001, ANOVA). Average rates in successive hours of: before expansion, 1.13 ± 0.08; after expansion, 0.53 ± 0.09 and 0.55 ± 0 . 1 3 ml/kg body weight per h (Fig lc). Six large lungs, obtained from fetuses over 100 g body weight (124.8 ± 12.1 g, SD), and 66 ± 1 days of gestation, were similarly expanded (70.7 ± 2.8 % SD). All lungs reabsorbed fluid in the hours following expansion; fluid production fell by 170.1 ± 21.1% in the middle hour. Average fluid 24 production rates in successive hours were: before expansion, 1.81 ± 0.17; after expansion, -1.27 ± 0.45 and -0.28 ± 0.30 ml/kg body weight per h (Fig Id). These decreases were significant by A N O V A (P< 0.001). The reductions in fluid production and reabsorption in the hour after expansion were quantified as percentage falls from the original rates. Data obtained from these studies were combined with additional data from expanded lungs from later studies of joint lung preparations (see below). The combined average weight of the fetuses was 99.3 ± 20.6 g (SD, range 63.4 to 145.8 g), and the average gestational age was 62 ± 3 days (SD, range 59 to 67 days). The average expansion was 70.6 ± 3.0% (SD). The percent fall in fluid production by lung preparations from fetuses below lOO.g (small) was compared with the percent fall by lung preparations from fetuses above 100 g (large). As shown in Fig 2, large preparations had significantly greater reductions in fluid production than small preparations (P < 0.0001, ANOVA). It was concluded that the fall in fluid production which is associated with lung expansion increases with fetal maturation. II Joint incubations These studies were based on 24 fetuses of 62 ± 2 days of gestation and 87.8 ± 24.4 g (SD) body weight. Preparations were incubated in pairs, each pair taken from fetuses of the same litter. Twelve control preparations were incubated in pairs and received no treatment. Their fluid production rates in successive hours were: first preparation, 1.31 ± 0 . 1 2 , 1.30 ± 0.13 and 1.29 ± 25 0.25; second preparation, 1.51 ± 0.26, 1.22 + 0.21 and 1.29 ± 0.22 ml/kg body weight per h. No significant changes in fluid production were seen (Fig 3 Ai; 3 Aii). Twelve further preparations were incubated in pairs, but one preparation in each pair was expanded by 70% (68.9 ± 2.8%) at the end of the first hour. All expanded lungs decreased fluid production; three turned to reabsorption. The fluid production rates in successive hours were: before expansion, 1.08 ± 0.24, after expansion, -0.28 ± 0.38 and -0.41 ± 0.23 ml/kg body weight per h (Fig 3 Bi). The fall immediately after expansion (125.4 ± 42.6%) was significant by A N O V A (P < 0.01). At the same time, all six unexpanded companion lungs reduced fluid production; three turned to slight reabsorption. The overall response was a 88.2 ± 10.7%> fall in fluid production in the middle hour. The fluid production rates in successive hours were: 0.98 ± 0.41, 0.12 ± 0.09 and 0.29 ± 0.20 ml/kg body weight per h (Fig 3 Bii). The reduction in fluid production in the middle hour was significant by A N O V A (P < 0.01). The results suggest that expansion liberates an agent that can inhibit fluid production in another lung preparation. 3.3.2 Somatostatin-like immunoreactivity Somatostatin-like immunoreactivity (SLI) was present in bathing saline obtained from all 20 preparations. There was no relationship found between the level of SLI in the bathing saline and either fetal weight or lung wet weight. SLI in samples obtained at the end of the hour immediately before lung expansion was 18.8 + 3.1 pg/ml of eluate. In samples obtained 20 min 26 after expansion of the same lungs, SLI was 13.5 ± 2.3 pg/ml of eluate. There was no significant change in SLI due to expansion (Fig 4). 27 Fig 1: Effect of expansion with Krebs-Henseleit saline, on lung liquid production by in vitro lungs from near-term fetal guinea pigs. Based on 24 fetuses of 62 ± 2 (SD) days of gestation and 98.9 ± 23.4 (SD) g body weight, (a) Six untreated control lungs, all from fetuses averaging 61 + 1 days gestation and 74.8 ± 10.4 g (SD) body weight (small); (b) untreated control lungs from fetuses of 63 ± 1 days gestation and 112.4 ± 13.5 g body weight (large); (c) preparations of lungs from small fetuses (61 + 1 days of gestation and 83.6 ± 9.0 g) expanded 73.6 ± 1.4% with saline at the end of the first hour; (d) preparations of lungs from large fetuses (65 ± 1 days gestation and 124.8 ± 12.1 g) expanded 70.7 ± 1.1% in the same way. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100%> was (a) 0.75 ± 0.15 (SD), (b) 1.07 ± 0.21, (c) 0.74 ± 0.14, and (d) 1.43 ± 0.17 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 28 130 120 110 100 90 80 130 120 110 100 90 80 120 110 100 90 80 120 110 100 90 80 a) Control (Small) n = 6 b) Control (Large) n = 6 c) 74 % Expansion (Small) n = 6 d) 71% Expansion (Large) n = 6 1.36 ± 0 . 2 1 0.28 ± 0 . 3 0 0 SALINE TREATMENT SALINE Time (Hours) 29 Fig 2: The effect of fetal weight on responses of in vitro lungs from fetal guinea pigs to saline expansion. Based on 39 fetuses of 62 ± 3 days of gestation and 99.3 ± 20.6 g (SD) body weight. The lungs produced fluid at 1.27 ± 0.10 ml/kg body weight per h in the initial hour and were expanded 70.6% ± 0.5% at the end of that hour. Ordinate: responses quantified as the percent reduction in liquid production from the resting (first hour) value following saline expansion (second hour). Error bars denote standard errors. The asterisk denotes a significant increase in the response to saline expansion (P < 0.001, ANOVA). Abscissa: Fetal weight (g); lungs were divided into two groups: those from fetuses below 100 g (61 ± 1 days gestation and 84.7 ± 1 0 . 1 g; small) and from fetuses above 100 g (65 ± 2 days gestation and 120.3 ± 11.2 g; large). 30 200 180 160 140 120 100 80 60 40 20 0 * (16) Small (< 100 g) Large (> 100 g) Fetal Size 31 Fig 3: Influence of expanded lungs on lung liquid production by unexpanded companion lungs supported in the same saline in vitro. Based on 24 fetuses of 62 ± 2 days of gestation and 87.8 ± 24.4 g (SD) body weight. Untreated control lungs taken from the same mother and incubated in pairs showed no significant changes in liquid production through three hours (Ai, Aii); 12 experimental lungs taken from the same mother and incubated in pairs, but with one expanded 68.9 ± 1.0% with saline at the end of the first hour (Bi), and the companion lung left unexpanded (Bii). The expanded lungs reabsorbed liquid; the companion lungs also reduced production. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (Ai) 0.68 ± 0.08 (SD), (Aii) 0.78 ± 0.23, (Bi) 1.09 ± 0.30, and (Bii) 1.00 ± 0.20 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 32 I Ai) lung 1: Control 130 120 110 100 90 80 120 110 100 90 h 80 110 100 90 80 L 110 r 100 -90 -Aii) lung 2: Control Bi) lung 1: 70% Expansion n = 6 . - ' ' Bii) lung 2: Unexpanded n = 6 . - ' * ' * 0.12 + 0.09 1.29 + 0.25 Joint incubations 1.29 ± 0 . 2 2 Joint incubations 0.29 ± 0 . 2 0 0 SALINE 1 TREATMENT 2 SALINE 3 Time (Hours) 33 Fig 4: Levels of somatostatin-like immunoreactivity (SLI) in the outer saline before and after expansion of in vitro lungs from fetal guinea pigs. Based on 20 fetuses of 62 ± 2 days of gestation (term = 67 days) and 101.6 ± 18.8 g (SD) body weight. After one hour of incubation in 100 ml of Krebs-Henseleit saline, lung preparations were transferred into 50 ml of fresh saline and expanded by 70% (71.4 ± 0.5%). SLI was estimated by radioimmunoassay using a monoclonal antibody. Values are mean ± SEM. Somatostatin-like immunoreactivity Before After Expansion Expansion 35 3.5 Discussion The results reported here confirm earlier observations that expansion of fetal lungs in vitro can reduce lung liquid production or turn it to reabsorption (Perks and Cassin, 1985a; Garrad Nelson and Perks 1996a) and liberate an agent that reduces liquid production in a second preparation (Garrad Nelson and Perks 1996b). The results also suggest that the agent liberated by expanded fetal lung preparations into the bathing saline is not somatostatin. In the first set of experiments, fetal lungs responded to intraluminal liquid expansion with a fall in production or with a reabsorption of lung liquid. Expansions were carried out with saline because air expansion causes the production of a foam which prevents further sampling of fluid. Expansion consistently produced reductions in fluid production or reabsorptions when lungs were expanded by approximately 70%, a level of expansion similar to the volume of the first breath (Garrad Nelson and Perks, 1996a). In that study, intraluminal pressures generated at 70% expansion were very small (0.5-1.0 cm H 2 0) and transient, and no discernable tissue damage occurred. In addition, they fully discussed the validity of the method used and assumptions made. In earlier studies on near-term fetal lambs, inflating lungs with saline by approximately 90% (peak inflation pressure = 7.7 cm H 2 0) only caused a slight increase in pore radius but slowed down secretion or caused reabsorption of lung liquid (Egan et ai, 1975). Thus the change observed in lung liquid volume may not involve non-electrolyte permeability or only do so to a slight extent. Furthermore, our lung preparations did not have an intact vascular perfusion; therefore, movement of fluid from lumen to interstitium could not be due to a colloid osmotic pressure. 36 The fall in fluid production associated with lung inflation is not surprising. Liquid expansion reduced lung liquid production by fetal goat lungs (Perks and Cassin, 1985). Other studies on perfused isolated lungs from fetal lambs (Ramsden et al, 1993) and on adult rabbit lungs (Vejlstrup et al, 1992) agree that inflation of lungs inhibits lung liquid production. Most recently the extensive study on in vitro lungs from fetal guinea pigs by Garrad Nelson and Perks (1996a) confirmed this observation. It is important to note that inflation of perfused or non-perfused isolated lungs produced the same responses as inflation of lungs in vivo. This suggests that the lungs themselves are capable of the response without neural influences or humoral influences from other organs. It is also evident from the in vitro studies that a patent pulmonary circulation is not required for the responses to occur. This suggests that the mechanism is driven primarily by processes within the pulmonary epithelial cells. When the fetuses were divided into two groups (below 100 g and above 100 g), there was a clear distinction in the relative change in fluid production in the hour after expansion. The responses of fetal lungs to intraluminal liquid expansion were clearly greater in lungs from larger fetuses. If expansion caused increased epithelial leakiness or epithelial damage, the results would imply that mature fetal lungs are leakier than immature ones. This is highly unlikely. Normand, Reynolds and Strang (1970) observed that when immature fetal lungs were ventilated, macromolecular transfer between alveoli and lymph was greater and the degree of molecular sieving was less than when mature fetal lungs were ventilated. In a later study, Goodman and 37 Wangensteen (1982) further investigated the permeability of the alveolar epithelium to small solutes at different stages of development and reported that immature fetal lungs had greater absolute permeability values than mature ones. The epithelium becomes less permeable to non-electrolytes with gestational maturation; therefore, it is clear that the increasing response to expansion would not be based on non-electrolyte permeability. Humphreys et al. (1967) had in fact suggested that increased lymph flow at the start of ventilation accounted for the removal of about 40% of the liquid present in the lungs of mature fetuses compared to only 25% of fluid in lungs of immature fetuses. Although our preparations had no functional lymph flow, my results fit with their observations since the percent fall in liquid production following expansion was far greater in lungs from larger (more mature) fetuses. Expansion must activate a mechanism of transepithelial fluid transfer that becomes more efficient with gestational maturation. This suggestion supports the conclusion of Garrad Nelson and Perks (1996a) that the reabsorptive response could be a specialization of the perinatal lung. This ability of the lung to transfer fluid is apparently reduced in the days after birth (Ramsden et al, 1992). When lung preparations were incubated in pairs, the expanded lungs responded with a large reduction in fluid production or with a reabsorption of fluid. Interestingly, the unexpanded companion lungs also responded with a substantial fall in liquid production. This response was a true one, as pairs of unexpanded lungs produced fluid steadily for three successive hours when incubated in the same saline. These observations suggest the release of an active, diffusible substance by the expanded lungs. The same observations were made in similar studies by Garrad 38 Nelson and Perks (1996b). In their study, the possibility that inhibition of the companion lungs was due to a toxic substance released by the expanded lung was examined and discounted This raised the question that if the active substance is not the result of pathological processes, does it then represent a physiological factor? Expansion with saline could be different from expansion with air; it does not produce pulmonary vasodilation unless the saline is well oxygenated (Dawes, Mott, Widdicombe and Wyatt, 1953). It is also known that intraluminal pressures reached by liquid expansion are far less than those reached by gas of an equivalent volume (Egan et al, 1975). Therefore, if the response is modulated by changes in pressure, my observations would be accurate qualitatively, but represent a relatively small part of the physiological response to be expected following expansion with air. The apparent age-dependent response to lung inflation supports the suggestion by Garrad Nelson and Perks (1996b) that it represents a physiological process. The responses of expanded lungs increased with fetal weight, suggesting that the mechanism involved could be dependent on gestational maturation. We would, therefore, expect the response of fetal lungs to be greatest at birth. Shortly after birth, lung expansion occurs inevitably as a consequence of the first breath. Estimation of somatostatin-like immunoreactivity by radioimmunoassay suggested that expansion did not increase liberation of somatostatin into the bathing saline. This means that somatostatin was not involved in the reduction of fluid by the unexpanded companion lung 39 preparations. The concentration of somatostatin in the bathing saline before expansion was 18.3 pg/ml (1.1 x 10"" M); this concentration did not change after expansion. According to Perks et al., 1992, the concentration of exogenous somatostatin that could reduce lung liquid production was at least 10"6 M ; at 10"7 M , somatostatin (SS-14) did not affect production significantly although the theoretical threshold was 9.0 x 10'" M . Thus the concentration I measured in the outer saline could not have inhibited liquid production. In addition, exogenous somatostatin could not turn fluid production into reabsorption even at concentrations as high as 10"5 M (Perks et al., 1992); the peak concentration of immunoreactive somatostatin-like substances measured in wet lung tissue was about 7000 pg/g (4.3 x 10"9 M). This concentration could theoretically reduce lung liquid production if carried from neuroepithelial bodies directly to alveoli, but it is unlikely that it could cause fluid reabsorption. Therefore, responses of expanded lung preparations can not be attributed to tissue release of somatostatin. These preliminary experiments did not investigate the possible ion transport mechanisms involved in the inhibition of fluid production by the companion lungs. Hence we still do not know the identity or mechanism of action of the released agent. The rest of this thesis investigates these questions. Chapter IV Effects of catecholamines on lung liquid production 41 4.1 Background The first suggestion of possible control of reabsorption of lung liquid in the perinatal period came from studies of adrenergic agonists. The p-adrenergic agonist, isoxsuprine, was implicated in the improvement of air retention in newborns (Wysogrodoski, Taeusch and Avery, 1974) and in the reduction of fetal lung liquid volume and wet weight (Enhorning, Chamberlain, Contreras, Burgoyne and Robertson, 1977). Similar effects were later observed in studies with other P-adrenergic agonists, terbutaline, prenaterol and ritodrine (Bergman, Hedner and Lundborg, 1978, 1980). The involvement of p-adrenergic agonists in the control of lung liquid in chronically-instrumented fetal sheep was confirmed in studies with terbutaline and isoproterenol (Walters and Olver, 1978; Brown, Olver, Ramsden, Strang and Walters, 1981; 1983; Chapman, Carlton, Cummings, Poulain and Bland 1991; Ramsden, Markiewicz, Walters, Gabella, Parker, Barker and Neil, 1992). In addition, the magnitude of responses as well as the sensitivity of fetal lungs to catecholamines increased towards delivery (Brown et al, 1983). Recent studies in fetal guinea pig lungs using specific a-adrenergic antagonists have shown that ot-adrenoreceptors are also involved in the fluid reabsorption (Woods, Doe and Perks, 1997; Doe and Perks, 1998). Catecholamines are present in the fetal circulation, and norepinephrine is the predominant fetal catecholamine (Comline and Silver, 1961; Lewis, Evans and Sischo, 1982; Cohen, Piasecki and Jackson, 1982; Bistoletti Nylund, Lagercrantz, Hjemdahl and Strom, 1983; Brown etal, 1983; Jelinek and Jensen, 1991). The catecholamine levels in fetal blood rise dramatically around 42 delivery (Jones, 1980; Olver, 1981; Lewis et al. 1982; Cohen et al, 1982;). Fetal lung tissue contains measurable levels of norepinephrine (Gardey-Levassort, Richard, de Lauture, Thiroux and Olive, 1981), and the concentrations are high in relation to other extra-adrenal tissues (Jelinek and Jensen, 1991). In addition, adrenergic innervation has been identified in lungs (O'Donnell, Saar and Wood, 1978). Furthermore, phenylethanolamine N-methyl transferase (PNMT) activity is high in fetal lungs (Padbury, Lam, Hobel and Fisher, 1983). Studies in our laboratory have recently shown that expanded fetal lungs release an agent that can inhibit lung liquid production in a second lung preparation in vitro (Garrad Nelson and Perks, 1996b). In the present study, I examine whether: (1) catecholamines have any internal effects on the lungs and (2) catecholamines are released by expanded fetal lung preparations in vitro. 4.2 Materials and Methods 4.2.1 Animals Pregnant albino guinea pigs of an inbred departmental stock were given food and water ad libitum (guinea pig chow, Ralston-Purina, supplemented with fresh vegetables and vitamin C). Studies were performed on 108 fetuses of 63 ± 2 days of gestation (term, 67 days) and 102.0 ± 19.6 g(SD) body weight. 43 4.2.2 Studies of lung liquid production To demonstrate the effects of catecholamines on liquid production by fetal lungs in vitro, studies were done as previously described. Epinephrine was tested at 2.2 x 10"6 M and norepinephrine at 10'5 M . In order to test the hypothesis that catecholamines are released by expanded lungs, six lung preparations were incubated singly in the presence of: (1) 10"5 M phentolamine (a-adrenergic antagonist) for 10 min before expansion, and during the hour following expansion, and (2) 10"7 M propranolol (P-adrenergic antagonist) used similarly. Other preparations were incubated in pairs; one preparation was expanded and the second was left unexpanded. The aim of these experiments was to test the possibility that catecholamines, if released from expanded lungs, affected fluid production in the unexpanded companion lung preparations. Twelve lung preparations were incubated in pairs in saline containing adrenergic blockers as above; one preparation from each pair was expanded by approximately 70% (69.7 ± 0.5%). To estimate the concentrations of catecholamines released by expanded fetal lungs, lungs were obtained from 24 fetal guinea pigs of 61 ± 1 days of gestation and 87.7 ± 17.1 g (SD) body weight. Twelve of the fetal lungs were incubated in vitro for three hours without treatment. At 5, 10, 20, 30, 40, and 50 min of the middle hour, 10 pl samples were withdrawn from the lungs for estimation of catecholamine concentration. The remaining 12 lungs were expanded by approximately 70% (71.0 ± 0.9%) 10 minutes into the middle hour; 10 pl aliquots were 44 withdrawn from the lungs, and 200 u.1 aliquots sampled from the bathing saline, at similar intervals, for estimation of catecholamine concentration. An extra sample was taken from the bathing saline two min after expansion. The samples were placed in 250 or 1000 ul polyethylene micro test tubes (Eppendorf, Brinkmann Instruments (Canada) Ltd., Rexdale, Ont), and the test tubes immediately placed in liquid nitrogen. The samples were then kept in a freezer at -70° until extracted for high pressure liquid chromatography (HPLC). The samples were kept frozen for no more than 14 days before extraction. 4.2.3 Analytical procedures: Sample extraction and analysis by HPLC The catecholamine standards used were HPLC grade norepinephrine (arterenol) bitartrate, epinephrine bitartrate, and dopamine hydrochloride (Sigma Chemicals, St. Louis, MO). 200 u.1 aliquots of these standards or experimental samples were mixed with 10 mg acid-washed aluminium oxide (alumina), 400 u.1 of 2 M tris/EDTA buffer (pH 8.6) and 50 u.1 of the internal standard, 20 ng/ml 3,4 dihydroxy benzylamine hydrobromide (DHBA). The mixture was shaken for 15 min, centrifuged for 1 min, and the supernatant aspirated and discarded. The mixture was washed three times with 0.2 % tris/EDTA solution (pH 8.1). Subsequently, catecholamines were eluted with 125 u.1 of 1% glacial acetic acid, 0.05% sodium disulphite and 0.025% EDTA in double distilled water; 20 u.1 aliquots of the eluate were injected into the HPLC column for analysis. Recovery of catecholamines in the extraction procedure was 69 ± 3%. 45 The concentrations of epinephrine, norepinephrine, and dopamine were determined by high pressure liquid chromatography (HPLC). The HPLC incorporates a Waters 460 Electrochemical Detector using a glassy carbon electrode (applied potential = + 0.60 V), a reverse-phase Waters Plasma Catecholamine Column, a Waters Model 510 HPLC pump with pulse dampeners and a Waters U6K Universal Liquid Chromatograph Injector (Waters Chromatography Division, Millipore Ltd., Mississauga, Ont). The concentrations were internally calculated from the area ratios of catecholamine/internal standard peaks by an integrator (Waters 746 Data Module), connected on-line to the electrochemical detector, and given directly in pg/ml. All concentrations were corrected for loss due to extraction. 46 4.3 Results 4.3.1 Effects of epinephrine and norepinephrine on lung liquid production Initial studies to demonstrate the effects of catecholamines on fetal lung liquid production were made on in vitro lungs from 18 fetal guinea pigs of 61 ± 1 days of gestation and 85.8 ± 19.1 g (SD) body weight. The preparations were divided into three groups; six untreated, six treated with 2.2 x 10"6 M epinephrine, and six treated with 10"5 M norepinephrine in the outer saline. Concentrations of norepinephrine were based on studies by Woods and Perks (unpublished observations); concentrations of epinephrine were based on studies by Woods et al., (1997) and on preliminary studies at 10"7 M , 10"6 M , and 2.2 x 10"6 M . At 10"7 M , lung liquid production did not change significantly. At 2.2 x 10"6 M , liquid production dropped consistently. Based on these trials, subsequent studies with epinephrine were done at 2.2 x 10"6 M . Six untreated control preparations produced fluid at 1.20 ± 0.13 ml/kg body weight per h in the first hour and showed no significant change through three successive hours (Fig 5a). All six lungs treated with 2.2 x 10"6 M epinephrine reduced fluid production; four turned to reabsorption. The fluid production rates were: before treatment, 1.33 ± 0.28; during treatment:-0.62 + 0.51; after treatment: 0.62 ± 0.44 ml/kg body weight per h (Fig 5b). The overall fall in fluid production in the second hour was 146.2 ± 23.7% and was judged significant by A N O V A and by Scheffe's test (P < 0.05). All six lungs treated with 10"5 M norepinephrine reduced fluid 47 production, with four turning to reabsorption. The fluid production rates were: before treatment: 1.24 ± 0.24; during treatment: -0.46 ± 0.48; after treatment: 0.32 ± 0.27 ml/kg body weight per h (Fig 5c). The fall in the second hour was 137.3 ± 21.7% and was significant (P < 0.05, ANOVA). These results suggest that both epinephrine and norepinephrine could reduce fluid production or cause reabsorption. These effects were reversible; fluid production increased when treatment was stopped. These results are similar to those obtained by other workers in the laboratory; the data were combined to give a linear log dose-response curve (r = 0.97, n = 54) with a theoretical threshold at 4.0 x 10"'° M (Fig 6). 4.3.2 Is there internal action of catecholamines in response to expansion? Studies with adrenergic blockers I Single incubations Untreated preparations produced fluid steadily through three successive hours (Fig 7a). Expanded lung preparations decreased fluid production, in the middle hour, by 170.7 ± 2 1 . 1 % . Fluid production rates in successive hours were: before expansion; 1.81 ± 0 . 1 7 , after expansion: -1.27 ± 0.45 and -0.28 ± 0.30 ml/kg body weight per h (Fig 7b). In order to test the possibility that if released by expanded lungs norepinephrine affected lung liquid production through cc-adrenoreceptor stimulation, 10"5 M phentolamine was placed in the outer saline 10 min before 70% (70.2 ± 0.3%) lung expansion, and maintained during the following hour. All expanded 48 lungs reduced fluid production; two turned to reabsorption. Fluid production rates in successive hours were: before expansion; 1.56 ± 0.28, after expansion, in the presence of phentolamine: 0.19 ± 0.19; in the final hour: 0.36 ± 0.24 ml/kg body weight per h (Fig 7c). The fall in fluid production in the middle hour was 87.7 ± 17.9% and was significant by A N O V A (P < 0.01). This fall was not significantly different from that of expanded lung preparations incubated similarly, but with no phentolamine in the outer saline (Fig 7b). Preparations incubated in the presence of 10"5 M phentolamine alone produced fluid in the hour of exposure and for an additional hour in saline with no significant change (Fig 7d). Therefore phentolamine did not abolish responses, and although some reduction in responses seemed to have occurred, there was no significant difference between the fall in fluid production in the presence and in the absence of phentolamine (ANOVA). In six lung preparations incubated in the presence of 10"7 M propranolol (P-adrenergic antagonist), and expanded by 70% (69.6 ± 0.8%), all lungs reduced fluid production with two turning to reabsorption. The fluid production rates in successive hours were: before expansion: 1.20 ± 0.11; after expansion, in the presence of propranolol: -0.02 ± 0.36; in the final hour: 0.45 ± 0.26 ml/kg body weight per h (Fig 7e). The overall fall in fluid production in the middle hour was 101.3 ± 30.9% (P < 0.05, ANOVA). This fall in fluid production was not significantly different from that of expanded lungs incubated in the bathing saline without propranolol (170.7 ± 2 1 . 1 % , Fig 7b). Six preparations incubated in the presence of 10"7 M propranolol alone produced fluid with no significant change in the middle and final hours (Fig If). It was concluded that despite some apparent reductions in the effects of expansion in the 49 presence of both blockers, responses persisted, and direct statistical comparisons showed that neither antagonist had any significant effect on the responses to expansion. II Joint incubations Lung preparations were incubated in pairs, each pair taken from the same litter; one preparation in each pair was expanded at the end of the first hour by approximately 70%. An adrenergic antagonist was placed in the outer saline as above. The aim of these experiments was to test whether the unknown agent released by expanded lungs affected fluid production in the companion lungs through adrenergic stimulation. Twelve preparations were incubated in pairs, and received no treatment (controls). No significant changes in fluid production were seen (Fig 8 Ai , Aii). a) Expanded preparations Twelve preparations were incubated in pairs; one preparation in each pair was expanded at the end of the first hour by approximately 70% (68.9 ± 2.8%). All six expanded lungs decreased fluid production; three turned to reabsorption. The fluid production rates in successive hours were: before expansion: 1.08 ± 0.24; after expansion: -0.28 ± 0.38, and -0.41 ± 0.23 ml/kg body weight per hour (Fig 8 Bi).The overall effect was a 125.4 ± 42.6% fall in production (P < 0.01, ANOVA). 50 Fig 5: Effect of epinephrine and norepinephrine on lung liquid production in vitro by lungs from near-term fetal guinea pigs. Based on 18 fetuses of 61 ± 1 (SD) days of gestation and 85.8 ± 19.1 (SD) g body weight. During the middle hour, the preparations were immersed in saline containing one of the following (a) saline alone (untreated controls), (b) 2.2 x 10"6 M epinephrine, and (c) 10"5 M norepinephrine. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (a) 0.75 ± 0 . 1 5 (SD), (b) 1.07 ± 0.21, and (c) 0.74 ± 0.14. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 51 80 L 120 a) Control n = 6 b) 2.2 x 10" M Epinephrine n = 6 80 L 120 110 h c) 10" M Norepinephrine n = 6 100 h 90 h 0.32 ± 0 . 2 7 0 SALINE TREATMENT SALINE Time (Hours) 52 Fig 6: The relationship between norepinephrine concentration and the fall in fluid production by in vitro lungs from fetal guinea pigs. Based on 54 fetal guinea pigs of 92.5 ± 18.8 g body weight and 62 ± 2 days of gestation (some data were obtained from previous workers in the laboratory). Percent fall in fluid production represents the change in fluid production rates between resting fluid production (hour 1), and during stimulation with norepinephrine (hour 2). Values are means ± SEM. Number of animals used at each dose are given above the error bars, r = 0.97; P < 0.001. 53 54 Fig 7: Influence of phentolamine and propranolol on responses to lung expansion in vitro. Based on 36 fetuses of 62 ± 3 days of gestation and 99.8 ± 21.8 g (SD) body weight, (a) untreated controls; (b) preparations expanded 70.7% with saline at the end of the first hour; (c) preparations expanded 70.2% with saline, but with 10"5 M phentolamine present in the outer saline; (d) preparations treated with 10"5 M phentolamine alone; (e) preparations expanded 70.2% with saline, but with 10'7 M propranolol present in the outer saline; (f) preparations treated with 10"7 M propranolol alone. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (a) 0.68 ± 0.08 (SD), (b) 0.78 ± 0.23, (c) 1.09 ± 0.30, (d) 1.00 + 0.20 ml, (e) 0.68 ± 0.08 (SD) and (f) 0.78 ± 0.23 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 55 130 120 110 -100 -90 -80 -120 110 100 90 80 «-120 110 100 90 80 L 120 110 100 90 80 «-120 110 100 90 J -80 L 120 110 100 90 80 a) Control n = 6 b) 71% Expansion n = 6 ± 0 . 1 7 c) 70% Expansion/10' 5 M Phentolamine n = 6 • — • --0.28 ± 0 . 3 0 0.19 ± 0 . 2 2 0.36 ± 0 . 2 4 d) 10'5 M Phentolamine n = 6 1.02 ± 0 . 0 9 e) 70% Expansion/10"7 M Propranolol n = 6 -0.02 ± 0 . 3 6 f) 10"7 M Propranolol n = 6 0 SALINE TREATMENT SALINE Time (Hours) 56 Fig 8: Influence of expanded lungs on lung liquid production by unexpanded companion lungs supported in the same saline in vitro; studies in the presence of phentolamine and propranolol. Based on 48 fetuses of 63 ± 2 days of gestation and 106.4 ± 19.3 g (SD) body weight. (Ai, Aii) preparations taken from the same mother and incubated in pairs without treatment (controls). (B) Six pairs of lungs incubated together in the same saline and transferred after 1 h to saline containing 10"5 M phentolamine: (Bi) one preparation from each pair was then expanded 70% with saline; (Bii) the second was left unexpanded. (C) Six pairs of lungs incubated together in the same saline and transferred after 1 h to saline containing 10"7 M propranolol: (Ci) one preparation from each pair was then expanded 70% with saline (n = 6); (Cii) the second was left unexpanded. (D) preparations incubated in pairs with one (Di) expanded 68.9% with saline at the end of the first hour, and the companion lung (Dii) left unexpanded. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (Ai) 0.68 ± 0.08 (SD), (Aii) 0.78 ± 0.23, (Bi) 1.09 ± 0.30, (Bii) 1.00 ± 0.20, (Ci) 0.68 ± 0.08, (Cii) 0.78 ± 0.23, (Di) 1.09 ± 0.30, and (Dii) 1.00 ± 0.20 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 57 O > —^> o H 130 120 110 100 90 i38 110 100 90 80 110 100 90 80 120 110 100 90 80 120 110 100 90 80 120 A i ) lung 1: Control n = 6 1.31 ± 0.12 A i i ) lung 2: Control n = 6 Joint incubations Bi) lung 1: 70% Expansion . -n = 6 Bii ) lung 2: No expansion n = 6 Joint incubations 0.12 ±0 .09 0.29 ±0 .20 Ci) lung 1:71% Expansion/10' 5 M Phentolamine n = 6 -0.36 ±0.36 Cii) lung 2: No Expansion/10' 5 M Phentolamine n = 6 . . - - • • * -0.21 ± 0.20 Joint incubations 120 110 100 -90 -80 -120 110 100 90 h 80 0.35 ±0 .09 Di) lung 1: 70% Expansion/10"7 M Propranolol n = 6 0.42 ±0 .16 0.10 ± 0.28 Dii) lung 2: No Expansion/10"7 M Propranolol n = 6 0.28 ±0.15 Joint incubations 0 S A L I N E j l j T R E A T M E N T I 2 I S A L I N E | 3 Time (Hours) 58 Another twelve preparations were incubated in pairs in the presence of 10"5 M phentolamine; one preparation in each pair was expanded at the end of the first hour by approximately 70% (71.0 ± 0.5%)). All expanded lungs decreased fluid production with four turning to reabsorption. The overall effect was a 124.0 ± 35.2% fall in production (P < 0.001, ANOVA). The fluid production rates in successive hours were: before expansion: 1.51 ± 0.26; after expansion: -0.36 ± 0.36, and -0.21 ± 0.20 ml/kg body weight per hour (Fig 8 Ci). Six pairs of lung preparations were similarly incubated in 10"7 M propranolol; one preparation from each pair was expanded by approximately 70% (68.9 ± 2. 1%). All expanded preparations reduced fluid production; one turned to reabsorption in the hour following expansion. Overall, fluid production was halted (94.7 ± 15.6% reduction) in the middle hour. This fall in fluid production was significant by ANOVA(PO.Ol); there was a tendency towards recovery in the final hour. The fluid production rates in successive hours were: before expansion: 1.83 ± 0.44; after expansion: 0.10 ± 0.28, and 0.28 ± 0 . 1 5 ml/kg body weight per h (Fig 8 Di). The usual fall in fluid production by expanded lungs (Fig 8 Bi) was not significantly altered by either phentolamine or propranolol (ANOVA). Even though the fall in fluid production was considerably less in the presence of propranolol, there was no significant difference when responses were compared directly by ANOVA. These results from jointly incubated preparations confirm the previous responses of expanded preparations incubated alone (Fig 7b). 59 b) Unexpanded companion preparations Six untreated companion lung preparations reduced fluid production; three turned to slight reabsorption. The overall fall in fluid production during the middle hour was 88.2 ± 10.7%. Fluid production rates in successive hours were: 0.98 ± 0.41, 0.12 ± 0.09 and 0.29 ± 0.20 ml/kg body weight per h (Fig 8 Bii). The companion lung preparations incubated in saline containing 10"5 M phentolamine decreased fluid production, but none reabsorbed fluid. The fluid production rates in successive hours were: first hour: 1.41 ± 0.35, in the presence of 10'$ Mphentolamine: 0.35 ± 0.09; in the final hour: 0.42 ± 0.16 ml/kg body weight per h (Fig 8 Cii). The overall fall in fluid production was 75.4 ± 14.3 %; this fall was significant by ANOVA(P < 0.05). When comparisons were made between the percent fall in fluid production by the companion lungs in the presence and in the absence of phentolamine, no significant difference was found (one-way ANOVA). Companion lungs incubated in the presence of 10"7 M propranolol reduced fluid production by only 37.0 ± 13.4% in the middle hour; none reabsorbed fluid. Fluid production recovered well in the final hour. Fluid production rates in successive hours were: first hour: 1.50 ± 0.43; in the presence of 10"^Mpropranolol: 0.94 ± 0 . 1 6 ; in the final hour: 1.09 ± 0.20 ml/kg body weight per h (Fig 8 Dii). The fall in fluid production was not significant. The percent fall in fluid production by the companion lungs in the presence of propranolol was significantly less (P < 0.01, one-way ANOVA) than the percent fall by companion lungs not treated with propranolol 60 (Fig 8 Bii). These observations imply that the P-adrenergic antagonist, propranolol, can reduce the fall in fluid production by the companion lungs. 4.3.3 Are catecholamines released by expanded lungs? Samples from both the bathing saline and lung liquid were analyzed. Norepinephrine was present in all samples examined. Epinephrine was detected in only three of six samples of bathing saline analyzed, while dopamine was present in only two samples. Neither epinephrine nor dopamine were detected in lung liquid. As such, only concentrations of norepinephrine are reported here (Fig 9). The concentrations immediately before lung expansion were: bathing saline: 1.45 ± 0.21 nM; lung liquid: 97.0 ± 12.0 nM. The concentration in bathing saline rose by 50% 10 min following expansion (Fig 9a). This increase was not significant by A N O V A and was not maintained. On the other hand, the concentration of norepinephrine in lung liquid increased significantly by 5 min after expansion (P < 0.05, Scheffe's test); the concentration was maintained for at least 45 min (Fig 9b). 61 Fig 9. Changes in concentrations of norepinephrine in the outer saline (a), and in lung liquid (b) in response to 71% lung expansion with Krebs-Henseleit saline. Based on 12 fetuses of 61 ± 1 days of gestation and 95.2 ± 21.5 g (SD) body weight. Two samples were taken before expansions were given 10 min into the middle hour (indicated by vertical bars); the rest were taken at intervals throughout the middle hour. The volume of the outer saline was 100 ml; lung liquid volume was 1.01 ± 0.22 ml. Ordinates: norepinephrine concentration (nM). The asterisk denotes a significant change from the pre-expansion concentration. Significance was accepted at P < 0.05 (ANOVA). 62 o o o U a o 1 r 0 400 300 200 100 0 a) Bathing saline n = 6 Expansion b) Lung fluid n = 6 Expansion 0 10 20 30 40 50 60 70 Time (min) 63 4.4 Discussion These results confirm that both epinephrine and norepinephrine reduce or reverse lung liquid production in fetal lungs, and demonstrate that norepinephrine is released by expanded fetal lungs. As observed before, norepinephrine reduces lung liquid production at a concentration higher than that of epinephrine (Higuchi, Murata, Miyake, Hesser, Tyner, Keenan and Porto, 1987, Woods and Perks, unpublished observations). The concentrations used in the present study were based on similar studies on in vitro lungs from fetal guinea pigs (Woods et al, 1997, Woods and Perks, unpublished observations). Norepinephrine produced a response comparable in magnitude to that of epinephrine, when used at five times the molar concentration of epinephrine. The overall effects of 2.2 x 10"6 M epinephrine and 10"5 M norepinephrine were strikingly similar; lung liquid production fell by 146.2 ± 23.7 % and 137.3 ± 21.7% respectively. This indicates that, at relatively high concentrations, norepinephrine is able to cause lung liquid reabsorption. Although higher than normal plasma catecholamine concentrations, this concentration of norepinephrine could be achieved in tissues through local release especially during hypoxemia or acidemia (Jelinek and Jensen, 1991). The concentration of epinephrine used in this study was also high, and reduced lung liquid production by 146%). The effect was greater than that seen in earlier studies, where reduction in fluid production was only up to 62% (Woods et al. 1997; Doe and Perks 1998). These earlier workers also found a fall-off in response at higher concentrations of epinephrine, with no response at 10"5 M . They hypothesized that at that concentration, epinephrine stimulated two populations of receptors with opposing effects; one 64 population (a 2 receptors) was sensitive at low epinephrine concentrations, and inhibited fluid secretion; the other (a, receptors) was sensitive only at high hormone concentrations, and opposed this effect. Therefore there is some inconsistency in the reported effects of epinephrine around 10"6 M . The maximal effect seen in previous studies was found at 10"7 M epinephrine, and at 10"6 M the responses had fallen off. In the work reported here 10"6 M epinephrine produced strong effects, so that the concentration producing maximal effects appeared to have changed. The reason for the change is not known; nevertheless, there is agreement in the fact that epinephrine could produce strong effects when at an appropriate concentration. In fetal guinea pig lungs, a-, not P-adrenergic stimulation is thought to be important in the reversal of lung liquid production (Woods et al, 1997). Specific oc-adrenergic receptor populations are known to change in rat lungs during perinatal development (Whitsett, Machulskis, Noguchi and Burdsell, 1982; Latifpour and Bylund, 1983). Therefore, the a,: a 2 -adrenoreceptor ratio could presumably differ between lungs from fetuses of different gestational ages. In studies using specific a, and a 2 adrenoreceptor antagonists on in vitro lungs from fetal guinea pigs, Doe and Perks (1998) showed that epinephrine treatment during a, receptor blockade resulted in increased fluid reabsorption, whereas treatment during oc2 receptor blockade stimulated production. They concluded that a, receptor stimulation caused fluid production while a 2 receptor stimulation caused reabsorption. It follows that for reabsorption to occur, treatment with epinephrine at physiological concentrations must stimulate mainly a 2 receptors; a, receptor stimulation must be insufficient, and cannot push fluid movement in the opposite 65 direction. According to Latifpour and Bylund (1983), the density of a 2 receptors is higher than that of a, receptors in lungs of newborn rats, but declines rapidly in the early postnatal period. They suggested that a 2 receptors may have an important developmental role in the lung. I speculate that the a 2 receptor density in lungs from mature fetal guinea pigs far exceed that of a, receptors and that this imbalance allows fluid reabsorption during stimulation with epinephrine. In contrast to studies of fluid balance in fetal guinea pig lungs, studies in fetal sheep lungs indicate an overwhelming importance of P-adrenoreceptors in the reduction of lung liquid secretion close to term (Walters and Olver, 1978; Lawson, Brown, Torday, Madansky and Taeusch, 1978, Bergman et al., 1980, Walters, Ramsden, Brown, Olver and Strang, 1982). Intravenous infusion of epinephrine in the fetal sheep slowed down lung liquid secretion in fetal lambs up to 129 days gestation; lungs of older fetal lambs reabsorbed fluid (Walters and Olver, 1978). In those same studies, norepinephrine was reported to be incapable of stimulating fluid reabsorption; however, norepinephrine was infused at a rate sufficient to give a concentration in fetal plasma similar to that of epinephrine. This resulted in a concentration of norepinephrine which was likely to be lower than that achieved in fetal lung tissue during hypoxemia (Jelinek and Jensen, 1991), a condition which is usually associated with labour (Lagercrantz and Slotkin, 1986). In our studies, norepinephrine was used at a concentration which was much higher than plasma concentrations achieved during hypoxemia (Cohen et al., 1982, Jelinek and Jensen, 1991) but likely to be achieved within lung tissue. This is because lung tissue norepinephrine levels are a combined result of adrenal and neural discharge (von Euler, 1954; Iversen, 1967, Jelinek and 66 Jensen, 1991). In any event, norepinephrine caused a 45% drop in liquid production by fetal lamb lungs in the study by Higuchi et al. (1987). This strengthens the suggestion here that norepinephrine can reduce production of or cause reabsorption of fetal lung liquid. The bathing saline and lung liquid were analyzed for their concentrations of epinephrine, norepinephrine, and dopamine. Norepinephrine appeared to be the predominant catecholamine in the lungs. Similar results have been reported from studies with fetal lambs (Cohen et ah, 1982) and humans (Bistolleti et al., 1983). Although the concentration of norepinephrine released in lung liquid was higher than that in the bathing saline, the total amounts discharged into the two spaces were even. This discharge was most probably from noradrenergic nerve endings terminating in the lungs. The lung liquid norepinephrine concentration rose shortly after expansion and remained high through the next hour; in the bathing saline, norepinephrine concentration fell to initial levels within 20 min. This implies that norepinephrine entered the lung liquid upon its release shortly after expansion, but could not exit the lungs subsequently. This could happen if epithelial permeability increased only briefly due to expansion, but recovered its tightness shortly afterwards. Such a phenomenon was reported in studies with fetal sheep lungs (Egan et al., 1975). Any possible effects of catecholamines on ion transport could be overshadowed by such an increase in epithelial permeability, which may allow all ions through. If epithelial tightness was not restored for as long as 10 min, any effects of catecholamines would be completely masked; in studies of electrical potential difference (Chapter 8), effects of catecholamines subsided after about 10 min. The possibility of some internal action of 67 catecholamines in expanded lungs appeared doubtful because neither phentolamine nor propranolol could block fluid reabsorption by expanded lungs. These observations suggested that either adrenergic receptors were not involved in the major response of fetal lungs to saline expansion, or their effect was overwhelmed by other factors. Similar observations have been made elsewhere. Garrad Nelson and Perks (1996a) reported that P-adrenergic blockade with propranolol could not affect the fall in fluid production following expansion of similar lung preparations. The use of propranolol in fetal lambs during labour could not decrease lung liquid clearance during birth (Bland et al, 1990, Chapman et al., 1994). These observations suggest that P-adrenergic stimulation does not necessarily play a dominant role in lung liquid clearance at birth. It is also possible that catecholamines do not affect epithelial ion transport when applied on the luminal side. On the other hand, any norepinephrine released on the serosal side was immediately diluted to a low concentration. At such a concentration, norepinephrine was unlikely to maintain the reductions in fluid production seen in the companion lungs. The ot-adrenergic blocker phentolamine did not block the fall in fluid production by unexpanded companion lungs. This suggests that a-adrenergic receptors were not involved in the response. Conversely, the p-adrenergic antagonist propranolol partially blocked the response of companion lungs. The usual drop in fluid production became smaller, and recovery in the final hour was almost complete. This suggested that (1) the expanded lungs released catecholamines and (2) the catecholamines functioned through p-adrenergic stimulation to decrease fluid production by the companion lungs. This observation was surprising for a number of reasons: (1) The 68 concentration of norepinephrine detected in the outer saline after lung expansion was only sufficient to cause a 20% fall in fluid production (Fig 6); by comparison, the companion lungs decreased fluid production by 88%. (2) Since the expanded lungs were not affected by propranolol, why were the companion lungs? (3) In contrast to the observations of Woods et al. (1997), the observation implies that P-adrenergic receptors are involved in liquid balance in fetal guinea pig lungs. At the concentrations measured in the bathing saline, norepinephrine was unlikely to maintain the reductions in fluid production seen in the companion lungs. Furthermore, effects of norepinephrine on lung liquid production are immediate, but are quickly lost when the hormone is withdrawn (Fig 5) or depleted (Woods et al., 1997, Woods and Perks, unpublished observations), yet fluid production was inhibited for two hours. Thus it seemed unlikely that norepinephrine alone could be responsible for the inhibition of fluid production by the companion lungs. If converted to epinephrine, which is more potent, effects may be pronounced even at such low concentrations. Phenylethanolamine N-methyl transferase (PNMT), the terminal enzyme in epinephrine synthesis, is present in fetal lungs; it is also known that lung PNMT is highly active, and has high substrate affinity and specificity (Padbury et al, 1983). Therefore, norepinephrine could have been converted to epinephrine in the companion lung tissue. This could have increased its activity. The fact that the companion lungs decreased fluid production despite the low concentration of norepinephrine in the bathing saline may also suggest synergism with other agents, possibly prostaglandins. Since it is unlikely that any local factors were operating in the companion lungs, 69 they could be influenced only by agents released by expanded lungs. The unexpected partial blockage by propranolol suggested that P-adrenergic receptors may act in fetal guinea pig lungs in the same way as in fetal sheep lungs. The use of propranolol and phentolamine in studies of transepithelial potential difference across fetal guinea pig bronchi indicate that effects of catecholamines are modulated through P-, and not a-adrenergic receptors (Chapter 8). This suggests that part of the fall in fluid production by the unexpanded companion lungs could be the result of fluid movement across the airway epithelium. Chapter V Effects of prostaglandins on lung liquid production 71 5.1 Introduction In the fetus during delivery, the plasma concentration of prostaglandin E 2 (PGE2) increases (Challis, Dilley, Robinson and Thorburn, 1976; Clyman, Mauray, Roman, Rudolph and Heymann, 1980) suggesting it has a physiological role in the term fetus. Indeed prostaglandins are known to have a role in the perinatal regulation of pulmonary vascular resistance (Olley, Heaton and Coceani, 1978; Cassin, Tyler, Leffler and Wallis, 1979; Lock, Olley and Coceani, 1980; Yoshimura, Tod, Pier and Rubin, 1989), maintaining the patency of the ductus arteriosus (Clyman, 1987), surfactant synthesis (Bustos, Ballejo, Giussi, Rosas and Isa, 1978), and alveolar development (Demello, Murphy, Aronovitz, Davies and Reid, 1987, Atsushi, Katayama, Thurlbeck, Matsui, Yasui and Konno, 1995). They are also thought to be involved in the regulation of fetal lung liquid production. In near-term fetal sheep, infusion of PGE 2 reduces lung fluid production (Kitterman, Liggins, Fewell and Lee, 1982). However, the use of inhibitors of prostaglandin synthesis has produced conflicting results. Indomethacin decreased lung liquid production in acute studies of fetal sheep (Cassin, 1984). Wlodek, Harding and Thorburn (1994) confirmed that indomethacin infusion significantly reduced lung liquid production by fetal sheep, but only late in gestation. In longer term studies neither indomethacin nor meclofenamate had any significant effects on fluid production (Kitterman, Liggins, Clements, Campos, Lee and Ballard, 1981). These conflicts may reflect the differences in the methods of estimation of lung liquid production, but could also reflect the ability of prostaglandin inhibitors to eliminate various prostaglandins that have opposing actions. In any 72 event, in all of those studies, the effects of prostaglandins were attributed to changes in the pulmonary vasculature or to other indirect influences on lung liquid production. Several studies have shown that prostaglandins are synthesized by fetal and adult lungs (Robinson, Hardy, and Holgate, 1985; Braunstein, Labat, Brunelleschi, Benveniste, Marsac and Brink, 1988; Abman and Stenmark, 1992; Hume, Cossar, Kelly, Giles, Hallas Gourlay and Bell, 1992). Specifically, epithelial cell cultures from human fetal lung synthesize and release prostaglandins into the culture medium (Hume et al., 1992). In perinatal lambs, prostacyclin (6-keto-PGIla) is apparently released spontaneously into the fetal lung lumen (Abman and Stenmark, 1992). In addition, inflation of fetal and adult lungs leads to the release of several prostaglandins (Berry, Edmonds and Wyllie, 1969; Said, Kitamura and Vreim, 1972; Leffler, Hessler and Terragno, 1980; Leffler, Hessler and Green, 1984). Recent evidence suggests the release, by expanded fetal lungs, of an agent that is capable of inhibiting fluid production in another lung preparation in vitro (Garrad-Nelson and Perks, 1996b). In the present study, I (1) present evidence that prostaglandins can alter lung liquid production in in vitro lung preparations lacking a functional vasculature and (2) show that inhibition of prostaglandin synthesis decreases the fall in fluid production by unexpanded lungs incubated alongside expanded ones in vitro. 73 5.2 Materials and Methods 5.2.1 Animals Pregnant albino guinea pigs of an inbred departmental stock were given food and water ad libitum (guinea pig chow, Ralston-Purina, supplemented with fresh vegetables and vitamin C). Studies were performed on 71 in vitro lungs from fetal guinea pigs. The fetuses averaged 90.8 ± 22.4 g (SD) body weight and 62 ± 2 days (SD) of gestation. 5.2.2 Lung liquid production To test the hypothesis that prostaglandins can affect lung liquid production in near term fetal lungs, lung preparations were incubated in the presence of various prostaglandins as follows: (1) 10"6 M PGE 2 , (2) 10"6 M PGI2 (3) 10"6 M PGD 2 and (4) 5 x 10"7 M PGF 2 c c , during the middle hour. PGD 2 and P G F 2 a were dissolved initially in a minimum of ethanol, and then diluted with Krebs-Henseleit saline to a final concentration of 0.05% and 0.1 %, respectively; suitable controls of ethanol diluted similarly were also tested, along with untreated controls. In order to test the possibility that prostaglandins are involved in the reduction in fluid production by the companion lung preparation, paired preparations were incubated in the presence of 1.5 x 10"5 M indomethacin. Indomethacin was dissolved in a minimum of methanol and diluted as above to a 74 final methanol concentration of 0.15%; appropriate methanol controls were also tested. The prostaglandin inhibitor was placed in the outer saline during the last 10 min of the initial hour and maintained during the middle hour; one lung preparation from each pair was expanded at the start of the middle hour as previously described. To test the hypothesis that PGE 2 is released by expanded fetal lungs, 1 ml aliquots of saline bathing the lungs were sampled immediately before, and 10 min after lung expansion for quantitative analysis of PGE 2 . The samples were placed in polyethylene micro test tubes (1.5 ml Eppendorf, Brinkmann Instruments Ltd., Rexdale, ON) and immediately frozen at -70°C until analyzed. Prostaglandin E 2 (PGE2) was purchased from Sigma Chemicals (St. Louis MO). All other prostaglandins were kindly provided by Dr. Sidney Cassin, University of Florida, Gainsville. Indomethacin was purchased from Research Biochemicals International (Natick, MA). 5.2.3 Analytical procedures: Enzyme-Linked Immunosorbent Assay (ELISA) for PGE2 The assay was done on unextracted samples using a PGE 2 ELISA Kit (Neogen 404110, Lexington, KY) having 100% cross reactivity with PGA„ PGA 2 , PGB„ PGB 2 , PGE, and 86% cross reactivity with 6-keto-prostaglandin E, . The test kit operates on the basis of competition between the enzyme conjugate and the PGE 2 for a limited number of binding sites on an antibody coated microplate. Standards were diluted in Krebs-Henseleit saline. Duplicate samples were read simultaneously using a microplate reader (Model 3550, Bio-Rad Laboratories, Richmond, 75 CA) set at dual wavelengths (measurement X - 650, reference X = 490) to reduce artificial well to well variation. Assay sensitivity was 0.15 ng/ml (4.3 x 10"10 M). 5.3 Results 5.3.1 Effects of various prostaglandins on lung liquid production To test the hypothesis that prostaglandins can reduce lung liquid production near term, six fetal lungs were incubated in the presence of various prostaglandins. Concentrations of prostaglandins were recommended by Dr. Sidney Cassin, University of Florida, Gainsville. Untreated control preparations produced fluid steadily through three successive hours with no significant change (Fig 10a). Six preparations were treated with 10"6 M PGE 2 during the middle hour. The lungs either decreased liquid production or reabsorbed fluid. The rates in successive hours were: before treatment, 1.09 ± 0 . 1 2 , during treatment, 0.05 ± 0.28, and after treatment, 0.17 ± 0.17 ml/kg body weight per hour (Fig 10b). The decrease in the middle hour represented a 95.2 ± 26.2% fall in production and was significant by A N O V A (P < 0.05). Six preparations, incubated in the presence of 10"6 M PGI2 during the middle hour, showed no significant Change in lung liquid production following treatment (Fig 10c). PGD 2 and PGF2 a were dissolved in ethanol and diluted to give a final concentration of 0.1%; ethanol alone (1.0%) did not significantly affect lung liquid production (Fig lOd). Nine preparations were incubated in the presence of 10"6 M PGD 2 , dissolved in ethanol (see Methods), during the middle hour. Altogether, eight of nine 76 preparations decreased fluid production; none reabsorbed fluid. Fluid production rates in successive hours were: before treatment, 1.15 ± 0.16, during treatment, 0.56 ± 0.11, and after treatment, 0.73 ± 0 . 1 6 ml/kg body weight per hour (Fig lOe). The overall fall in fluid production was 51.4 ± 13.3%; this fall was significant by A N O V A (P < 0.005). There was a tendency towards recovery in the final hour. Six preparations were incubated in the presence of 5 x 10"7 M PGF 2 a , dissolved in ethanol as above. All preparations decreased fluid production; four preparations turned to reabsorption in the middle hour. There was difficulty withdrawing fluid from two of the reabsorbing preparations during the final hour. The combined rates in successive hours were: before treatment, 0.91 ± 0.20, during treatment, -0.28 ± 0.36, and after treatment, -0.24 ± 0.54 ml/kg body weight per hour (Fig lOf). The overall fall in fluid production in the middle hour was 130.8 ± 48.5%) and was significant by A N O V A (P < 0.05). It was concluded that while PGE 2 and P G F 2 a could halt lung liquid production or cause reabsorption, PGD 2 could only reduce flow slightly; PGI2 had no effect. 5.3.2 Effects of indomethacin on responses to expansion I Single incubations In order to test whether the reduction in fluid production by expanded or companion lung preparations was due to prostaglandins, preparations were incubated in the presence of indomethacin, dissolved in methanol to give a final concentration of 0.15%. Neither methanol 77 nor 1.5 x 10"5 M indomethacin used alone (controls) had any significant effect on lung liquid production (Fig 11). II Joint incubations a) Controls Lung preparations were incubated in pairs, each pair taken from the same litter. Twelve untreated control preparations, incubated in pairs, produced fluid for three successive hour with no significant change in rate (Fig 12 Ai , Aii). b) Expanded preparations Twelve preparations were incubated in pairs; one preparation in each pair was expanded at the end of the first hour by approximately 70% (68.9 ± 2.8%). All six expanded lungs decreased fluid production; three turned to reabsorption. The overall effect was a significant (125.4 ± 42.6%), P < 0.01, ANOVA) fall in production (Fig 12 Bi). Twelve preparations were incubated in the presence of 1.5 x 10"5 M indomethacin, placed in the outer saline 10 min before one preparation from each pair was expanded by approximately 70 % (69 2 + 2.1%), and maintained during the following hour. All expanded preparations decreased fluid production or reabsorbed fluid in the middle hour. The fluid production rates in successive hours were: before expansion, 78 1.56 ± 0.37, after expansion, -0.74 ± 0.55, and-0.19 ± 0.64 (Fig 12 Ci). The overall fall in fluid production in the middle hour was 147.8 ± 18.7%; this fall was significant by A N O V A (P < 0.05), but was not significantly different from that of expanded preparations incubated with no indomethacin in the outer saline (Fig 12 Bi). This suggests that indomethacin had no effect on the expanded preparations. c) Unexpanded companion preparations Six untreated companion lung preparations reduced fluid production; three turned to slight reabsorption. The overall fall in fluid production during the middle hour was 88.2 ± 10.7% (Fig 12 Bii). Six companion preparations incubated with indomethacin in the bathing saline decreased fluid production, but none reabsorbed fluid. The fluid production rates were: first hour: 1.36 ± 0.14, in the presence of 1.5 x 10~s M indomethacin: 0.88 ± 0.06; final hour: 0.64 ± 0.12 ml/kg body weight per h (Fig 12 Cii). The overall fall in fluid production was (34.9 ± 2.9%), and was significant by A N O V A (P 0.001). This fall was significantly less (P < 0.001, ANOVA) than that by companion lungs incubated with no indomethacin the outer saline (Fig 12 Bii). It was concluded that prostaglandins may be involved in the reduction of fluid production by expanded lungs. This implied that prostaglandins could have been released by expanded lungs into the bathing saline. However, no PGE 2 was detected by ELISA in the saline bathing the lung preparations before or after expansion. If present, the concentration of PGE 2 was below the assay limit of 0.15 ng/ml (4.3 x 10"'° M). 79 These observations suggest that some prostaglandins can reduce fluid production, but the use of the cyclooxygenase inhibitor, indomethacin, implies that either (1) they are not concerned in the effects of expansion on expanded lungs, or (2) in their absence, other mechanisms continue to maintain the full response. The apparent reduction of effects of expansion on companion lung preparations in the presence of indomethacin suggests that prostaglandins could be involved in the reduction of fluid production by companion preparations. 80 Fig 10: Effect of various prostaglandins on lung liquid production in in vitro lungs from fetal guinea pigs. Based on 39 fetuses of 61 ± 1 (SD) days of gestation and 85 ± 19.1 g body weight. During the middle hour, the preparations were immersed in saline containing one of the following: (a) saline alone (untreated controls), (b) 10"6 M PGE 2 , (c) 10"6 M PGI2, (d) 1.0% ethanol, (e) 10"6 M PGD 2 and (f) 10"6 M PGF 2 a . Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (a) 0.68 + 0.08 (SD), (b) 0.78 ± 0.23, (c) 1,09 ± 0.30, (d) 1.00 ± 0.20, (e) 0.68 ± 0.08, (f) 0.78 ± 0.23, ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 81 a > +-> o H 120 110 100 90 80 110 100 h a) Control n = 6 b) 1 0 ' b M P G E 2 n = 6 90 120 r 110 -100 -90 -120 r 0.05 ± 0 . 2 8 c) 1 0 ° M P G I n = 6 0.79 ± 0 . 1 7 0.17 ±0 .17 0.70 ±0 .31 cf) 1.0% Ethanol n = 6 no -100 -90 -80 -120 r e) 1 ( T M P G D : n = 9 f )5x 1 0 ' 7 M P G F n = 6 •2a -0.28 ± 0 . 3 6 -0.24 ± 0 . 5 4 0 SALINE TREATMENT SALINE Time (Hours) 82 Fig 11: Lung liquid production in in vitro lungs from fetal guinea pigs. Based on 39 fetuses of 61 ± 1 (SD) days of gestation and 85 ± 19.1 g body weight. During the middle hour, the preparations were immersed in saline containing one of the following: (a) saline alone (untreated controls), (b) 0.5% methanol, and (c) 1.5 x 10"5 M indomethacin Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100%) was (a) 0.68 ± 0.08 (SD), (b) 0.78 ± 0.23, (c) 1.09 ± 0.30, All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. 83 a) Control n = 6 1.36 ±0.27 b) 0.5% Methanol n = 6 0.97 ±0.27 0.84 ±0.14 c) 1.5 x 10" M Indomethacin n = 6 100 h 90 h 0 SALINE TREATMENT SALINE Time (Hours) 84 Fig 12: Influence of expanded lungs on lung liquid production by unexpanded companion lungs supported in the same saline in vitro; studies in the presence of indomethacin. Based on 48 fetuses of 63 ± 2 days of gestation and 106.4 ± 19.3 g (SD) body weight. (Ai, Aii) preparations taken from the same mother and incubated in pairs without treatment (controls). (Bi) expanded 68.9% with saline at the end of the first hour, and the companion lung (Bii) left unexpanded. (C) Six pairs of lungs incubated together in the same saline and transferred after 1 h to saline containing 1.5 x 10"5 M indomethacin: (Ci) one preparation from each pair was then expanded 70%o with saline; (Cii) the second was left unexpanded. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100%) was (Ai) 0.68 ± 0.08 (SD), (Aii) 0.78 ± 0.23, (Bi) 1.09 ± 0.30, (Bii) 1.00 ± 0.20, (Ci) 1.09 ± 0.30, and (Cii) 1.00 ± 0.20 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 85 i i O > o H (TVS 130 120 110 100 90 80 120 110 100 90 80 110 100 90 80 L 120 110 100 90 80 120 r Ai) lung 1: Control Joint incubations Aii) lung 2: Control Bi) lung 1: 70% Expansion n = 6 Bii) lung 2: No expansion n = 6 Joint incubations - • — • — • 0.12 ±0.09 Ci) lung 1: 70% Expansion/1.5 x 10"5 M Indomethacin n = 6 • — • — • -0.29 ±0.20 ±0.37 -0.19 ±0.64 - • — • 80 L 120 110 100 90 80 Cii) lung 2: No Expansion/1.5 x 10"5 M Indomethacin n = 6 . . - * " * *~~ 0.64 1-36 ±0.06 ±0.14 i I Joint incubations 0 SALINE l i | TREATMENT | ? SALINE Time (Hours) 86 5.5 Discussion These observations suggest that (1) some prostaglandins are capable of reducing lung liquid production (2) the use of the cyclooxygenase inhibitor, indomethacin, fails to stop the effects of expansion on expanded lungs, either because prostaglandins are not involved in the response, or because other parallel mechanisms maintain full effects even when effects of prostaglandins are abolished, and (3) indomethacin can reduce but not abolish effects on unexpanded companion lungs. Therefore, prostaglandins can be involved in responses to expansion, but are not necessary for their full expression. Exogenous PGE 2 halted lung liquid production in our in vitro fetal guinea pig lung preparations; PGD 2 reduced it. These results agree with those of Kitterman et al. (1982) and Cummings (1995) who reported decreased lung liquid production following infusion of those prostaglandins into fetal lambs. Our in vitro lung preparations had no functional circulation, therefore the fall in fluid production was not likely to be the result of changes in the pulmonary vasculature. In our model, the lungs comprise two fluid compartments, the luminal and interstitial compartments. Thus any change in intraluminal fluid volume is a direct result of changes in the pulmonary epithelium which remains the only barrier to fluid movement, except for some resistance from supporting tissues. Most of the discussion on the effects of the prostaglandin system in fetal lungs has focused on the tone and permeability of the pulmonary vasculature (Cassin 1984; Yoshimura et al., 1989; Wallen, Murai, Clyman, Lee, Mauray, Ballard and Kitterman, 1989; 87 Velvis, Moore and Heyman, 1991; Cummings, 1995). Little attention has been paid to the direct effects of prostaglandins on fluid movement across the lung epithelium. This study suggests that both PGE 2 , PGD 2 and P G F 2 a can act directly on the epithelium to alter the direction of lung liquid movement. This fact is also suggested by the work of McCray and Bettencourt (1993), although they observed stimulation, not inhibition, of lung liquid production. Furthermore, immunohistochemical studies in human fetal lung have shown that epithelial cells are just as reactive to PGE 2 as endothelial cells (Hume et al., 1992). To reduce fetal lung liquid production, prostaglandins probably act on the pulmonary epithelium by stimulating production of adenosine 3', 5' cyclic monophosphate (cAMP) in the cells. PGE, increases cAMP in human lung tissue (Tauber, Kaliner, Stechshulte and Austen, 1973), while PGE 2 increases cAMP in isolated tracheal epithelial cells from adult rabbits (Liedtke, 1986), and from infant and adult ferrets (Kercsmar, Chung and Davis, 1991). The cAMP pathway also modulates fetal lung liquid reabsorption in both sheep and guinea pigs (Olver, Ramsden and Walters, 1987; Barker, Brown, Ramsden, Strang and Walters, 1988; Walters Ramsden and Olver, 1990; Kindler, Ziabaksh and Perks, 1992). Fluid production by in vitro lungs treated with PGE 2 tended to recover when preparations were transferred into fresh saline. This response is similar to that seen following treatment with catecholamines (Woods, Doe and Perks, 1997), which stimulate lung liquid reabsorption through the cAMP system (Strang, 1991). 88 As indicated above, McCray and Bettencourt (1993) observed that PGE 2 caused stimulation, not inhibition, of fetal lung liquid production. There were methodological differences between my experimental method and theirs, which could be responsible for these different results. First, whereas they had used first trimester peripheral lung tissue explants, I used an in vitro preparation of whole lungs from late gestation fetal guinea pigs. Second, they had estimated lung liquid production indirectly by measuring luminal surface area of the explants; I measured liquid production more directly by a dye dilution technique. They concluded that PGE 2 participated in the stimulation of fetal lung liquid production. The response they observed was similar to that of P-adrenergic agonists on premature fetal lungs (McCray, Bettencourt and Bastacky, 1992a; 1992b). Lungs from late gestation fetuses such as ours reabsorb fluid in response to p-adrenergic stimulation (Brown, Olver, Ramsden, Strang and Walters, 1983). This phenomenon probably reflects the maturation of pulmonary epithelial expression of Na + transport proteins (Tchepichev, Ueda, Canessa, Rossier and O'Brodovich, 1995; Crump, Askew, Wert, Lingrel and Joiner, 1995). If PGE 2 and P-adrenergic agonists trigger similar intracellular events, they may both stimulate fluid production early in gestation, and cause reabsorption when the lung epithelium develops its ability to reabsorb Na + later in gestation. The discrepancy between the two studies probably reflects a difference between the signaling mechanisms for PGE 2 at these two stages of fetal lung development, or more simply, in the age of the tissues studied. The use of inhibitors of prostaglandin synthesis by other workers has produced equivocal results. In contrast to the present results, infusion of meclofenamate into fetal lambs significantly 89 reduced plasma PGE 2 concentration yet did not affect tracheal fluid production (Wallen et al, 1989). This may mean that at normal plasma concentrations, PGE 2 does not affect fluid production so that its removal is without effect on production rates. In acute studies on fetal sheep, indomethacin reduced fetal lung liquid production (Cassin 1984); longer term use of these inhibitor was without effect (Kitterman et al, 1981). In contrast, infusion of indomethacin into late gestation fetal sheep changed fetal lung liquid flow from net efflux to net influx (Wlodek et al., 1994). These discrepancies are probably because inhibitors of prostaglandin synthesis decrease the concentrations of all prostaglandins; this may complicate interpretation of results if the prostaglandins have opposing effects. The in vitro lung preparation allowed us to (1) test the effects of exogenous prostaglandins in a system that does not have a functional vasculature, and (2) avoid altogether the use of prostaglandin inhibitors. Thus we could directly test the effects of specific prostaglandins on fluid transport across the pulmonary epithelium. Prostaglandin D 2 (PGD2) is reported to be a potent bronchoconstrictor in adult humans (Robinson et al., 1985; Cassin, 1987). The use of PGD 2 did not prevent sampling of fluid or reduce apparent fluid volume in our in vitro lung preparations. If observed, these events could have indicated an adverse effect of PGD 2 on airway muscular tone and integrity. The small change in fluid production seen following treatment with PGD 2 most probably reflected events within the epithelium itself. The significance of the change is not clear, but prostaglandins E 2 and D 2 act in the same fashion in the fetal pulmonary vasculature (Cassin, Tod, Philips, Frisinger, Jordan and Gibbs, 1981; Cassin, 1987); therefore the similarity between their actions here, is not surprising. 90 Prostaglandin F 2 a (PGF 2 a) reduced lung liquid production significantly or caused fluid reabsorption; however, it also increased the resistance to withdrawal of lung liquid in some preparations, suggesting perhaps that it may have constricted the airways. Like PGD 2 , P G F 2 a is a potent bronchoconstrictor, and can cause intense bronchospasm in adult human lungs (Cuthbert, 1973). However, since fetal lungs often respond differently from adult lungs, it is not certain that, in our lung preparations, P G F 2 a constricted airways and restricted access to lung liquid in the distal airspaces. In contrast to PGE 2 , PGD 2 and PGF 2 a , prostacyclin (PGI2) did not affect lung liquid production in our in vitro lung preparations. In lungs of newborn lambs, PGI2 increased fluid filtration by increasing pulmonary vascular surface area (Yoshimura et al., 1989); however, it was not clear whether the filtered fluid increased lung liquid production. Nevertheless this action on the pulmonary vasculature is likely to increase, not reduce, lung liquid production; therefore, our results are not in conflict. Furthermore, in human lungs, PGI 2 is either without effect, or can induce slight bronchodilation, and antagonize bronchoconstriction by other agents (Bianco et al., 1978). If occurring, none of these actions was likely to reduce lung liquid production significantly in our preparations. There was no measurable concentration of PGE 2 , in the fluid bathing expanded lungs. The assay also suggested that PGA,, PGA 2 , PGB,, PGB 2 or PGE, 6-keto-prostaglandin E, were not released 91 in measurable quantities. This was surprising because the release of PGE 2 and other prostaglandins by inflated lungs has been demonstrated (Berry et al, 1969; Said et al, 1972; Leffler et al, 1980). However, the amount of prostaglandins released, if any, would have been dispersed in 100 ml of saline. The volume of bathing saline used was 100 ml in order to mimic the experiments in which expanded lung preparations affected unexpanded ones incubated in the same saline. The concentration of exogenous PGE 2 that completely inhibited liquid production by in vitro lungs was 10'6 M , at least 2000-fold the sensitivity limit of the assay (0.15 ng/ml or 4.3 x 10"'° M). Even though the threshold concentration of PGE 2 was not determined, it is unlikely that the unknown agent was PGE 2 . This conclusion was supported by the failure of indomethacin to block or reverse the fall in liquid production by the companion lungs. However, the fall in production in the presence of indomethacin, though significant, was significantly smaller than the fall that occurred in its absence. This suggests that even though indomethacin did not cancel the usual fall in fluid production by the companion lungs, it could reduce it. I am faced with apparently conflicting data within the experiments. While inhibition of prostaglandin synthesis with indomethacin failed to eliminate effects of expansion on expanded lungs, it did reduce (but not abolish) responses in unexpanded companion lungs. This suggests that prostaglandins are not responsible for effects in expanded lungs, but may influence companion lungs. Since the prostaglandins, if released, would be expected to be in the highest 92 concentration where they are produced, not diluted by external saline, their influence would be greatest in the expanded lungs. We must conclude that although prostaglandins are capable of acting within expanded lungs, and can spread their influence to other lungs, they are not a vital part of the mechanisms involved, and in their absence, other mechanisms triggered by expansion can produce full effects. The apparent inhibition of companion lungs by indomethacin could also mean that prostaglandin release occurs in the lung tissue as a result of stimulation by an intermediate substance that diffuses from the expanded preparation to the unexpanded one. What this substance could be is not clear at this point. As reported in Chapter 3, treatment of companion preparations with phentolamine had no effect, treatment with propranolol inhibited the usual fall in fluid production, p-adrenergic stimulation by norepinephrine may cause the release of a prostaglandin in lung tissues. This is possible since there is evidence of functional coupling of P-adrenergic agonists and PGE 2 in the production of cAMP, in airway epithelia of adult and infant ferrets (Kercsmar et al, 1991). cAMP is known to modulate fluid reabsorption by lungs late in gestation (Kindler et al., 1992). Prostaglandins of the dilator type decrease lung liquid production in fetal sheep with an intact pulmonary circulation; however, the dilation probably has little to do with the reduction, and may be related instead to the stimulation of cAMP (Cummings, 1995). However, since indomethacin caused only a very slight reduction in the fall in fluid production by companion lungs it must be concluded that prostaglandins are not the only factors involved in the response. 93 In conclusion, prostaglandins may well have a direct role in altering transepithelial fluid transport in mature fetal lungs. The concentrations of PGE 2 in fetal plasma increase during delivery (Challis et al, 1976, Clyman et al, 1980) and we show here that it can reduce lung liquid production by in vitro lung preparations that lack a functional circulation. The effect, on fluid reabsorption, of prostaglandin released by inflated fetal lungs may be overshadowed by other factors since lungs expanded after preincubation in indomethacin still reabsorbed fluid. Still, prostaglandins seem to be linked to the remote effect of expansion on unexpanded companion lungs. Our companion lungs may have been detecting prostaglandin mechanisms which can operate in expanded lungs; however, these effects are probably overwhelmed by other factors triggered by expansion. Prostaglandins may only be an insurance for fluid transport in case other mechanisms fail. Chapter VI Effects of calcium and its secretagogues on lung liquid production 95 6.1 Background The production of fetal lung liquid slows down late in gestation and turns to reabsorption just prior to birth to clear the future air space in preparation for postnatal gas exchange (Kitterman, Ballard, Clement, Mescher and Tooley, 1979; Dickson, Maloney and Berger, 1986; Perks, Dore, Dyer, Thorn, Marshall, Ruiz, Woods, Vanderhorst and Ziabaksh, 1989). A significant amount of the fluid is reabsorbed during labour, but some fluid remains in the lung and is reabsorbed following the onset of breathing (Adams, Yanagisawa, Kuzela, and Martinek, 1971; Bland, McMillan, Bressack and Dong, 1980). The reabsorption of fluid during labour is thought to be linked to the concurrent surge of catecholamines in the fetal circulation (Walters and Olver, 1978; Brown, Olver, Ramsden, Strang and Walters, 1983; Lagercrantz and Slotkin, 1986). In sheep, epinephrine in particular, is thought to interact with (3-adrenoreceptors and raise intracellular cyclic AMP levels, which in turn increase amiloride-sensitive Na + transport across the fetal lung epithelium, resulting in fluid reabsorption (Olver, Ramsden and Walters, 1987; Walters, Ramsden and Olver, 1990). The importance of cyclic AMP in fluid reabsorption has been further demonstrated in studies of lung liquid production by in vitro lungs from fetal guinea pigs (Kindler, Ziabaksh and Perks, 1992). Although oc2 receptor stimulation generally activates an inhibitory G protein (G() which inhibits adenylate cyclase (Bylund, 1988), it also increases cAMP in other tissues including adult guinea pig lungs (Palmer, 1971). Recent evidence suggests that stimulation of a 2 receptors may cause 96 fluid reabsorption in lungs from fetal guinea pigs, a response that is linked to increased intracellular cAMP levels (Woods, Doe and Perks, 1997; Doe and Perks, 1998). The lungs are considerably expanded following the first breath, an event which is associated with a very rapid reabsorption of lung liquid into the interstitium (Humphreys, Normand, Reynolds and Strang, 1967). This reabsorption continues for several hours after birth (Adams et al, 1971; Bland et al, 1980; O'Brodovich et al, 1990). On the other hand, whereas fetal plasma catecholamine concentrations are high during birth, they return to normal only 2 hours later (Lagercrantz and Slotkin, 1986). In addition, while catecholamine-induced fluid reabsorption is believed to depend on amiloride-sensitive Na + transport, a significant component of reabsorption following the first breath appears to be insensitive to amiloride (O'Brodovich, Hannam, Seear and Mullen, 1990; Song, Sun, Curstedt, Grossmann and Robertson, 1992). Studies from our laboratory have shown that lung expansion with a volume of saline approximating the first breath causes significant reabsorption without any damage to lung tissue; fluid reabsorption can continue for two hours after expansion (Garrad Nelson and Perks, 1996a). Such a level of saline expansion does not increase aqueous pore radius in the fetal sheep pulmonary epithelium (Egan, Olver and Strang, 1975). In addition, expanded lungs release several prostaglandins (Berry, Edmonds and Wyllie, 1969; Said, Kitamura and Vriem, 1972; Leffler, Tyler and Cassin, 1978; Leffler, Hessler and Green, 1984; Leffler, Hessler and Terragno, 1980; Hasan, Olson, Rigaux, Bano, Pankovich and Connors, 1996). Prostaglandin release in 97 guinea pig lungs is thought to be mediated, among other things, by bradykinin (Piper and Vane, 1969). Tissues release bradykinin in response to a variety of factors including shear stress (Wachtfogel, DeLa Cadena and Colman, 1993), and bradykinin is known to increase intracellular calcium concentration in cultured tracheal epithelial cells (Harris, Baimbridge, Bridges and Phillips, 1991; Harris and Hanrahan, 1993). Histamine can be released whenever there is non-specific cell damage or trauma from any cause (Douglas, 1980), and like bradykinin, histamine increases intracellular calcium concentration in cultured tracheal epithelial cells (Harris and Hanrahan, 1993; 1994). Recently, Garrad Nelson and Perks (1996b) demonstrated the release, from expanded fetal lungs, of an unknown agent that can reduce fluid production in another fetal lung. Mechanical distention of cultured rat alveolar type II cells mobilizes cytosolic Ca 2 + only briefly, but this mobilization is usually followed by a sustained secretion of pulmonary surfactant (Wirtz and Dobbs, 1990). I hypothesize that prolonged fluid reabsorption following fetal lung expansion is due to the action of an active agent, released by expanded lungs, that can increase intracellular calcium-dependent processes within the epithelium. In this study, I examine the effects of agents such as A23187, bradykinin, and histamine, which are known to increase intracellular Ca 2 + concentration, on lung liquid production. I also test the effects of a histamine antagonist on the expansion-induced inhibition of fluid production 98 6.2 Materials and Methods 6.2.1 Animals Pregnant albino guinea pigs of an inbred departmental stock were given food and water ad libitum (guinea pig chow, Ralston-Purina, supplemented with fresh vegetables and vitamin C). Studies were performed on 73 fetuses of 61 ± 2 days of gestation (term, 67 days) and average body weight 87.9 ± 22.2 g (SD). 6.2.2 Studies of lung liquid production Initial studies to demonstrate the effects of calcium on fetal lung liquid production were made on in vitro lungs from 6 fetal guinea pigs using 10'6 M A23187 (calcimycin), a calcium ionophore. The ionophore was dissolved in a minimum of ethanol and was diluted to a final concentration of 0.05%. Bradykinin was tested on fetal lungs at two concentrations: 10"8 and 10"7 M . At least 10"8 M bradykinin is needed in the bathing medium to cause a significant increase in intracellular Ca 2 + concentration (Harris et al., 1991). Histamine was used at 10"6 M ; these concentrations were based on Harris and Hanrahan (1993; 1994). In order to test the hypothesis that the fall in fluid production by the companion lung was due to the action of histamine released by the expanded lungs, preparations were incubated in pairs; one preparation was expanded and the second was left unexpanded. 10"6 M mianserin (H, histamine antagonist) was placed in the 99 bathing saline 10 min before one of the lungs was expanded by 73% (72.9 ± 3.2%);the antagonist was maintained during the following hour. All agents were obtained from Research Biochemicals International (RBI, Natick, MA). 6.3 Results 6.3.1 Effects of calcium secretagogues Untreated lung preparations produced fluid steadily for through three consecutive hours (Fig 13a). To study the effects of calcium on fetal lung liquid production, the calcium ionophore, A23187 (IO6 M), was tested on 6 lung preparations. The ionophore was dissolved in ethanol and diluted to a final concentration of 0.05%; ethanol alone did not affect lung liquid production significantly (Fig 13b). All preparations treated with A23187 strongly reduced fluid production in the middle hour (overall 95.0 ± 8.3% fall) and reabsorbed fluid in the final hour (140.5 ± 15.8%) fall). The average rates of fluid production in successive hours were: before treatment: 1.87 ± 0.34; during treatment: 0.09 ± 0 . 1 2 and after treatment: -0.47 ± 0.28 ml/kg body weight per hour (Fig 13c). These changes were significant by A N O V A (P < 0.001) and by Scheffe's post hoc test (hour 1 vs hour 2: P < 0.005, hour 1 vs hour 3; P < 0.001). The ionophore had a prolonged effect on liquid production. 100 Fig 13: Effects of various calcium secretagogues on lung liquid production in vitro by lungs from near-term fetal guinea pigs. Based on 36 fetuses of 61 ± 1 (SD) days of gestation and 85.8 ± 1 9 . 1 (SD) g body weight. During the middle hour, the preparations were immersed in saline containing one of the following (a) saline alone (untreated controls), (b) 1.3% ethanol, (c) 10"6 M A23187, (d) 10"8 M bradykinin, (e) 10"7 M bradykinin, and (f) 10-6 M histamine. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (a) 0.75 ± 0 . 1 5 (SD), (b) 1.07 ± 0.21, (c) 0.74 ± 0.14, (d) 0.74 ± 0.14, (e) 1.07 ± 0.21, and (f) 0.74 ± 0 . 1 4 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 101 120 110 100 -90 -80 -120 r 110 100 90 80 120 r 110 -100 -90 -80 -110 r 100 -90 L 110 100 90 L 110 r 100 -90 -80 -a) Control n = 6 b) 1.3% Ethanol n = 6 c) KTMA23187 n = 6 0.09 ± 0 . 1 2 ± 0 . 3 4 d) 10'8 M Bradykinin n = 6 1.33 • ± 0.08 e) 10 ' 7 M Bradykinin n = 4 1.30 ±0 .13 f) H T M Histamine n = 6 ± 0 . 1 7 -0.47 ±0 .28 0.35 ±0 .15 0 I SALINE 11 I TREATMENT | ? | SALINE Time (Hours) 102 Bradykinin was tested on lung preparations at 10"8 M and 10"7 M respectively. Six preparations incubated in the presence of 10"8 M BK reduced fluid production but none reabsorbed fluid; another six preparations responded similarly to 10"7 M BK. Average rates in successive hours were: 10"8 M BK: before treatment: 1.33 ± 0.08; during treatment: 0.64 ± 0.14 and after treatment: 0.57 ± 0 . 1 5 ml/kg body weight per h (Fig 13d); 10"7 M BK: before treatment: 1.30 ± 0.13; during treatment: 0.61 ± 0 . 1 7 and after treatment: 0.45 ± 0 . 1 6 ml/kg body weight per hour (Fig 13e). Fluid production rates fell as follows : 10"8 M BK: middle hour, 52.2 ± 11.4%; final hour, 57.1 ± 11.6%; 10'7 M BK: middle hour, 53.0 ± 18.5%; final hour, 65.4 ± 13.4%. These reductions were all significant by A N O V A (P < 0.01). The fluid production rates in the final hour appeared to be lower than second hour rates; however, there was no significant difference in fluid production rates between the two hours. As with A23187, these results show persistent effects of bradykinin on lung liquid production. Histamine was used at 10"6 M. Two of eight lung preparations studied reabsorbed fluid in the middle hour; two reabsorbed in the final hour; the rest decreased fluid production in both hours. There was no evidence of recovery in all lungs except for one preparation. Fluid production rates in successive hours were: before treatment: 1.27 ± 0.13; during treatment: 0.46 ± 0 . 1 5 and after treatment: 0.35 ± 0 . 1 5 ml/kg body weight per hour (Fig 13f). Fluid production fell in the middle hour by 63.9 ± 13.0% and in the final hour by 72.4 ± 12.4%. These falls were significant by A N O V A (P < 0.01 and 0.005, respectively). As in the cases of A23187 and bradykinin, these reductions in fluid production were long lasting. 103 6.3.2 Effects of Mianserin In order to test the hypothesis that histamine is released from expanded lungs, and could influence the unexpanded companion lungs, lung preparations were incubated in pairs and treated with 10"6 M mianserin as described (see Experimental Procedures). Untreated lung preparations, incubated in pairs for three consecutive hours showed no significant change in fluid production rates (Fig 14 Ai , Aii). I Expanded preparations Of twelve further preparations incubated in pairs, one preparation in each pair was expanded by approximately 70% at the end of the first hour; the other left unexpanded. All expanded lungs decreased fluid production; three turned to reabsorption. Overall, fluid production immediately after expansion fell by 125.4 ± 42.6% (Fig 14 Bi). This fall was significant by A N O V A (P < 0.01). At the same time, all six unexpanded companion lungs reduced fluid production by an average of 88.2 ± 10.7% in the middle hour (Fig 14 Bii). The reduction in fluid production in the middle hour was significant by A N O V A (P < 0.01). Of twelve lung preparations, incubated in pairs and treated with 10"6 M mianserin, one preparation in each pair was expanded by 73% (72.9 ± 3.2%) at the end of one hour. All expanded preparations decreased fluid production and three turned to reabsorption in the middle hour. The fluid production rates in successive hours were: before expansion: 2.46 ± 0.78; after expansion: -0.79 ± 0.56, and 0.83 ± 0.51 (Fig 14 Ci). The 104 fall in fluid production by expanded lungs was 132.0 ± 39.3% and was significant by A N O V A (P < 0.01); in comparison, lung preparations expanded similarly, but incubated without mianserin in the outer saline, reduced fluid production by 125%) (Fig 14 Bi). There was no significant difference between the two groups; therefore, mianserin did not affect the percent falls in fluid production by expanded lungs. II Unexpanded companion preparations All unexpanded companion preparations incubated in the presence of mianserin decreased fluid production; one reabsorbed fluid. The overall fluid production rates were: first hour: 1.34 ± 0.15; in the presence of 10'^ M mianserin: 0.46 ± 0.25 and final hour: 0.82 ± 0.38 ml/kg body weight per h (Fig 14 Cii). The fall in fluid production was 65.4 ± 14.3 % and was significant by regression analysis. By comparison, companion lung preparations incubated in the absence of mianserin reduced fluid production by 88.2% (Fig 14 Bii). No significant difference was found between the two groups (one-way ANOVA). It was concluded that histamine was not essential to responses of either expanded or unexpanded lung preparations. Preparations incubated singly in saline containing 10"6 M mianserin showed no significant reduction in fluid production in the hours following exposure (Fig 13 D). Fluid production rates in successive hours were: before treatment: 1.07 ± 0.09; during treatment: 0.76 ± 0 . 1 0 and after treatment: 0.58 ± 0.09 ml/kg body weight per hour. 105 Fig 14: Influence of expanded lungs on lung liquid production by unexpanded companion lungs supported in the same saline in vitro; studies in the presence of mianserin". Based on 43 fetuses of 63 ± 2 days of gestation and 106.4 ± 19.3 g (SD) body weight. (A) Twelve untreated (control) preparations incubated in pairs (B) 12 experimental lungs taken from the same mother and incubated in pairs, one was expanded by 70 % with saline at the end of the first hour (Bi); the companion preparation was left unexpanded (Bii). (C) Six pairs of lung preparations incubated together in the same saline and transferred after 1 h to saline containing 10"6 M mianserin: (Ci) one preparation from each pair was then expanded 71% with saline; (Cii) the second was left unexpanded. (D) Seven lung preparations incubated singly in 10"6 M mianserin. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (Ai) 0.68 ± 0.08 (SD), (Aii) 0.78 ± 0.23, (Bi) 1.09 ± 0.30, (Bii) 0.68 ± 0 . 0 8 , (Ci) 0.78 ± 0.23, (Cii) 1.09 ± 0.30, and (D) 1.00 ± 0.20 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 106 130 120 110 100 90 80 120 110 100 90 80 110 100 90 80 110 Ai) lung 1: Control n = 6 1.31 ± 0.12 Aii) lung 2: Control n = 6 Joint incubations Bi) lung 1: 70% Expansion n = 6 *m • ' ' * Bii) lung 2: No expansion. n = 6 100 -90 -120 r 110 100 -90 -80 -120 110 100 90 0.12 ± 0.09 0.98 • ± 0 . 1 4 Ci) lung 1: 73 % Expansion/10"6 M Mianserin n = 6 0.29 ± 0.20 Cii) lung 2: No expansion/10" M_Mianserin n = 6 * ' . 0.82 ± 0.38 80 120 110 100 90 80 D) 10 M Mianserin n = 7 0.56 ± 0.09 S A L I N E 1 1 T R E A T M E N T 12 S A L I N E Time (Hours) Joint incubations Joint incubations 107 6.4 Discussion These results demonstrate that agents known to increase intracellular calcium concentration can reduce fetal lung liquid production, and that their effects are relatively long lasting. The calcium ionophore (A23187) caused strong reductions in fluid production, or produced fluid reabsorption; these effects were long lasting. This could mean that (1) the ionophore remained in the tissue and exerted its influence on fluid production, or (2) it triggered intracellular events that continued long after initial stimulation. A23187 can in fact remain in cells for long periods (R. Harris, personal communication). It also increases cytosolic Ca 2 + in cultured rat alveolar type II cells with no increase in intracellular cyclic AMP (Dobbs, Gonzalez, Marinari, Mescher and Hagwood, 1986). We might expect A23187 to sustain a high concentration of cytosolic free Ca 2 + and exert its prolonged effect on fluid production through calcium-dependent transport mechanisms. Wirtz and Dobbs (1990) demonstrated that mechanical distention of cultured rat alveolar type II cells mobilizes cytosolic Ca 2 + only briefly, but this mobilization is accompanied by a sustained secretion of pulmonary surfactant. They concluded that an elevation of cytosolic free Ca 2 + initiates complex, long lasting cellular events that outlive the Ca 2 + stimulus. We may reasonably expect prolonged intracellular processes following an initial increase in cytosolic free Ca 2 + after exposure to A23187. The possible persistence of A23187 in tissue, and the possibility of 108 prolonged Ca2+-dependent intracellular events could explain the strong effect on fluid production in the final hour. Calcium-dependent ion transport does occur in fetal lung epithelial cells. The elevation of intracellular calcium concentration ([Ca2+]j) increases whole cell ion conductances in cultured rat fetal distal lung epithelial cells (Nakahari and Marunaka, 1995). These authors proposed that this elevation of [Ca2+]j increases K + and CI" conductances which in turn activate nonselective cation channels that play an important role in Na + absorption and water clearance in the perinatal lung air space. The occurrence of nonselective cation channels has been demonstrated in rat fetal distal lung epithelial cells (Orser, Bertlik, Fedorko and O'Brodovich, 1991; Marunaka, Tohda, Nagiwara and O'Brodovich, 1992) and could be widespread. These channels are apparently inactive in nominally Ca 2 + free solutions and are activated when cytosolic free Ca 2 + concentrations increase to the mM range. Wirtz and Dobbs (1990) measured [Ca2+]j after a single stretch of cultured alveolar type II cells from adult rats and determined it to be in the pM range. Due to technical limitations, the measurements were taken only after relaxation was complete. Still, the measured [Ca2+] ( was 3.5 times higher after stretch. These concentrations must be regarded as conservative since intracellular Ca 2 + release patterns in airway epithelial cells are often biphasic. Intracellular Ca 2 + release in cultured human tracheal and nasal epithelial cells may involve a sharp initial spike followed by a smaller, relatively longer lasting release phase(Harris et al, 1991; Harris and Hanrahan, 1993). This suggests the release of Ca 2 + from multiple pathways. If Ca 2 + release patterns in these cells are similar, Wirtz and Dobbs could 109 have missed the initial spike of cytosolic Ca 2 + release. Sustained or rythmic expansion could maintain high [Ca2+]j presumably through the slower Ca 2 + release pathway; this may produce long lasting cellular responses. Our in vitro fetal lung preparations reduce fluid production or reabsorb fluid in response to saline expansion, and often maintain reabsorption for two hours after expansion. The preparations lack a rigid rib cage, therefore, their compliance is high. This allows a greater distribution of pressure to the surrounding medium and reduces intraluminal pressure. 70% expansion with saline resulted in a small (0.5-1.0 cm H 2 0) and transient increase in intraluminal pressure (Garrad Nelson and Perks, 1996a). The amount of stretch of the epithelial cells in response to liquid expansion is not known; however, even in the absence of any significance increase in intraluminal pressure, it must be sufficient to influence fluid production. Expansion of lungs with fluid by up to 70% does not damage lungs yet fluid reabsorption induced by lung expansion is not fully inhibited with amiloride (Garrad Nelson and Perks, 1996a). We have concluded that transport processes triggered by expansion probably include amiloride-insensitive Na + transport (see Chapter 7). Marunaka and his co-workers have demonstrated that amiloride has no effect on open probability of nonselective Na + channels when [Ca2+]j is in the mM range (Marunaka et al., 1992). Thus a sufficient increase in intracellular Ca 2 + could cause amiloride-insensitive Na + transport. If expansion sufficiently increases Ca 2 + , it could explain the failure of amiloride to stop expansion-induced fluid reabsorption. 110 The Ca 2 + secretagogues, bradykinin and histamine produced smaller but, like A23187, long lasting effects. In all three cases there was evidence of a prolonged reduction in fluid production; there was no recovery in the third hour. On the other hand, lung preparations treated with analogues of cyclic AMP, or with hormones that exert their effects by stimulating cAMP synthesis, do have a tendency to recover fluid production (Kindler, Ziabakhsh and Perks, 1992; Perks, Kindler, Marshall, Woods, Craddock and VonderMuhll, 1993; Woods et al., 1997). Both bradykinin and histamine are known to elevate intracellular calcium levels in cultured human airway epithelial cells (Harris et al., 1991; Harris and Hanrahan, 1993; 1994). Sustained stimulation of these cells with bradykinin elevated cytosolic free Ca 2 + , but for less than 3 min, yet the effect of bradykinin treatment on fluid production in our lung preparations lasted more than 1 h. In similar fashion, bradykinin treatment could have elevated cytosolic free Ca 2 + in our preparations. If so, the intracellular events following the initial increase in cytosolic free Ca 2 + would have to be long lasting. As mentioned earlier, mechanical distention of cultured rat alveolar type II cells, mobilizes cytosolic Ca 2 + briefly, but this mobilization is accompanied by a prolonged secretion of pulmonary surfactant (Wirtz and Dobbs, 1990). Prolonged effects, as seen following treatment with bradykinin, could mean the initiation of complex cellular events that outlive the Ca 2 + stimulus. The overall effect mirrored the lung liquid response to A23187. Histamine decreased fluid production and sustained the decrease for at least 2 hours, possibly by increasing pulmonary epithelial permeability through H, and H 2 receptor stimulation as it does in adult human lungs (Braude, Coe Royston and Barnes, 1984; Chan, Eiser, Shelton and Rees, Ill 1987). Both H, and H 2 histamine receptors are known to occur in guinea pig lungs (Chand and DeRoth, 1979). Histamine also increases cytosolic free Ca 2 + through H, receptor stimulation (Harris and Hanrahan, 1993), and is released following treatment of human lung and nasal fragments with A23187 (Ott, Lohse, Klotz, Vogt-Moykopf and Scwabe, 1992; Austin, Dear, Neighbour, Lund and Foreman, 1996). In the present experiments, a relatively high histamine concentration (10"6 M) reduced fluid production only modestly, but the effect was sustained. This concentration of histamine was not unusually high, as the concentration that elevates cytosolic free Ca 2 + in cultured human tracheal epithelial cells is at least 10"6 M , and the half-maximal concentration of free Ca 2 + is only achieved with 1.2 x 10"4 M histamine (Harris and Hanrahan, 1993). Thus histamine appears to be a fairly weak Ca 2 + secretagogue, and it seems unlikely that A23187 would exert its huge effects through histamine unless tissue release of histamine following treatment with A23187 is very strong. The data shows that histamine could produce prolonged and significant reduction in lung liquid production. However, in the presence of mianserin, which antagonizes H[ and H 2 histamine receptors (Leitch, Boura and King, 1992), both expanded and unexpanded companion lung preparations responded, as usual, with a significant reduction in fluid production. This suggested that the full responses of the expanded and companion lungs were not based on LL, receptor stimulation, therefore if expanded lungs liberated histamine, it could only cause partial responses in either preparation. Chapter VII Transport mechanisms related to catecholamine stimulation and lung expansion: Effects of ion transport blockers and intraluminal glucose concentration. 112 113 7.1 Introduction The clearance of fetal lung liquid is a critical event in the transition from intrauterine to extrauterine life. Studies with amiloride have suggested that the removal of lung liquid at the time of birth involves the activation of a Na+-based reabsorptive mechanism (Olver, Ramsden and Strang, 1981; Olver, Ramsden, Strang and Walters, 1986; O'Brodovich, Hannan, Seear and Mullen, 1990; Bland, 1991; Cassin and Perks, 1993). This reabsorption can be stimulated by hormones such as epinephrine (Olver et al., 1981) or by expansion (O'Brodovich et al., 1990; Song, Sun, Curstedt, Grossman and Robertson, 1992; Garrad-Nelson and Perks, 1996a). In addition, arginine vasopressin (AVP) can reduce lung liquid secretion through a mechanism that is completely blocked with amiloride (Perks, Kindler, Marshall, Woods, Craddock and Vonder Muhll, 1993). However, amiloride does not completely block the reabsorption associated with the first breath (O'Brodovich et al., 1990; Song et al., 1992) suggesting the involvement of another transport mechanism. There is some evidence from studies of isolated lung epithelial cells from fetal rats that fetal lung liquid can be reduced by mechanisms involving amiloride-insensitive Na + transport (Matalon, Bridges and Benos, 1991; Matalon, Bauer, Benos, Kleyman, Lin, Cragoe Jr., and O'Brodovich, 1993). The possibility of fluid clearance through Na+-glucose cotransport is doubtful because (1) glucose concentration in fetal lung liquid is low (Barker, Boyd, Ramsden, Strang and Walters, 1989) and (2) phloridzin has no effect on Na + uptake or liquid clearance (O'Brodovich, Hannam 114 and Rafii, 1991, Smedira, Gates, Hastings, Jayr, Sakuma, Pittet and Matthay, 1991; Russo, Lubman and Crandall, 1992). The possibility of a reversal of active CI" secretion has been raised. The gastric peptide, somatostatin, reduces lung liquid production (Perks, Kwok, Mcintosh, Ruiz and Kindler, 1992); this reduction is not blocked with amiloride (Perks and Kindler, unpublished observations). Somatostatin presumably inhibits CI" secretion as it does in the mammalian intestine and stomach (Carter, Biter, Zfass and Makhlouf, 1978; Dobbins, Dharmsathphorn, Racusen and Binder, 1981). Perks and coworkers also showed that neonatal guinea pig lungs released somatostatin following the first few breaths; somatostatin-like immunoreactivity was still detectable in lungs after 30 min of postnatal life. In recent studies using in vitro lungs from fetal guinea pigs, Garrad-Nelson and Perks (1996a) have demonstrated that expanded fetal lungs reduce fluid production, and release an unknown agent that can inhibit liquid production in another preparation. In the present study, I use the Na + channel blocker, amiloride, and CI' channel blocker, 5-nitro-2-(3-phenylpropylamino) benzoic acid (NPPB), to (1) investigate the ion transport mechanisms involved in fluid reabsorption by expanded lungs, and (2) identify transport mechanisms that reduce fluid production by unexpanded lungs stimulated by the unknown agent thought to be released by expanded lungs. 115 7.2 Materials and Methods 7.2.1 Animals Pregnant albino guinea pigs of an inbred departmental stock were given food and water ad libitum (guinea pig chow, Ralston-Purina, supplemented with fresh vegetables and vitamin C). These experiments were carried out on 67 guinea pig fetuses of 61 ± 1 days (SD) of gestation and 92.2 ± 20.8 g (SD) body weight. 7.2.2 Studies of lung liquid production Removal of the lungs from the fetuses and preparation of in vitro lungs occurred as previously described. In order to test the hypothesis that amiloride-sensitive Na + transport is involved in liquid reabsorption by expanded lungs, lung preparations were expanded as before but with Krebs-Henseleit saline containing amiloride in both the outer saline and luminal lung liquid. Final concentrations were 10"6 M or 10"5 M after expansion. Amiloride hydrochloride was obtained from Sigma Chemicals (St. Louis, MO). Other preparations were incubated in pairs; one preparation in each pair was expanded and the second was left unexpanded. The aim of these experiments was to test the role of amiloride-sensitive Na + transport in fluid production by the companion preparation. Each pair was transferred to saline containing 10"6 M amiloride before one lung preparation from each pair was expanded by approximately 70% with saline to 116 give a final luminal concentration of 10"6 M amiloride. Amiloride was placed in the bathing saline in order to stop the loss, by diffusion, of intraluminal amiloride. The chloride channel blocker, 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB, RBI, Natick, MA) was used in studies to investigate the role of chloride channels in the reversal of lung liquid production seen following treatment with catecholamines or expansion. NPPB was added directly into the lungs to give a final intraluminal concentration of 10"5 M after expansion. The concentration was based on the work of Timann, Kunzelmann, Krobe, Cabantchik, Lang, Englert and Greger (1991) which indicated that open probability of CI" channels was zero at 10"5 M . To test whether the reabsorption by expanded lungs was due to glucose-linked sodium transport, isotonic NaCl (0.15 mM) was used as the instillate for lung expansion instead of Krebs-Henseleit saline (glucose concentration =11.1 mM); since Krebs-Henseleit saline increased lung glucose concentration, while NaCl solution reduced it, comparison of effects between the two expansion fluids was expected to indicate effects of glucose. Amiloride was dissolved in isotonic NaCl and instilled into lungs during expansion; final concentration in lungs was 10"5 M . Fetal lung liquid and plasma concentrations of glucose were determined (see below). 117 7.2.3 Glucose concentrations in fetal plasma and lung liquid I Sample collection Blood and lung liquid samples for the determination of glucose concentration were obtained from fetuses immediately after their delivery by Cesarian section. After delivery, the fetuses were transferred to Krebs-Henseleit saline; no fetal breathing movements were seen. The fetal trachea was cannulated as previously described, but the lungs were not removed from the fetal thorax. The lung liquid was completely withdrawn into a glass syringe (Hamilton Co., Reno, NV), transferred into a test tube (1 ml Eppendorf, Brinkmann Instruments, Rexdale, ON) and frozen at -20°C until analyzed for glucose. The fetal thorax was quickly opened by a midline incision, and a 1 ml blood sample was drawn directly from the ventricle using a heparinized syringe (Becton Dickinson and Co., Rutherford NJ). The blood was centrifuged (Eppendorf Model 3200) for 10 min and the plasma was decanted and frozen until analyzed for glucose. 118 II Glucose Assay The assay was a modification of the method of Kunst, Draeger and Ziegenhorn (1983), and was based on the following reactions: hexokinase, Mg a) Glucose + ATP -> ADP + Glucose-6-phosphate glucose-6-phosphate dehydrogenase b) Glucose-6-phosphate - » 6-phosphogluconolactone + NADPH + FT Plasma samples were deproteinized with 17.5 % perchloric acid (PCA) and the supernatant neutralized with 2M K 2 C 0 3 . Triplicate samples of deproteinized plasma (20 pl + 20 pl distilled water) and lung liquid (40 pl) were mixed in microplate wells with 200 pl cocktail containing Tris salt (60 mM), Tris-HCl (40 mM), MgS0 4 .7H 2 0 (1.0 mM), N A D + (2.0 mM), ATP (1.0 mM) and glucose-6-phosphate dehydrogenase (2,500 U/ml). Finally 10 pl hexokinase (300 rnU/ml) in assay buffer (pH 7.5) consisting of 5.57 g/100 ml triethanolamine and 0.0998 g/100 ml MgS0 4 .7H 2 0 were added to the mixture and the absorbance read immediately on a microplate reader (THERMOmax™, Molecular Devices, Sunnyvale, CA) and at 5 min intervals until the reaction was complete. Sample absorbances were internally converted to glucose concentration on a SOFTmax Data Module (Molecular Devices, Sunnyville, CA). Results are reported as mM (± SEM). 119 7.3 Results 7.3.1 Effects of amiloride on fluid responses to lung expansion I Joint incubations Studies to demonstrate the effects of amiloride on the response to fetal lung expansion were done on lung preparations incubated in pairs. Twelve untreated control preparations, incubated in pairs, produced fluid for three consecutive hours without any significant change (Fig 15 Ai , Aii). a) Expanded preparations Twelve other preparations were incubated in pairs; one preparation in each pair was expanded by approximately 70% at the end of the first hour, and the other left unexpanded. All expanded lungs decreased fluid production; three turned to reabsorption. Overall, fluid production immediately after expansion fell by 125.4 ± 42.6% (Fig 15 Bi). This fall was significant by A N O V A (P < 0.01). Six preparations were incubated in pairs as above. Three preparations were expanded as before, but were given intraluminal amiloride estimated to give a final concentration of 10"6 M ; the other three preparations were left unexpanded but were similarly instilled with amiloride. All expanded lungs reduced fluid production in the middle hour, and two turned to 120 reabsorption. The fluid production rates in successive hours were: before expansion, 1.58 ± 0.11; after expansion: -0.44 ± 0.42, and -0.47 ± 0.22 ml/kg body weight per hour (Fig 15 Ci). Fluid production fell by 128.1 ± 23.5% in the hour after expansion; this fall was significant (P < 0.01, ANOVA), and was not different from the percent fall in production by expanded lung preparations, similarly incubated, but with no amiloride in the outer saline (Fig 15 Bi). b) Unexpanded companion preparations In the absence of amiloride, six unexpanded companion lungs reduced fluid production by an average of 88.2 ± 10.7%> in the middle hour (Fig 15 Bii). The reduction in fluid production in the middle hour was significant by A N O V A (P < 0.01). By comparison, companion lung preparations incubated in the presence of 10"6 M amiloride did not decrease fluid production significantly. The fluid production rates in successive hours were: 1.49 ± 0.42, in the presence of lO'6 Mamiloride, 1.05 ± 0.64, and 1.03 ± 0.89 ml/kg body weight per h (Fig 15 Cii). The change in fluid production during the middle hour was a 29.9% fall. However, there was no significant difference in the percent fall in fluid production in the presence, and in the absence of amiloride (one-way ANOVA). Six preparations incubated singly, with 10"6 M amiloride in the lungs and outer saline, produced lung liquid steadily for three successive hours with no significant change in rate (Fig 15 D). These results suggest that amiloride could slow effects in the companion lung preparations; however, since the number of experiments done was small, these results must be taken with caution. 121 7.3.2 Is there significant glucose in guinea pig fetal lungs? Glucose concentrations in fetal plasma and lung liquid The possibility of fluid movement by glucose-linked Na + transport has been suggested but eliminated in studies of fetal sheep lungs. To ascertain that glucose transport has no influence in fluid movement by fetal guinea pig lungs, glucose concentrations in fetal lung liquid and plasma were determined. These studies were based on 6 fetuses of 113 ± 17.8 g body weight (SD) and 63 + 1 days of gestation. Glucose was present in both fetal plasma and lung liquid in all six preparations. Plasma glucose concentration was 1.35 ± 0.11 mM; lung liquid glucose concentration was 0.19 ± 0.03 mM. By comparison, Krebs-Henseleit saline contains 11.1 mM glucose resulting in a final intraluminal glucose concentration of approximately 4.6 mM. There was no correlation between plasma glucose concentration and lung liquid glucose concentration. Lung expansion with Krebs-Henseleit saline raised glucose concentration in the lungs; therefore, it introduced the possibility of glucose-linked sodium transport. To test the effects of lung expansion with slightly lowered glucose concentration, I expanded lungs using 0.9% NaCl (isotonic saline). 122 Fig 15: Influence of intraluminal amiloride on lung liquid production by expanded lungs and unexpanded companion lungs supported in the same saline in vitro. Based on 36 fetuses of 63 ± 2 days of gestation and 106.4 ± 19.3 g (SD) body weight. (A) Twelve untreated control preparations incubated in pairs. (B) 12 experimental lungs taken from the same mother and incubated in pairs; one was expanded by 70 % with saline at the end of the first hour (Bi); the companion preparation was left unexpanded (Bii). (C) Three pairs of lung preparations incubated together in the same saline and instilled with 10"6 M amiloride at the end of the first hour. (Ci) one preparation from each pair was then expanded 69 % with saline; (Cii) the second was left unexpanded. (D) Six lung preparations incubated singly, and instilled with 10"6 M amiloride at the end of the first hour. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (Ai) 0.68 ± 0.08 (SD), (Aii) 0.78 ± 0.23, (Bi) 1.09 ± 0.30, (Bii) 0.68 ± 0.08, (Ci) 0.78 ± 0.23, (Cii) 1.09 ± 0.30, and (D) 1.00 ± 0.20 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 123 130 120 10 100 90 80 120 10 100 90 80 10 100 -90 -A i ) lung 1: Control n = 6 A i i ) lung 2: Control n = 6 B i ) lung 1 : 70% Expansion n = 6 ' " " * -m 9 • u -0.28 ± 0.37 80 10 100 90 80 120 r B i i ) lung 2: N o expansion n = 6 0.12 ± 0.09 C i ) lung 1: 69% Expansion/10" 6 M Ami lo r ide n = 3 10 -100 -90 -80 -120 r 10 -100 -90 -80 0 C i i ) lung 2: N o Expansion/10" M Ami lo r ide n = 3 D) 10"" M A m i l o r i d e n = 6 Joint incubations -0.41 ± 0.23 0.29 ± 0.20 Joint incubations 0.47 ± 0.22 Joint incubations S A L I N E T R E A T M E N T S A L I N E Time (Hours) 124 7.3.3 Effects of lung expansion with 0.9% NaCl Six untreated lung preparations produced fluid steadily for three successive hours with no change in fluid production (Fig 16a). Another six preparations were expanded by approximately 70% (68.2 ± 0.7%)) with 0.9% NaCl in order to minimize effects due to possible glucose transport. 70% lung expansion with NaCl was expected to reduce intraluminal glucose concentration from 0.19 to 0.11 mM, a 42% reduction (70% expansion with Krebs-Henseleit saline raised intraluminal glucose concentration 24-fold). All lungs reduced fluid production; three turned to reabsorption. Fluid production rates in successive hours were: before expansion: 1.1 ± 0.2, after expansion: -0.24 ± 0.25, and 0.27 ± 0 . 1 4 ml/kg per h (Fig 16b). The overall fall was 121.7 ± 28.7%>; this fall was significant (P < 0.001, ANOVA). By comparison, lung preparations from fetuses of comparable size, similarly incubated, but expanded using Krebs-Henseleit saline reduced fluid production by 170.1 ± 2 1 . 1 % (Fig 16c). There was no significant difference between these two effects; this suggested that there was no significant glucose-linked Na + transport in lung preparations expanded with Krebs-Henseleit saline. 7.3.4 Effects of raising amiloride, and lowering glucose concentrations in the lungs Since 10"6 M amiloride appeared ineffective in preventing responses to expansion, the concentration was increased to 10"5 M . Amiloride was prepared in 0.9% NaCl solution to avoid 125 introducing any glucose into the lungs. Six untreated lung preparations produced fluid steadily with no significant change in rate (Fig 17a). Another six preparations treated with amiloride alone produced fluid for three successive hours with no significant change in rate (Fig 17b). Six further preparations were expanded by 70% (68.2 ± 0.7%) with 0.9% NaCl in order to eliminate any effects due to additional glucose in the intraluminal space, and treated with amiloride to give a final concentration of 10'5 M intraluminally. All lungs reduced fluid production; five turned to reabsorption. The rates in successive hours were: before expansion, 1.26 ± 0.36, after expansion, -0.64 ± 0 . 3 1 , and 0.35 ± 0.20 ml/kg body weight per hour (Fig 17c). The overall fall in fluid production was 151.0 ± 29.7%; this fall was significant by A N O V A (P < 0.005. By comparison, lung preparations expanded with 0.9% NaCl, but with no amiloride in the intraluminal space, reduced fluid production by 121.7 ± 28.7%> (Fig 17d). There was no significant difference between the two treatments, suggesting that the fall in lung liquid production was not due to amiloride-sensitive Na + transport. In addition, no significant difference was found between the effects of amiloride in lungs expanded with Krebs-Henseleit saline (170.1 ± 2 1 . 1 % fall) and preparations expanded with 0.9% NaCl (151 ± 29.7% fall). This means that the fall in fluid production by lung preparations expanded with Krebs-Henseleit saline, was not the result of glucose transport made possible by additional glucose introduced into lungs through the saline itself. I conclude that neither the use of glucose-free expansion fluid nor the presence of a relatively high concentration of intraluminal amiloride affected the response of the fetal lungs to expansion. 126 Fig 16: Comparison of effects of expansion with Krebs-Henseleit saline and 0.9% NaCl on lung liquid production by in vitro lungs from near-term fetal guinea pigs. Based on 18 fetuses of 61 ± 1 (SD) days of gestation and 85.8 ± 1 9 . 1 (SD) g body weight. At the end of the first hour, the preparations were immersed in fresh saline and treated as follows: (a) left unexpanded (untreated controls), (b) 69% expansion with 0.9% NaCl and (c) 70% expansion with Krebs-Henseleit saline. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (a) 0.75 ± 0.15 (SD), (b) 1.07 ± 0.21, and (c) 0.74 ± 0.14 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 127 120 r 110 100 90 80 120 -110 -a) Control n = 6 1.14 ± 0 . 1 4 b) 69% Expansion with 0.9% NaCl n = 6 1.08 ± 0 . 2 4 100 90 -0.24 ± 0 . 2 5 -0.27 ± 0 . 1 4 80 120 110 h 100 90 c) 70%) Expansion with Krebs-Henseleit saline n = 6 0.28 ± 0 . 3 0 80 0 SALINE TREATMENT SALINE Time (Hours) 128 Fig 17: Effects of amiloride on lung liquid production by in vitro lungs expanded with 0.9% NaCl. Based on 25 guinea pig fetuses of 61 ± 1 (SD) days of gestation and 85.8 ± 19.1 (SD) g body weight. At the end of the first hour, the preparations were immersed in fresh saline and treated as follows: (a) saline alone (untreated controls), (b) 10"5 M amiloride in 0.9% NaCl (c) 68% expansion with 0.9%> NaCl containing 10"5 M amiloride, and (d) 69% expansion with 0.9%) NaCl. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (a) 0.75 ± 0.15 (SD), (b) 1.07 + 0.21, (c) 0.74 + 0.14, and (d) 1.00 ± 0.20 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 129 O o H N O 120 110 100 90 80 120 110 100 90 80 120 110 100 90 80 120 110 100 90 80 a) Control n = 6 b) 1(T M Amiloride in 0.9% NaCl n = 6 c) 68% Expansion / 10 M Amiloride (0.9% NaCn n = 6 . . -d) 69% Expansion with 0.9% NaCl n = 7 -0.24 ± 0 . 2 5 0.27 ± 0 . 1 4 0 SALINE TREATMENT SALINE Time (Hours) 130 7.3.5 Studies with catecholamines Treatment of expanded lungs with amiloride did not reduce fluid reabsorption, therefore lungs were treated with intraluminal amiloride and incubated in the presence of epinephrine (2.2 x 10"6 M) or norepinephrine (10"5 M). These agents are known to cause significant reductions in fluid production, or reabsorption in the guinea pig (see Chapter 4) and the sheep fetus; fluid reabsorption in the latter is known to involve amiloride-sensitive Na + transport (Olver et al., 1986). The aim of these experiments was to test the efficacy of amiloride used in the expansion experiments, given above. I Basic effects Untreated control preparations produced fluid for three hours with no significant change in rates (Fig 18a). Likewise, six lung preparations, treated with amiloride alone, produced fluid steadily for three hours with no significant change in rate (Fig 18b). Another six preparations were treated with epinephrine (2.2 x 10"6 M) in the outer saline. All preparations reduced fluid production; four turned to reabsorption. The fluid production rates in successive hours were: before treatment, 1.33 ± 0.28; during treatment, 0.62 + 0.51; after treatment, 0.61 ± 0.44 ml/kg body weight per h (Fig 18c). The overall fall in fluid production was 146.2 ± 23.7% and was significant by A N O V A (P < 0.05). In similar experiments, six lungs were treated with 10"5 M norepinephrine placed in the outer saline. All lungs reduced fluid production, with four turning to 131 reabsorption. The fluid production rates were: before treatment: 1.24 ± 0.24; during treatment: -0.46 ± 0.48; after treatment: 0.32 ± 0.27 ml/kg body weight per h (Fig 18e). The fall in fluid production in the middle hour (137.3 ± 21.7%) was significant (P < 0.001, ANOVA). This fall was not significantly different from the fall in preparations treated with epinephrine. II Effects of amiloride Six lung preparations were treated simultaneously with 2.2 x 10"6 M epinephrine (in the bathing saline) and 10"5 M amiloride (placed in the bathing saline and intraluminally). All lungs reduced fluid production; five turned to reabsorption. The fluid production rates were: before treatment, 1.21 ± 0.33; during treatment: -0.66 ± 0.17; after treatment: 0.70 ± 0.28 ml/kg body weight per h (Fig 18d). The overall fall in fluid production in the middle hour was 155.0 ± 29.1%; this fall was significant (P < 0.001, ANOVA). There was no significant difference between this fall and the fall in fluid production by lung preparations treated with epinephrine, but with no intraluminal amiloride. Another six preparations were similarly treated with 10"5 M norepinephrine and 10"5 M amiloride. All preparations reduced fluid production, with four turning to reabsorption. The fluid production rates were: before treatment: 1.30 ± 0 . 1 3 ; during treatment: -0.23 ± 0.22; after treatment: 0.21 ± 0 . 1 8 ml/kg body weight per h (Fig 18f). The overall fall in fluid production in the middle hour was 117.8 ± 14.1%; this fall was significant (P < 0.001, ANOVA). However, there was no significant difference between this fall and the fall in fluid production by lung preparations treated with norepinephrine, but with no intraluminal 132 amiloride. These observations suggest that either (1) catecholamines stimulate lung liquid reabsorption through a mechanism that is not blockable with amiloride or (2) the concentration of amiloride required to eliminate effects of catecholamines in these lung preparations is greater than 10'5 M . However, 10'6 M amiloride does block effects of other agents on similar preparations (Perks, Kindler, Marshall, Woods, Craddock and Vonder Muhll, 1993a; Perks, Ruiz, Chua, Vonder Muhll, Kindler and Blair, 1993b). Ill Effects of the CI transport inhibitor, NPPB To test whether the lung liquid reabsorption induced by norepinephrine might be based on epithelial chloride transport rather than Na + transport, 5-nitro-2-(3-phenylpropylamino) benzoic acid (NPPB), a potent blocker of intermediate-conductance outwardly rectifying chloride channels, was used simultaneously with norepinephrine. Untreated preparations produced fluid for three successive hours with no significant difference in rates (Fig 19a). Six lung preparations were treated with NPPB, estimated to give a final intraluminal concentration of 10"5 M ; at this concentration, apical CI" channel inhibition is complete (Timann et al., 1991). NPPB alone did not decrease fluid production significantly. Fluid production rates in successive hours were: before treatment: 1.01 ± 0.09; with intraluminal NPPB: 0.50 + 0.12 and 0.55 ± 0.34 ml/kg body weight per h (Fig 19b). The overall fall was by 50% (50.3 ± 13.5%), but this fall was not statistically significant; therefore, despite some indication that CI" channels are involved in fluid production, this could not be clearly demonstrated. Four preparations treated simultaneously with 133 10"5 M norepinephrine in the outer saline, and 10'5 M NPPB applied directly into the lungs, reduced fluid production; two turned to reabsorption. The fluid production rates in successive hours were: before treatment, 1.35 ± 0.19, during treatment, -0.16 ± 0.20, and after treatment, 0.38 ± 0.11 ml/kg body weight per hour (Fig 19c). The overall fall in fluid production in the middle hour was 111.5 ± 17.2%; this fall was judged significant (P < 0.05, ANOVA). By comparison, preparations treated with norepinephrine, but with no intraluminal NPPB, decreased fluid production by 137.3 ± 21.7% (Fig 19d). There was no significant difference between the effects of norepinephrine used alone, and norepinephrine used simultaneously with intraluminal NPPB. The results suggest that CI" channel blockade with NPPB could not inhibit the fall in fluid production caused by stimulation with norepinephrine. 134 Fig 18: Effect of amiloride on lung liquid production by in vitro lungs treated with epinephrine and norepinephrine. Based on 36 guinea pig fetuses of 61 ± 1 (SD) days of gestation and 85.8 ± 19.1 (SD) g body weight. During the middle hour, the preparations were immersed in saline containing one of the following (a) saline alone (untreated controls), (b) 10"6 M amiloride, (c) 2.2 x 10"6 M epinephrine, (d) 2.2 x 10"6 M epinephrine in presence of 10"6 M amiloride, (e) 10"5 M norepinephrine, and (d) 10"5 M norepinephrine in the presence of 10"6 M amiloride. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (a) 0.75 ± 0.15 (SD), (b) 1.02 ± 0.23, (c) 0.83 ± 0.17, (d) 0.77 ± 0.21, (e) 1.01 ± 0.21, and (f) 0.76 ± 0.09 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 135 s O > 13 o H O N 120 r 110 -100 -90 -80 -120 -110 -100 -90 -80 -120 r a) Control n = 6 1.14 ± 0.14 1.08 ± 0.21 b) 10'6 M Amiloride n = 6 l . n ± 0.19 0.98 ± 0.18 c ) 2.2 x 10"6 M Epinephrine n = 6 d) 2.2 x 10 M E p i n e p h r i n e / 1 0 ° M Amiloride n = 6 ^ 110 100 90 I-80 L 120 r e) j_p_" M Norepinephrine n = 6 f) 10"5 M Norepinephrine/10 0 M Amiloride •0.46 ± 0 . 4 8 -5 0.32 ± 0 . 2 7 n = 6 -0.23 + 0.22 0.21 + 0.18 1 | TREATMENT | 2 I SALINE Time (Hours) 136 Fig 19: Effects of intraluminal application of the chloride channel blocker, 5-nitro-2-(3-phenylpropylamino) benzoic acid (NPPB) on lung liquid production by in vitro lungs treated with norepinephrine. Based on 22 fetuses of 61 ± 1 (SD) days of gestation and 85.8 ± 19.1 (SD) g body weight. At the end of the first hour, the preparations were immersed in fresh saline and treated as follows: (a) saline alone (untreated controls), (b) 10"5 M NPPB alone (c) 10"5 M norepinephrine in the presence of 10"5 M NPPB, and (d) 10"5 M norepinephrine alone. Ordinates: total volume of lung liquid expressed as a percentage of that present at the end of the first hour where 100% was (a) 0.75 ± 0 . 1 5 (SD), (b) 1.07 ± 0 . 2 1 , (c) 0.74 ± 0 . 1 4 , and (d) 1.00 ± 0.20 ml. All regressions are lines of best fit; the slopes represent lung liquid production rates; rates in ml/kg body weight per h (±SEM) are given below the lines. Asterisks above the lines show significant changes from the original slope (dotted lines). Significance is accepted at P < 0.05. Standard errors of individual points are omitted for clarity. 1 3 7 a > o H 120 110 100 90 80 120 110 100 90 80 120 110 100 90 80 120 110 100 90 80 a) Control n = 6 b) 10'5 M NPPB (intraluminal) n = 6 c) 10" M Norepinephrine/10 M NPPB (intraluminal) n = 4 . . - - ' " * « • * • m --0.16 ± 0 . 2 0 0.38 ± 0 . 1 1 d) 10"5 M Norepinephrine n = 6 0 SALINE 1 TREATMENT 2 SALINE 3 Time (Hours) 138 7.4 Discussion These results provide evidence that (1) expansion of in vitro lungs from fetal guinea pigs triggers a unique mechanism of lung liquid reabsorption that does not appear to be sensitive to amiloride, (2) changes in glucose concentration in the intraluminal space do not appear to influence this fluid response to expansion, and (3) catecholamine-induced fluid resbsorption by in vitro lungs from fetal guinea pigs, is not blocked by amiloride and the CI" channel blocker, NPPB. The use of 10"6 M amiloride intraluminally did not prevent, or reduce, the reabsorption typical of expanded fetal lung preparations. In the present study, the concentration of amiloride was in the range in which amiloride is believed to inhibit Na + channel transport, and possibly some Na + -H + exchange; however, it should not inhibit the Na + -K + ATPase pump (Cuthbert, Edwardson, Ceves and Wilson, 1979). The present results differ from those of Garrad Nelson and Perks (1996a) which indicated that although amiloride could not restore fluid production by expanded lungs, it canceled the reabsorption. The same work repotted that although the overall result was an apparent abolition of fluid reabsorption, amiloride in some cases did not completely abolish the reabsorptions normally seen following expansion. No correlation was reported between fetal size and the response to amiloride, although the magnitude of the response to lung expansion was shown to increase with fetal size. This might explain the discrepancy between the two studies. In the present study, all lungs used were from large fetuses (105.2 ± 14.2 (SD) g body weight and 62 ± 3 days gestation); the previous studies were made on lungs from smaller fetuses. In studies 139 of adult rabbit lungs, it was reported that amiloride blocked expansion-induced fluid reabsorption only when the lungs were not overinflated (Vejlstrup, Boyd and Dorrington, 1993). However, the expansions in the present study were not excessive. The volumes of saline used to expand the lungs by 70% were 0.64 ± 0.25 ml. This represents a volume that is equivalent to the first breath of newborn rabbits (Lachmann, Grossman, Nilson and Robertson, 1979), which are considerably smaller than newborn guinea pigs. In addition, the pressures generated by our expansions were quite small and transient (see Garrad and Perks, 1996a). Therefore the effects of our expansions should be small in comparison to those expected of guinea pig neonates taking the first breath. It is possible that (1) fluid reabsorption following expansion might be based on Na + transport, but the apical channels involved may be insensitive to amiloride, (2) the mechanism of this fluid reabsorption is based on transport of another ion, possibly CI", and (3) amiloride is able to inhibit the effect of expansion only when used at very high concentrations. In fact, in studies of liquid clearance by lungs of newborn guinea pigs and rabbits, lung water clearance was only reduced by half when intraluminal amiloride concentrations of 10'4 M and 10"3 M were used respectively (O'Brodovich et al, 1990; Song et al, 1992). Unexpanded companion lungs incubated with no intraluminal amiloride decreased fluid production by 88%> while those with intraluminal amiloride decreased fluid production by only 30%). This suggested that amiloride-sensitive Na + channels were responsible for part of the fall in fluid production. Amiloride-sensitive Na + channels are known to be involved in lung fluid reabsorption induced by catecholamines (Olver et al, 1986). The current study reports the 140 release of norepinephrine into fetal lungs in response to expansion (see Chapter 4); however, the concentration of norepinephrine measured in the outer saline was not expected to cause any significant fall in fluid production. Thus, it was surprising that amiloride appeared to block the effects in the companion lung preparations. Furthermore, amiloride had no effect on fluid reabsorption by expanded lung preparations. This could mean that (1) expanded lung preparations reabsorbed fluid by mechanisms which may include epithelial Na + transport, but another mechanism, possibly a reversal of CI" secretion, may overshadow Na + transport, and (2) unexpanded companion preparations reduced fluid production by a mechanism involving mainly Na + transport, possibly stimulated by norepinephrine. However, amiloride could not block direct effects of catecholamines. Thus it seems that effects in the companion lung preparations could have been the result of another released substance which was able to stimulate epithelial Na + transport. The failure of amiloride to block fluid reabsorption by catecholamines was surprising, but closely mirrored those of Chua and Perks (unpublished observations) who recently observed that dopamine, another catecholamine, induced reabsorptions which could not be stopped with intraluminal amiloride or its analogue, benzamil. Previous studies of arginine vasopressin (AVP) and aldosterone on similar in vitro lung preparations clearly indicated the presence of an amiloride-sensitive fluid transport mechanism (Kindler, Chuang and Perks, 1991; Perks et al., 1993a). In those studies, amiloride was used at a final concentration of 10"6 M intraluminally. In the current report, amiloride was used at similar or higher concentrations, thus it seems that the 141 fall in fluid production stimulated by both catecholamines and expansion occurred by mechanisms not involving amiloride-sensitive Na + transport. In parallel studies of electrical potential difference (PD, lumen negative) across in vitro lungs from fetal guinea pigs, catecholamines, but not lung expansion, decreased PD (Chapter 8). This implies that (1) the fall in fluid production by expanded lungs may have occurred as a result of simultaneous influx of Na + and CI" ions caused perhaps by a increased permeability of the epithelium, and (2) catecholamines reduced fluid production, not by stimulating Na + transport from lumen to interstitium, but by stimulating the reversal of CI" secretion. This contrasts results from studies of fetal lamb lungs, which showed that fluid reabsorption occurs only as a result of Na + reabsorption (Olver et al., 1986; Chapman et al., 1994). The failure of amiloride to block the catecholamine-induced fall in fluid production could mean that adrenergic stimulation caused the opening of Na + channels with low affinity for amiloride; however, the reduction in PD following adrenergic stimulation could only mean that the fall in liquid production was a result of CI" transport. Epithelial Na + channels may exhibit low affinity for amiloride (Moran, Asher, Cragoe and Garty, 1988). In fetal lambs, moderate concentrations of amiloride (4.3 x 10"6 M), inhibits epinephrine-induced fluid reabsorption by only 50% (Olver et al., 1986); in newborn guinea pigs, relatively high concentrations of amiloride (10'4 M) inhibited lung fluid clearance by a similar amount (O'Brodovich et al., 1990). Studies of membrane vesicles from fetal sheep alveolar type II cells also indicate that Na + uptake is only impaired by amiloride at very high concentrations (Butcher, Steel, Ward and Olver, 1989; Shaw, 142 Steele, Butcher, Ward and Olver, 1990). More recently, 10"4 M amiloride inhibited 2 2 Na + flux across plasma membrane vesicles from rat fetal epithelial cells by only 30% (Matalon, Bauer, Benos, Kleyman, Lin, Cragoe and O'Brodovich, 1993). These studies suggest that the occurrence of amiloride-insensitive or low affinity Na + channels could be common. There could be species differences in the occurrence of these channels. It is also possible that the responses of fetal lung liquid to luminal expansion and catecholamine stimulation could be due to the elevation of intracellular Ca 2 + concentration. It is known that P-adrenergic agonist stimulation of cultured rat distal lung epithelial cells increases [Ca2 +] i 5 and that an elevation of [Ca2+]j eliminates the blocking action of amiloride on nonselective cation channel (NSC) activity (Marunaka, Tohda, Hagiwara and O'Brodovich, 1992). Similarly, mechanical distention of cultured rat alveolar type II cells mobilizes cytosolic Ca 2 + (Wirtz and Dobbs, 1990). Thus, both expansion and adrenergic stimulation may increase epithelial [Ca2+]j and stimulate nonselective cation channel activity. This could result in fluid reabsorption due to amiloride-insensitive Na + transport. In fact, stimulation of cultured fetal rat distal lung epithelial cells with the P-adrenergic agonist, terbutaline, increases the open probability of nonselective cation channels in the apical membrane (Tohda, Foskett, O'Brodovich and Marunaka, 1994); terbutaline also increases whole cell currents via these channels (Nakahari and Marunaka, 1995). Thus fluid transport based on Na + reabsorption through such channels is a possibility; however, PD would be expected to increase, not decrease as observed, unless PD reflected only bronchial, 143 and not alveolar events (Chapter 8). Thus I explored the possibility that another ion was primarily involved in this seemingly amiloride-insensitive fluid reabsorption. Based on the relative concentrations of individual ions in fetal lung liquid, I hypothesized that other than Na +, only CI" ions would be present in amounts that could be sufficient to facilitate fluid reabsorption. Consequently, lungs instilled with the CI' channel blocker, 5-nitro-2-(3-phenylpropylamino) benzoic acid (NPPB), were treated with norepinephrine to test the possibility of Cl'-based reabsorption. The possibility of Cl'-based fluid reabsorption was considered to be unlikely, since studies of PD across fetal lamb and rat lungs had indicated that PD increases, rather than decreases, as would be expected if CI" secretion was inhibited or reversed (Olver et al, 1986; Krochmal-Mokrzan, Barker and Gatzy, 1993; Chapman et al, 1994). The CI" channel blocker did not affect the reabsorption induced by norepinephrine; NPPB itself appeared to reduce the basal rate of lung liquid production although this was not statistically significant. However, studies with other CI" channel blockers have suggested that they do reduce fluid production (Tognotti and Perks, unpublished observations). Although these observations suggested that the reversal of fluid production was not based on CI" transport from lumen to interstitium, they did not eliminate the possibility of inhibition of CI" secretion from interstitium to lumen. Studies with bumetanide have shown that the inhibition of the Na + CI" cotransporter can result in slight fluid reabsorption (Cassin, Gause and Perks, 1986; Thorn and Perks, 1990). Hence CI" transport may still be involved in the reversal of lung fluid transport seen 144 in guinea pig lungs following adrenergic stimulation. It is also possible that CI" channel transport across the basal, not the apical membrane is important. In the initial experiments in which fetal lungs were expanded, Krebs-Henseleit saline was used as the agent of expansion. The glucose concentration of the saline was 50-fold higher than that of the lung liquid of fetal guinea pigs. Consequently, I lowered the glucose concentration of lung liquid by introducing glucose-free saline as the agent of expansion. There was no significant difference in the response to expansion, indicating that Na+-linked glucose transport was not important in the fluid reabsorption. This suggested that the fall in fluid production by expanded lung preparations was not the result of glucose transport made possible by additional glucose introduced into lungs through Krebs-Henseleit saline itself. Although a Na+-glucose cotransporter exists in lungs of adult rats (Basset, Crone and Saumon, 1987) and may be present in fetal lungs, it is not likely to be important in the rapid clearance of liquid from fetal lungs, since the glucose concentration of fetal lung liquid is very low. At the concentration measured in the fetal guinea pig lungs, glucose would not significantly affect the osmotic balance across the lung epithelium. Furthermore, studies using the Na+-glucose cotransport inhibitor phloridzin demonstrated that Na+-glucose cotransport had little effect on liquid production in fetal sheep lungs (Barker et al, 1989) and on lung water clearance by newborn rat lungs (O'Brodovich et al, 1991). Therefore, as in previous studies of fetal sheep lungs, the present studies suggest that glucose transport is not likely to be important in lung liquid clearance. 145 In conclusion, these results suggest the following: (1) Fluid reabsorption by expanded fetal lungs is not blockable using amiloride. This suggests that expansion may cause increased epithelial permeability and allow influx of CI" and Na + ions, or lower the amiloride sensitivity of the lung epithelium. (2) Catecholamines reduce fluid production or cause reabsorption in fetal guinea pig lungs through a transport mechanism which is insensitive to amiloride. This could mean that the decline in fluid production is based on the reversal of CI" transport. Chapter VIII Transepithelial electrical potentials across fetal guinea pig lungs: Responses to catecholamines, expansion and other factors 116 147 8.1 Introduction Through most of gestation, the fetal lung epithelium transports CI", against its electrochemical gradient, from plasma to lung lumen (Olver and Strang, 1974). This secondary active CI" transport contributes to the generation of a transepithelial potential difference (PD, lumen negative), and drives net movement of liquid into the lung lumen. At birth, this liquid must be cleared to prepare the lungs for gas exchange in the postnatal period. Fluid clearance is thought to involve the activation of a Na+-based reabsorptive mechanism (Olver, Ramsden and Strang, 1981; Olver, Ramsden, Strang and Walters, 1986; O'Brodovich, Hannan, Seear and Mullen, 1990; Bland, 1991; Cassin and Perks, 1993) triggered by hormones such as epinephrine and norepinephrine, which surge at birth (Olver et al., 1981; Woods, Doe and Perks, 1997, Woods, doe and Perks, unpublished observations), or by expansion (Perks and Cassin, 1985; O'Brodovich et al, 1990; Song, Sun, Curstedt, Grossman and Robertson, 1992; Garrad Nelson and Perks, 1996a) which occurs during the first few breaths. In sheep, epinephrine-induced fluid reabsorption is associated with an increase in PD (lumen negative), an increase in Na + flux from lumen to plasma, and an associated decrease in active CI" secretion; the increase in PD and Na + flux can be inhibited with 10"4 M amiloride (Olver et al, 1986). 148 Although amiloride completely blocks effects of other hormones such as arginine vasopressin (Cassin and Perks, 1993; Hooper, Wallace and Harding, 1993), some studies suggest that amiloride-sensitive Na + transport accounts for only part of fluid reabsorption (O'Brodovich et ah, 1990; Song et al., 1992; Garrad Nelson and Perks, 1996a). Recently, in studies of lung liquid production using in vitro fetal lung preparations, amiloride could not completely block fluid reabsorption induced by luminal expansion, and could not slow effects of epinephrine, norepinephrine and dopamine (Perks et ah, unpublished observations). No reason is known for this failure. There is some evidence of amiloride-insensitive Na + transport in fetal lung epithelia (Yankaskas, Cotton, Knowles, Gatzy and Boucher, 1985; Matalon, Bauer, Benos, Kleyman, Lin, Cragoe and O'Brodovich, 1993), but its importance has not been established. To gain further insights into these questions, studies of bioelectric properties of stimulated lungs are essential. It is believed that the entire fetal pulmonary epithelial surface, including the trachea and bronchi, contributes to lung liquid production (Cotton, Boucher and Gatzy, 1988a; Krochmal, Ballard, Yankaskas, Boucher and Gatzy, 1989), whereas Na+-led liquid reabsorption is confined to the distal (alveolar) epithelium (Krochmal-Mokrzan, Barker and Gatzy, 1993; Tessier, Lester, Langham and Cassin, 1996). Due to the difficulty in accessing the distal lung epithelium of whole lungs for direct study, most studies of bioelectric properties of fetal lungs have been done on excised trachea and bronchi, and on cultured monolayers or cysts from distal lung epithelia (Cotton, Lawson, Boucher and Gatzy, 1983; Zeitlin, Loughlin and Guggino, 1988; 149 O'Brodovich, Rafii and Post, 1990; Krochmal-Mokrzan et al, 1993; Barker, Boucher and Yankaskas, 1995). In this study, I measure transepithelial PD changes across in vitro whole lung preparations from fetal guinea pigs. PD is measured before and during stimulation with catecholamines and other active agents, and during luminal expansion. The effects of adrenergic antagonists and ion channel blockers are examined. 8.2 Materials and Methods 8.2.1 Animals Pregnant albino guinea pigs of an inbred departmental stock were given food and water ad libitum (guinea pig chow, Ralston-Purina, supplemented with fresh vegetables and vitamin C). Studies were performed on 36 fetuses of 62 ± 2 days of gestation (term = 67 days) and 102.1 ± 18.1 g body weight (SD). 8.2.2 Surgical Procedures and lung preparation Pregnant guinea pigs were anaesthetized with halothane (Fluothane; Ayerst, Montreal, Que) until full inhibition of the corneal reflex; final euthanasia was accomplished by severing the 150 carotid arteries. The fetuses were removed by Cesarean section, and transferred to Krebs-Henseleit saline (see Burton, 1975) at 37°C. No fetal breathing movements were seen. The fetal thorax was opened by a midline incision to expose the lungs and trachea. The trachea was ligated rostrally and cannulated caudally with polyethylene tubing which was flared on one end. The tubular stem was inserted into the trachea and pushed down to about 5 mm above the bifurcation of the bronchi; the flared end was connected to the trunk of a Y-tube (V16", VWR Canlab, Edmonton AB). One arm of the Y-tube was attached to a 1.0 ml tuberculin syringe via an 18-gauge hypodermic needle and a 3-way stopcock (K75, Pharmaseal, Puerto Rico); the other arm was attached to a polyethylene tube to guide a KCl-agar salt bridge into the airway. All tubing was filled with saline. The cannula was tied in place above the bifurcation of the bronchi, and the trachea was severed rostral to the cannula. The esophagus and vascular attachments to the lung were then cut. During these procedures, care was taken to keep the lungs warm and moist with frequent washes with Krebs-Henseleit saline warmed to 37°C. The heart was removed, then the preparation was rinsed with fresh saline and suspended in 100 ml bath of Krebs-Henseleit saline maintained at 37°C. The saline was oxygenated and maintained at pH 7.4 with 95% 0 2 and 5% C 0 2 . All preparations were set up within 3-4 minutes. Approximately 0.35 ml of lung liquid were withdrawn into the reservoir syringe and mixed thoroughly with 100 ul of Blue Dextran 2000 (50 mg.mi"1 in 0.9% NaCl) introduced in another syringe via the 3-way stopcock. The mixture was then passed into the lungs. The dye was used only to make the liquid mixture in the lung visible, and was not for the purpose of measuring 151 lung liquid production. The preparations were left to equilibrate for 30 min, and in this period, lung fluid was withdrawn and returned every 5 min; the outer saline was replaced every 15 min. Preparations were given various active agents (see below) for 15 min each, then the bathing saline was changed and the preparation left to recover for at least 15 min before the next dose or treatment. Expansions were carried out using Krebs-Henseleit saline (approximately 0.7 ml per 100 g fetal body weight). To expand the lung preparations, a small amount of lung liquid mixture was withdrawn into the reservoir syringe, then Krebs-Henseleit saline, warmed to 37°C, was introduced in another syringe through the 3-way stopcock. The fluids were mixed thoroughly and the mixture was passed into the lungs. 8.2.3 Drugs Test substances were dissolved in saline or in ethanol then placed in the bathing saline and/or directly into the lungs. The following agents were used: Epinephrine, norepinephrine and dopamine (Sigma Chemicals, St. Louis MO); phentolamine (Ciba-Geigy Mississauga, ON); propranolol (Ayerst Laboratories, Montreal, Que); amiloride, benzamil, calcimycin (A23187), loperamide, 5-Nitro-2-(3-phenylpropylamino)benzoic acid (NPPB), N-phenylanthranilic acid (DPC), SQ22536, and verapamil from Research Biochemicals International (Natick MA); PGE 2 (Neogen, Lexington, KY); PGI2, a gift from Dr. Sidney Cassin, University of Florida, Gainsville. 152 8.2.4 Measurement of electrical potential difference Potential differences (PD) were measured using a modification of the method of Hanrahan, Meredith, Phillips and Brandys (1983). KCl-agar salt bridges were prepared in P.E.50 (I.D., 0.58 mm; O.D., 0.965 mm) or PE 10 (I.D., 0.28 mm; O.D., 0.61mm) Intramedic polyethylene tubing (Clay Adams, Parsipany, NJ). Pairs of salt bridges of identical diameter were used for any one measurement. Prior to use, all salt bridges were placed in a common reservoir of 3 M KC1. The exploring salt bridge was guided into the trachea through a larger polyethylene tube (P.E. 90, I.D., 0.86 mm; O.D., 1.27 mm) attached to one arm of the Y tubing. The salt bridge was gently advanced into the lungs until resistance was felt. It was then pulled back about 5 mm and anchored to the guide tubing, using plasticine, so that it was set in place during the experiment. The position of the exploring salt bridge was altered at 5 mm intervals for measurements of PD in different regions of the airway. Maximal depth was 25 mm below the bifurcation. The final position of the salt bridge for all other measurements was approximately 15 mm below the bifurcation of the bronchi. The reference salt bridge was anchored in the bathing cup, 5 mm or less from the pleural surface of the lungs; the position of the reference bridge in the bathing saline did not affect PD measurements at all. As shown in Fig. 20, the bridges were connected via saturated calomel half-cells connected to a high impedance voltmeter, and the signal was recorded on a polygraph (Soltec 220, San Fernando CA). All measurements were corrected for minimal electrode assymetry. 153 8.2.5 Quantification of results and statistical analysis Comparisons of PD in different regions of the airway were were done by analysis of variance (ANOVA) and Scheffe's post-hoc test. Absolute PDs before and after treatment were compared. Changes were also expressed as a percentage of baseline PD (assumed 100%) for proportionality; when data from similar experiments were combined, these percentages were averaged. Differences between treatments were done by two-way A N O V A . All statistical analyses were done by computer using SYSTAT for Windows (Version 5, SYSTAT, Inc., Evanston IL). All mean values are given with their standard errors, unless otherwise stated. Statistical significance was accepted at P < 0.05. 154 Fig 20: Schematic diagram of the apparatus for the measurement of electrical potential difference (PD) across in vitro lungs from fetal guinea pigs. PD was measured using KCl-agar salt-bridges connected via saturated KC1 solutions to calomel half-cells. Salt bridges were prepared in P.E.50 (I.D., 0.58 mm; O.D., 0.965 mm) or PE 10 (I.D., 0.28 mm; O.D., 0.61mm). 156 8.3 Results These studies were carried out on 36 fetuses of 62 ± 2 days of gestation, and 102.1 ± 18.1 g body weight (SD). Recordings had no discernable electronic noise, and all preparations gave steady state recordings after the 30 min equilibration period. Baseline PD was -7.5 ± 0.4 mV, lumen negative (range, -1.9 to -11.4 mV). A typical recording is shown in Fig 21. 8.3.1 Recordings of unstimulated PD at varying positions in the airway These studies were done on 6 fetuses of 63 ± 1 days of gestation and 111.4 ± 19.6 g (SD) body weight. The uppermost position of the exploring salt bridge in exposed trachea was 6.8 ± 1 . 3 mm above the bifurcation; the exploring salt bridge was then advanced at 5 or 10 mm intervals. The lowest position of the bridge was 16.8 ± 1.1 mm below the bifurcation, but was still within the bronchi. This position was confirmed after the experiment using a dissecting microscope. PD increased ten-fold from -1.0 ± 0.3 mV in the trachea to -10.8 ± 0.8 mV in the lower airways; the increase was linear (Fig 22). Measured PD leveled off as the salt bridge met resistance from narrowing airways. Since the magnitude of PD measured corresponded with the depth of the exploring salt bridge, lung preparations from the largest fetuses gave the highest maximum PDs. 157 8.3.2 Effects of catecholamines on PD I Basic effects Six lung preparations were used to study the effects of increasing catecholamine concentration on PD. Norepinephrine was tested on 3 preparations and epinephrine was tested on the remaining 3. Salt bridges used had an outside diameter of 0.97 mm (P.E. 50). The salt bridge was positioned about 10 mm below the bifurcation of the bronchi. Baseline PD was -7.6 ± 0.7 mV. Preliminary experiments with epinephrine and norepinephrine had shown that maximal change in PD occurred within 5 min; therefore 15 min periods were allowed for monitoring effects at each dose. The bathing saline was replaced between doses, and recovery was always complete before the higher dose of catecholamine was applied. Both catecholamines reduced PD in a dose-dependent manner (Fig 23). The minimum concentration of norepinephrine required to affect PD was 10"8 M ; the change in PD was significant (P < 0.05, ANOVA) at 10"7 M . Epinephrine reduced PD at 10"9 M ; the PD change was significant (P < 0.001) at 10"8 M . The dose-response curves for the two catecholamines show that epinephrine was 10 times more effective than norepinephrine at medium concentrations; at 10"6 M , both catecholamines reduced PD by a similar magnitude of about 30%; there was no significant difference between their maximal effects (ANOVA). Additional studies of effects of norepinephrine on PD were carried out on 18 fetuses (62 ± 2 d of gestation; 100.1 ± 21.5 g body weight (SD)) using slender salt bridges (P.E. 158 10). These salt bridges could be inserted deeper into the bronchi. The average position of the salt bridge was 15 mm below the bifurcation. Baseline PD was -7.2 ± 0.5 mV. 10"6 M norepinephrine decreased PD significantly (P < 0.001, ANOVA) to -3.9 ± 0.4 mV. The overall change was a 51.7 ± 2.8% decrease in PD. The fall in PD appeared to be large when baseline PD was large. In preliminary experiments, the P-adrenergic agonist, isoproterenol (10"5 M), decreased PD by 57%, but dopamine (5 x 10"5 M) decreased PD only slightly; however these findings were based on only one or two experiments, and cannot be accepted with any confidence (see Appendix A). II Effects of adrenergic antagonists To determine the receptors involved in the inhibition of transepithelial PD by norepinephrine and epinephrine, the effects of phentolamine (ot-antagonist) and propranolol (P-antagonist) on catecholamine-induced PD changes were studied. Both propranolol (10"? M) and phentolamine (10~5 M) had no effect on baseline PD. However, propranolol blocked effects of both norepinephrine (10"6 M) and epinephrine (10'7 M). Both catecholamines significantly reduced PD (P < 0.05) in preparations pretreated with 10"5 M phentolamine (Table I). It was concluded that catecholamines influenced transepithelial PD through P-receptor stimulation. 159 III Effects of ion channel blockers The epithelial sodium channel blocker, amiloride (IO 4 M), applied directly into the lung lumen, did not affect baseline PD (Table II). Norepinephrine (10"6 M) alone reduced PD significantly (P < 0.05); when applied after amiloride, it still reduced PD significantly (P < 0.05, Table II), and interestingly, the inhibition was sustained for a longer period than in the absence of amiloride (Fig 24). In parallel studies of lung liquid production, amiloride did not block catecholamine-induced lung fluid reabsorption (Chapter 7) These observations suggested that catecholamines inhibited transepithelial PD through an amiloride-insensitive mechanism, but amiloride could delay the redistribution of charge following catecholamine inhibition. Two trials with the amiloride analogue, benzamil (10'4 M), like amiloride, showed Na + channel blockade had no effect on baseline PD (see Appendix). The effects of the epithelial chloride channel blocker, N-Phenylanthranilic acid (DPC), on PD were tested. This blocker was used since NPPB, used earlier, did not reduce fluid production significantly. 10"4 M DPC alone had little effect on baseline PD (Table II). Norepinephrine (106 M), applied after DPC, significantly reduced PD as before (P < 0.05, Table II); this suggested that the norepinephrine-induced change in PD did not depend on apical CI" channels. In two trials, the loop diuretic, bumetanide, reduced baseline PD; norepinephrine when applied after 160 bumetanide, further inhibited PD (see Appendix); the inhibition by bumetanide and norepinephrine did not appear to be additive. 8.3.3 Effects of expansion Five fetuses (63 ± 1 d of gestation; 107.2 ± 8.8 g body weight (SD)) were used to test the effects of luminal expansion on PD. Baseline PD was -8.4 ± 0.8 mV. The lung preparations were expanded by approximately 70% with saline. In contrast to the effects of epinephrine and norepinephrine, there was no significant change in PD (Table III). This implies that expansion had no effect on ion transport across fetal bronchi. However we have shown that expansion causes total lung liquid reabsorption (Chapter 3). Therefore, expansion either (1) triggered fluid reabsorption, across alveolar, not bronchial epithelium, or (2) caused fluid reabsorption across bronchi through the simultaneous influx of Na + and CI" ions. 8.3.4 Effects of other substances Lungs from 3 fetuses were treated with 10"6 M PGE 2 . PD significantly decreased (P < 0.05, Table III).) and did not return to baseline suggesting a long lasting effect of PGE 2 . In preliminary studies PGI 2 had no effect at all on PD, and the calcium ionophore, calcimycin (5 x 10"5 M), decreased PD irreversibly by 93% (see Appendix A). These results mirror those from studies of lung liquid production (Chapter 5). 161 Fig 21: Typical recording of electrical potential difference (lumen -ve, mV) across in vitro lungs from fetal guinea pigs. Recordings had little electronic noise; all preparations gave steady state recordings after the 30 min equilibration period; chart speed, 6 cm per h. 162 > c o o o Baseline 10"6 M N E 163 Fig 22: Potential difference (lumen -ve, mV) across in vitro lungs from fetal guinea pigs. Studies were done on 6 fetuses of 63 ± 1 days of gestation and 111.4 ± 19.6 g (SD) body weight. Measurements were made in airways at increasing depth; relative positions of the electrode tips can be judged from the corresponding diagram of the lungs. The horizontal bars represent trachea (A) and bronchi (B); the vertical bar indicates the bifurcation of the bronchi. Values are mean ± SEM. * P < 0.05, significant difference from tracheal PD (above bifurcation). 164 165 Fig 23: Relationship between catecholamine concentration and change in potential difference (PD) across in vitro lungs from fetal guinea pigs. Studies were done on 6 fetuses of 61 ± 1 days of gestation and 100.6 ± 25.5 g (SD)body weight. Measurements were made approximately 10 mm below the bifurcation of the bronchi. All measurements were maximal effects seen within 15 minutes of stimulation. Open circles represent norepinephrine; closed circles represent epinephrine. Values are mean ± SEM. * P < 0.05, % fall in PD significantly different from zero. Concentration (M) 167 Table I: Blocking effects of adrenergic antagonists, propranolol and phentolamine, on epinephrine- and norepinephrine-induced inhibition of transepithelial PD across in vitro lungs from fetal guinea pigs. Treatment M n PD (mV) baseline during treatment % change Propranolol alone lO"7 6 -7.7 ± 0 . 5 -7.7 ± 0 . 5 0.0 Phentolamine alone lO"5 6 -7.7 ± 0.6 -7.7 ± 0.6 0.0 epinephrine alone lO"7 3 -6.5 ± 0.4 -4.7 ± 0.4* 27.4 ± 2.4 after 10"7 M propranolol lO"7 3 -6.5 ± 0.4 -6.2 ± 0.5 5.5 ±2.8" after 10"5 M phentolamine lO"7 3 -6.7 ± 0 . 3 -5.0 ± 0 . 5 * 24.8 ± 3 . 7 norepinephrine alone 10"6 3 -8.9 ± 0 . 4 -6.1 ± 0 . 2 * 31.4 ± 3.7 after 10"7 M propranolol lfj 6 3 -8.9 ± 0.4 -8.7 ± 0 . 5 1.9 ± 1.9* after 10"5 M phentolamine 10"6 3 -8.8 ± 0 . 4 -6.2 ± 0.2* 29.1 ± 3 . 2 Values are means ± SEM. PD, potential difference (lumen negative); n, number of animals. * Significantly different from baseline (P < 0.05); * Significantly different from %change with catecholamine alone (P < 0.05) 168 Table II: Effects of the sodium channel blocker, amiloride, and the chloride transport blocker, DPC, on norepinephrine-induced inhibition of transepithelial PD across in vitro lungs from fetal "guinea pig. Treatment M n PD (mV) baseline during treatment % change amiloride alone IO"4 4 -6.4 ± 1.1 -6.4 ± 1.1 0.0 DPC alone IO"4 6 -8.9 ± 0 . 9 -7.8 ± 0 . 8 12.5 ± 2 . 1 norepinephrine alone IO"6 10 -8.4 ± 0 . 8 -3.4 ± 0 . 5 * 59.1+3.6 after 10"4 amiloride IO"6 5 -8.9 ± 1.0 -3.3 ± 0 . 3 * 60.8 ± 6.2 after 10" 5MDPC IO"6 5 -7.9 ± 1.2 -3.4 ± 0 . 1 * 61.2 ± 1.8 Values are means ± SEM. PD, potential difference (lumen negative); n, number of animals. * Significantly different from baseline (P < 0.05). 169 Fig 24: Comparison of norepinephrine-induced change in transmural PD (lumen -ve, mV) in the absence (A) and in the presence (B) of intraluminal amiloride. Arrows indicate the time of drug application; chart speed, 6 cm per h. 170 171 Table III: Effects of lung expansion, and of PGE 2 on transepithelial PD across in vitro lungs from fetal guinea pigs. Treatment PD (mV) n Baseline during treatment % change 70% expansion 5 -8.4 ± 0 . 8 -7.8 ± 0 . 7 6.9 ± 3 . 0 10" 6 MPGE 2 3 -6.8 ± 1.1 -4.7 ± 1 . 3 * 33.6 ± 7 . 5 Values are means ± SEM. PD, potential difference (lumen negative); n, number of animals. * Significantly different from baseline (P < 0.05). 172 8.4 Discussion These results provide evidence that (1) fetal guinea pigs, unlike adults, have higher transepithelial electrical PD in the bronchi than in the trachea, (2) catecholamines may reduce fetal bronchial PD without affecting apical amiloride-sensitive Na + channels, and (3) lung expansion does not affect fetal bronchial PD. Our PD measurements were made under open-circuit conditions in the trachea, bronchi and small airways of in vitro lung preparations from fetal guinea pigs. PD values from small airways are similar to those from in vivo adult guinea pig trachea (Boucher, Bromberg and Gatzy, 1980), and are within the range of PDs measured across excised fetal sheep trachea both in open-circuit and short-circuit conditions (Cotton et al., 1983; Zeitlin et al., 1988). However, they are higher than PDs reported from in vivo fetal sheep studies (Olver et al., 1986). Initial PD measurements were made in the trachea, about 7 mm above the bifurcation of the bronchi. This was approximately midway between the larynx and the carina. Final measurements were made about 17 mm below the bifurcation (approximate distance from the bifurcation to the inferior border of the lungs, 20 mm). In their studies of fetal sheep, Olver and coworkers indicated that the final position of their exploring salt bridge was 18 cm below the larynx, which was itself 16.1 cm above the carina (bifurcation of the bronchi). This means that their measurements were made approximately 2 cm (20 mm) below the carina. The actual distance from the carina was similar to ours, but because fetal sheep lungs are much bigger than fetal guinea pig lungs, the position of their exploring salt 173 bridge was anatomically proximal to ours. This may explain the difference between the values although, as previously suggested, there could be differences in electrical PD between species (Boucher et al, 1980). Interestingly, PD changes recorded by Olver and coworkers were consistent with alveolar events, although it is probable that they recorded bronchial PD. In the current study, events in the bronchi appear to differ from those expected of alveoli. It seems likely that there are species differences between the sheep and guinea pig. In the present studies, bronchial PD was higher than tracheal PD, unlike studies of adult lungs which show the reverse (Boucher et al, 1980; Cotton et al., 1983). The magnitude of baseline PD corresponded with the depth of the exploring salt bridge, therefore, lung preparations from the largest fetuses gave the highest maximum PDs. PD increased ten-fold from -1.0 ± 0.3 mV in the trachea to -10.8 ± 0.8 mV, in the lower passages. This distal increase in PD probably reflects differences in the transport of Na + and CI" ions between different regions of fetal and adult airway epithelia. Although I cannot make direct inferences about directional ion movements based on PD measurements alone, the PD values reflect the sum total of ion fluxes across the epithelium and support results from short circuit current studies. The tracheae of most adult mammals studied both secrete CI" and absorb Na +. The bronchi on the other hand absorb Na + at rest but do not typically secrete CI"; the distal lung epithelium absorbs Na + but does not secrete CI" (Boucher, Stutts and Gatzy, 1981; Knowles etal, 1982; Nielson and Lewis, 1990; Krochmal-Mokrzan et al, 1993). Accordingly, unless Na + flux from lung lumen to plasma is much higher in alveolar epithelium, transepifhelial PD would decrease distally in adult lungs. In contrast to 174 their air-breathing adult counterparts, luminal surfaces of fluid-filled fetal lungs secrete CI" but do not absorb Na + significantly until close to term. This probably accounts for the higher PD in adult trachea compared to fetal trachea (Cotton et al., 1983; Zeitlin et al., 1988). As expected, our tracheal PDs were lower than those from adult guinea pig trachea (Boucher et al., 1980). Since little or no active Na + reabsorption was expected in fetal trachea, it was not surprising that PD was not higher in trachea than in the small airways. On the contrary, the distal increase in PD suggests that total CI" secretion is greater in the lower surfaces, perhaps because of the increase in total surface area arising from airway proliferation. Interestingly, in studies of alveolar buds and tracheal cysts from 14 and 16 day old rat fetuses in submersion culture, fetal tracheal PD was found to be higher than alveolar PD (Krochmal-Mokrzan et al., 1993). In those studies, and in most studies of cultured and excised tissues, measurements reflect only events in the immediate vicinity of the studied cells. In our whole lung preparation, it is likely that PD measurements reflected not only influences from epithelial cells close to the tip of the electrode, but also some effects from cells further down the airway. Treatment of the lung preparations with catecholamines surprisingly decreased lung lumen negativity, and pretreatment with propranolol abolished effects of both epinephrine or norepinephrine; pretreatment with phentolamine did not stop these effects. This was very surprising for two reasons: (1) fetal guinea pig lungs are thought to reduce lung liquid production mainly through ot-receptor stimulation (Woods et al., 1997; Doe and Perks, 1998), and (2) the direction of PD change following adrenergic stimulation is opposite that from previous studies of 175 fetal sheep and rats (Olver et al, 1986; Krochmal-Mokrzan et al, 1993) and is inconsistent with the model of stimulated apical Na + reabsorption. I will address these two problems separately. Previous experiments using guinea pig fetuses from the same inbred stock indicate that catecholamines decrease lung liquid production or cause liquid reabsorption (Woods et al, 1997). However, unlike studies on fetal sheep (Walters and Olver, 1978), stimulation with a specific |3-agonist (isoproterenol) could not reduce fluid production significantly. In addition, phentolamine, but not propranolol, blocked the effects of epinephrine and norepinephrine on lung liquid production. Thus it seemed odd that in the current study, pretreatment of lung preparations with propranolol, but not phentolamine, blocked the change in P D . The inconsistency may arise from two factors: (1) the lungs were immature and had not developed sufficient numbers of a-receptors which are thought to control fluid movement in lungs of this species, or (2) the PDs reflect events in the airways and do not reflect ion movements in the alveolar region. The first possibility is unlikely for a number of reasons. Firstly, the studies of Woods and co-workers were done on fetuses of comparable gestational maturation or younger. Secondly, in comparison to other mammals, fetal guinea pig lungs are known to mature early (Lechner and Banchero, 1982; Collins, Kleinerman, Moessinger, Collins, James and Blanc, 1986; Sosenko and Frank, 1987). Finally and perhaps most importantly, in parallel studies of lung liquid production (Chapter 4), adrenergic stimulation caused lung liquid reabsorption, a result similar to that of Woods and co-workers. The lungs were clearly able to respond to adrenergic stimulation as they had before. Thus the second possibility is a more likely 176 explanation. It was improbable that the PD measured was alveolar PD because the exploring salt bridge was relatively large. The airways branch into very narrow bronchioles and eventually tiny alveolar sacs which are virtually inaccesible to a salt bridge with an outside diameter of 0.61 mm. Since the pulmonary epithelium of the fetus is relatively leaky (Olver and Strang, 1974), the PD measured would relate to events close to the tip of the measuring electrode, i.e., the bronchial epithelium. There is no clear evidence of any functional differentiation of the fetal respiratory tract before birth. On the contrary, most of the available evidence suggests that the entire respiratory tract secretes CI' through most of gestation (Cotton, et al., 1983; McCray and Welch, 1991; McCray, Bettencourt and Bastacky, 1992a; 1992b; Barker et al, 1995) and only switches to Na + reabsorption during labour (Bland, Hansen, Haberken, Bressack, Hazinski, Raj and Goldberg, 1982; Brown, Olver, Ramsden, Strang and Walters, 1983; Chapman et al, 1994). In the postnatal period, Na + reabsorption continues (Ramsden et al, 1992), but the trachea and, in some cases bronchi, continue to secrete CI" (Boucher et al, 1981; Knowles et al, 1982; Cotton et al, 1983). However, the signaling mechanisms that control ion transport in fetuses differ from those in adults, and could differ in regions of the same lung. In tracheae of adult humans and dogs, P-adrenergic receptors are linked to CI' secretion (Zeitlin et al, 1988); in the alveolar epithelium of fetal lambs and rats, they are thought to be linked to Na + reabsorption (Olver et al, 1986; Krochmal-Mokrzan et al, 1993). Signaling mechanisms may also differ in various stages of fetal development. In cultured distal lung epithelial cells from developing human and rat fetuses, 177 P-adrenergic stimulation causes CI" secretion and is thought to drive lung fluid production (McCray et al, 1992a; 1992b; Barker et al, 1995); in term fetuses, P-adrenergic stimulation is known to cause alveolar Na + and fluid reabsorption (Brown et al., 1983). As mentioned above, studies of in vitro lungs from near term fetal guinea pigs suggest epinephrine-induced fluid reabsorption is mediated through oc2-, and not P-adrenergic receptors (Doe and Perks, 1998; Woods et al., 1997). The evidence provided in the current study, of the involvement of P- and not a-adrenergic receptors in the PD response, may reflect a regional distribution of adrenergic receptors in the lung epithelium. Both receptor types occur in guinea pig lungs (Barnes, Karliner, Hamilton and Dollery, 1979; Siegl and Orzechowski, 1981; Engel, 1981). I speculate that, in near term fetal guinea pigs, P receptors may predominate in the airways and a receptors may be concentrated in the alveolar epithelium. Studies of rat and rabbit lungs suggest that P receptor density increases continuously even after birth (Giannopoulos, 1980; Whitsett, Machulskis, Noguchi, and Burdsall, 1982). Their role in modulating CI" and fluid secretion in adults airways is well documented (Al-Bazzaz and Cheng, 1979; Boucher and Gatzy, 1982; Liedtke and Tandler, 1984; Welsh, 1986; 1987), and underlines their importance in postnatal lung function. In contrast, a 2 receptor density in rat lungs peaks in late gestation and falls right after birth (Latifpour, Jones and Bylund, 1982; Latifpour and Bylund, 1983). This suggests that a 2 receptors may have an important developmental role in lung function during the perinatal period. 178 Studies on fetal sheep have shown that catecholamine-induced inhibition of fetal lung fluid secretion and fluid reabsorption are associated with an increase in lumen negativity (Olver et ah, 1986). Based on this, and on evidence of increased Na + flux from lumen to plasma, it was proposed that Na + flux itself explained the behaviour of PD during adrenergic stimulation. The reasons for the stark contrast between those observations and our results are not clear. Olver and co-workers noted that the PD increase and Na + reabsorption were associated with a decrease in active CI" transport in the opposite direction. This would itself tend to decrease, not increase, lumen negativity. The increase in Na + reabsorption must indeed far outweigh the decline in CI" secretion to cause increased lumen negativity. It is possible that in our lung preparations, the catecholamine-induced decline in CI" secretion was greater than Na + reabsorption, at least in the airways. Ironically, in parallel studies of lung liquid production on similar preparations, adrenergic stimulation clearly caused liquid reabsorption (Chapter 4). This could mean that adrenergic stimulation triggers two simultaneous events: (1) an increase in Na + reabsorption from lung lumen to plasma and (2) an inhibition of CI" secretion in the opposite direction. These events might occur in different regions of the fetal respiratory tract. If stimulation of Na + reabsorption predominates in the alveoli, and inhibition of CI' secretion in the airways, adrenergic stimulation would increase PD only in the alveoli; in the airways, PD would decrease. This could also explain the failure of amiloride to stop the change in PD. The net reabsorption of lung liquid seen in response to catecholamine treatment suggests the overwhelming importance of the distal alveoli in fluid reabsorption. 179 In the current study, amiloride alone had virtually no effect on baseline PD; this mirrors previous studies (Cotton et al., 1988a; Krochmal-Mokrzan et al., 1993). Although short circuit current studies are needed for confirmation, this result suggests that Na + channel transport did not contribute significantly to the electrical potential at rest. Norepinephrine decreased PD in preparations pretreated with amiloride, suggesting that adrenergic stimulation could reduce CI' secretion with little or no effect on Na + reabsorption. Interestingly, the usual return of PD to baseline following adrenergic stimulation was delayed. Thus amiloride appeared to inhibit the redistribution of charge following adrenergic stimulation. In addition, the apical CI" channel blocker, DPC, did not alter baseline PD significantly, and could not block the norepinephrine-induced fall in PD. These observations suggest two things: (1) p-adrenergic stimulation causes inhibition of CI" secretion leading to a decline in PD. (2) The return of PD to baseline probably depends on the gradual restoration of CI" secretion as well as a tonic flux of Na + from lumen to epithelium through apical Na + channels. The amiloride studies suggest that this Na + flux is small and is overwhelmed by CI" secretion under resting conditions. However, when added before adrenergic stimulation, amiloride prevented the small Na + flux and delayed the restoration of baseline PD. In this scheme, adrenergic stimulation inhibits CI" secretion; this unmasks Na + flux from lung lumen to epithelium. As mentioned earlier, in studies of lung liquid production using similar lung preparations, norepinephrine caused significant fluid reabsorption (Chapter 4). This means that adrenergic stimulation may well produce fluid reabsorption without increasing PD. Fluid reabsorption 180 could not be inhibited with 10"5 M amiloride, implying that reabsorption was accomplished without significant amiloride-sensitive Na + reabsorption. The latter agrees with previous observations that amiloride-sensitive Na + transport accounts for only 50% or so of reabsorbed fluid (O'Brodovich et al, 1990; Song et al., 1992). The amiloride-insensitive component of fluid reabsorption could reflect changes in CI" rather than Na + transport. The changes in PD reported here suggest either that p-adrenergic stimulation (1) acts only to inhibit CI" secretion or (2) allows CI' transport from lumen to interstitium, or some element of both. However, PD probably reflects local changes in deeper bronchi, and not in alveoli which may move fluid. Treatment of lung preparations with 10"6 M PGE 2 caused a significant reduction in lumen negativity. These results mirror those from studies of lung liquid production; 10"6 M PGE 2 significantly reduced lung liquid production (Chapter 5). However, the response was opposite to that reported from studies of first trimester peripheral lung tissue explants in which PGE 2 increased lumen negativity (McCray and Bettencourt, 1993). This study concluded that PGE 2 participates in the stimulation of fetal lung fluid secretion. This response was similar to that of P-adrenergic agonists on premature fetal lungs (McCray et al, 1992a; 1992b); later in gestation, fetal lungs increasingly reabsorb fluid in response to p-adrenergic stimulation (Brown et al, 1983). It has been suggested that this phenomenon reflects the development of a step beyond receptors in the signal transduction chain (Olver, Ramsden and Walters, 1987; Walters, Ramsden and Olver, 1990). Recent evidence points to increased expression of Na + transport proteins in the pulmonary epithelium (Tchepichev, Ueda, Canessa, Rossier and O'Brodovich, 1995; Crump, 181 Askew, Wert, Lingrel and Joiner, 1995). The signaling events triggered by PGE 2 are probably similar to those triggered by catecholamines. Like P-adrenergic agonists, PGE 2 may stimulate fluid production early in gestation, but cause fluid reabsorption later in gestation. The difference in the relative gestational maturation between fetuses used in the present study and those used by McCray and Bettencourt is large. Perhaps the signaling mechanisms for PGE 2 at these two stages of fetal lung development are different. This might account for the observed differences in the actions of PGE 2 . Lung expansion did not change PD significantly although expansion caused powerful lung liquid reabsorption in studies of lung liquid production (Chapter 3). That PD did not change does not necessarily mean the absence of active ion transport; however, the change in luminal fluid volume must reflect net movement of either electrolytes or non-electrolytes from the luminal compartment. The absence of any change in PD could mean that (1) fluid reabsorption was due to the movement of non-electrolytes, or (2) expansion caused symmetrical movement of anions and cations from lumen to interstitium. Since the concentration of non-electrolytes such as glucose is very low in fetal lung liquid (Chapter), it is unlikely that fluid reabsorption was due to the movement of non-electrolytes. In addition, polarity across the epithelium was unchanged, and the epithelium could still respond to adrenergic stimulation in the same way. Therefore it is unlikely that expansion caused any damage to the epithelium. Since bronchial PD responded to stimulation with catecholamines but not to expansion, it may also mean that the rigid airways do 182 not expand; therefore expansion may trigger ion transport but only in alveoli, which are more elastic. In summary, P-adrenergic stimulation caused a reduction in PD, and intraluminal amiloride could not stop the effect. Both observations do not support the Na+-based fluid reabsorption proven in sheep. The fall in PD reflects changes in transepithelial CI", not Na + transport. In addition, studies using adrenergic antagonists show that P receptors are active in bronchi, but their failure to block net fluid reabsorption (Woods et al., 1997) suggests that a receptors could be more important in alveoli, at least in fetal guinea pig lungs. While measured PD did not change with expansion, parallel studies of lung liquid production showed that expansion could cause massive fluid reabsorption (Chapter 3). This suggests that bronchi are not affected by expansion, but that expansion may trigger ion transport mechanisms and fluid reabsorption in alveolar cells, far from the measuring electrode. Chapter I X General Discussion 184 This thesis attempts to identify the mechanisms involved in fluid reabsorption following lung expansion, or treatment with catecholamines. First, the results confirm earlier observations that expansion of fetal lungs in vitro can reduce lung liquid production or turn it to reabsorption, and liberate an agent that can reduce liquid production in a second lung preparation (Garrad Nelson and Perks 1996a, 1996b). Second, my observations suggest that the responses of the companion lung preparations may be due to a combination of factors, and not just one agent. Third, the results show that in guinea pigs lungs, catecholamines may cause lung liquid reabsorption with little or no involvement of amiloride-sensitive Na + channels. When expanded with saline by approximately 70%, fetal lungs consistently reduced fluid production or reabsorbed fluid. This amount of expansion approximates that achieved by the first breath; however, in contrast to air expansion, saline expansion generates only small (0.5-1.0 cm H 2 0) and transient intraluminal pressures. In addition, it does not cause any discernible damage to lung tissue (Garrad Nelson and Perks, 1996a). The lung preparations did not have an intact pulmonary circulation; therefore, movement of fluid from lumen to interstitium could not be due to a colloid osmotic pressure. Thus the reduction in liquid production, or reabsorption probably involved epithelial transport. However, expansion did not change PD. This could mean that (1) fluid reabsorption occurred due an influx of non-electrolytes, (2) PD was measured in a region of the lungs that did not participate significantly in fluid reabsorption, and (3) expansion triggered the simultaneous influx of cations and anions. The concentration of macromolecules such as 185 glucose is very low in fetal lung liquid, therefore the first option is highly unlikely. The PD recordings were most likely taken from deeper bronchi and did not necessarily reflect events in the distal lung spaces. The bronchi are fairly rigid and are not likely to expand significantly; therefore, expansion may have triggered directional ion transport, but only in cells of alveoli, which are far more elastic. This may explain the large fluid reabsorptions occurring despite the apparent absence of PD change. Still, there is a paradox. The use of intraluminal amiloride did not prevent or reduce fluid reabsorption even though the concentration of amiloride was in the range in which it inhibits Na + channel transport (Cuthbert et al., 1979). This failure of amiloride to block the effects of expansion suggests that if apical Na + channels were involved, they must be insensitive to amiloride. It may also mean that the fluid reabsorption was based on transport of another ion, possibly CI". Finally, it could mean that lung expansion caused the simultaneous influx of Na + and CI" ions, perhaps by increasing the permeability of the epithelium. These observations were surprising, but were confirmed in parallel studies using catecholamines and amiloride. Like expansion, catecholamines reduced lung liquid production; amiloride did not significantly reduce these effects. This suggested that catecholamines did not cause reduction of lung liquid production through amiloride-sensitive Na + transport, but possibly through an amiloride-insensitive Na + transport mechanism, or the inhibition of CI' secretion. However, studies with CI" channel blockers proved inconclusive since the blockers themselves reduced both liquid production and baseline PD. Even though the responses of expanded lungs resemble those of lungs treated with catecholamines, there is a major difference. Whereas catecholamines 186 reduced lumen negativity, expansion did not. Therefore, amiloride did not inhibit fluid reabsorption by expanded lungs, or by lungs treated with catecholamines, but perhaps for different reasons. Lung expansion may have moved fluid by increasing epithelial permeability. If expansion activated directional ion transport, these mechanisms were masked by direct effects of expansion. Fluid reabsorption most likely occurred in the alveolar epithelium and not in airway regions that may not expand significantly. Catecholamines on the other hand probably reduced fluid production or caused reabsorption by the inhibition of active CI" secretion. These responses were most likely carried out by all pulmonary epithelial cells including tracheal and bronchial cells. Some involvement of Na + ions cannot be ruled out since responses of the companion lungs to amiloride suggest apical Na + transport at low catecholamine concentration. These results suggest a mechanism of fluid reabsorption that could be different from that proven in sheep (Olver et al., 1986; Chapman et al., 1994). This conclusion is supported by recent experiments using in vitro lungs from fetal guinea pigs, which showed that amiloride and its analogue, benzamil, could not block fluid reabsorption following stimulation with dopamine (Chua and Perks, unpublished observations). This discrepancy between transport processes observed in the sheep and guinea pig lungs suggests species differences. The current studies also confirm that expanded lungs liberate an agent that can influence another lung. However, various blocking agents including phentolamine and propranolol could not stop reabsorption by expanded lungs; on the other hand, propranolol appeared to inhibit significant effects in the unexpanded companion lungs. This suggested that catecholamines, if released, 187 could not be important to the response of expanded lungs; however, p receptors have a role in the companion lungs. It could mean that in expanded lungs, other factors overshadowed any effects of adrenergic stimulation, but in the absence of direct effects of expansion, adrenergic stimulation could be significant. Since the concentration of norepinephrine in the bathing saline was low, it was not expected to reduce fluid production significantly; therefore, norepinephrine may have been converted by PNMT within the companion lungs to epinephrine, which is a more potent agent. Alternatively, there could have been synergism between norepinephrine and another agent. From direct analysis of the outer saline, I ruled out any significant involvement of either somatostatin or PGE 2 in the responses of the companion lung preparations. Somatostatin and prostaglandins, including PGE 2 , can be released into lung tissue following expansion (Berry et al, 1969; Said et al, 1972; Leffler et al, 1980; 1984; Perks et al, 1992 ). However, somatostatin-like immunoreactivity did not increase in the outer saline following expansion, and PGE 2 was not detected at all in the outer saline. The direct use of various prostaglandins on in vitro lung preparations suggested that some prostaglandins can reduce lung liquid production or cause reabsorption. Furthermore, the use of indomethacin, an inhibitor of prostaglandin synthesis, suggested that a prostaglandin may be involved in the response of the companion lungs. I do not know whether one or more prostaglandins were released into the bathing saline, but analysis of the bathing saline suggested that it was not PGE 2 . Of the other prostaglandins 188 tested directly on lungs, only P G F 2 a would be most likely to cause reductions in fluid production comparable to the companion lungs. Agents other than catecholamines or prostaglandins can also influence lung liquid production. The direct use of Ca 2 + secretagogues on the in vitro lung preparations suggested that elevated intracellular Ca 2 + ion concentration could reduce lung liquid production; these effects were long lasting. If elevated, intracellular Ca 2 + may be important in regulating the open probability of nonselective apical Na + channels as suggested by Marunaka et al. (1992).I did not test the effects of any inhibitors of intracellular Ca 2 + ; however, the use of mianserin, a histamine antagonist, suggested that histamine was not important to the responses of either expanded or unexpanded companion lungs. Collectively, the experiments with various agonists and antagonists suggest the possibility that a variety of agents may be involved in the responses of the companion lungs. These agents may together inhibit liquid production in the companion lungs, but cannot individually account for the total response. Studies with amiloride suggested that apical Na + transport could be involved in the inhibition of fluid production in the companion lungs, but not in the the expanded lungs, possibly because the superimposition of other effects masked effects of amiloride. Likewise, in lungs treated with catecholamines, amiloride had no apparent effect. The mechanisms of fluid reabsorption in expanded lungs and in those treated with catecholamines are similar in some respects, but they are both different from the mechanism involved in the unexpanded companion 189 lungs. Responses of companion lungs were blocked with amiloride, suggesting apical Na + transport similar to that identified in fetal lamb lungs (Olver et al, 1986; Chapman et al, 1994). Like amiloride, propranolol inhibited significant responses in the companion lungs; it also prevented catecholamine-induced PD change in the bronchi. The results suggest that (1) the inhibition of fluid production by companion lungs may occur mainly in the airways, and not in alveoli, which may move most of the fluid, possibly through ot2 receptor stimulation (Doe and Perks, 1998), and (2) the mechanism of inhibition of fluid production involves, at least in part, the activation of amiloride-sensitive Na + transport. However, there is a paradox. Amiloride did not inhibit the reduction in catecholamine-induced PD; these measurements were presumed to be in the airways. Perhaps amiloride can inhibit effects of catecholamines at lower concentrations, but not at concentrations such as those used here. In conclusion, expanded lungs reduce fluid production and reabsorb fluid through major influences. These influences mask lesser ones, which could be detected by the companion preparations. The lesser influences, perhaps occurring in fetal airway epithelium, may be present as an insurance for the drainage of lungs, a vital process in the transition from intrauterine to extrauterine life. 190 References Abman, S.H. and K.R. Stenmark. 1992. Changes in lung eicosanoid content during normal and abnormal transition in perinatal lambs. Am. J. Physiol. 262 (Lung Cell. Mol. Physiol. 6):L214-L222 Adams, F.H., T. Fujiwara and C. Rowshan. 1963. The nature and origin of the fluid in the fetal lamb lung. J. Pediatr. 63(5):881-888. Adams, F.H., M . Yanagisawa, D. Kuzela and H. Martinek. 1971. The disappearance of fetal lung fluid following birth. J. Pediatr. 78:837-843 Adamson, T.M., R.D.H. Boyd, H.S. Piatt and L.B. Strang. 1969. Composition of alveolar liquid in the fetal lamb. J. Physiol. 204:159-168 Adamson, T.M.,V. Brodecky, T.F. Lambert, J.E. Maloney, B.C. Ritchie and A . M . Walker. 1975. Ajebak 53(l):65-67 Aherne, W. and M.J.R. Dawkins. 1964. The removal of fluid from the pulmonary airways after birth in the rabbit, and the effect on this of prematurity and prenatal hypoxia. Biol Neonat. 7:214-229 Al-Bazzaz, F.J. and E. Cheng. 1979. Effect of catecholamines on ion transport in dog tracheal epithelium. J. Appl. Physiol. 47:397-403 Alcorn, D.T., M . Adamson, T.F. Lambert, J.E. Maloney, B.C. Ritchie and P.M. Robinson. 1977. Morphological effects of chronic tracheal ligation and drainage in the fetal lamb. J. Anat. 123:649-660 Atsushi, N., M . Katayama, W.M. Thurlbeck, R. Matsui, S. Yasui and K. Konno. 1995. Effect of indomethacin on lung development in postnatal rats: possible role of prostaglandin in alveolar formation. Am. J. Physiol. 268 (Lung Cell. Mol. Physiol. 12):L56-L62 Austin, C.E., J.W. Dear, H. Neighbour, V. Lund and J.C. Foreman. 1996. The contribution of histamine to the action of bradykinin in the human nasal airway. Immunopharmacol. 34:181-189 Balaguer, L. , J. Romano and P. Ruiz-Pesini. 1992. Localization of serotonin, cholecystokinin and somatostatin immunoreactivity in the lower respiratory tract of embryonic, foetal and postnatal sheep. Histol. Histopath. 7:703-708 191 Barker, P.M., R.C.Boucher and J.R. Yankaskas. 1995. Biolelectric properties of cultured monolayers from epithelium of distal human fetal lung. Am. J. Physiol. 286 (Lung Cell Mol. Physiol. 12):L270-L277. Barker, P.M., C.A.R. Boyd, C A . Ramsden, L.B. Strang and D.V. Walters. 1989. Pulmonary glucose transport in the fetal sheep. J. Physiol. 409:15-27 Barker, P.M., M.J. Brown, C A . Ramsden, L.B. Strang and D.V Walters. 1988. the effect of thyroidectomy in the fetal sheep on lung liquid reabsorption by adrenaline or cyclic AMP. J. Physiol. 407:373-383 Barnes, P., J. Karliner, C. Hamilton and C. Dollery. 1979. Demonstration of a^adrenoreceptors in guinea pig lung using 3H-prazosin. Life Sci. 25:1207-1214 Basset, G., C. Crone and G. Saumon. 1987. Fluid absorption by rat lung in situ: pathways for sodium entry in the luminal membrane of alveolar epithelium. J. Physiol. Lond. 384:325-345 Bergman, B., T. Hedner and P. Lundborg. 1978. Effects of terbutaline on the pressure-volume relationship in fetal rabbit lung. Acta Obst. Gynecol. Scand. 57:233-326 Bergman, B., T. Hedner and P. Lundborg. 1980. Pressure-volume relationship and fluid content in fetal rabbit lung after p-receptor stimulating drugs. Ped. Res 14:1067-1070 Berry, E.M. , J.F. Edmonds and J.H. Wyllie. 1969. Release of prostaglandin E 2 and unidentified factors from ventilated lungs. Brit. J. Surg. 53(3): 189-192 Bianco, S., M . Robuschi, R. Ceserani and C. Gandolfi. 1978. Prevention of aspecifically induced bronchoconstriction by prostacyclin (PGI2) in asthmatic subjects. Int. Res. Commun. Syst. Med. Sci. 6: 256 Bistoletti, P., L. Nylund, H. Lagercrantz, P. Hjemdahl and H. Strom. 1983. Fetal scalp catecholamines during labour. Am. J. Obstet. Gynecol. 147:785-788 Bland, R.D. 1988. Lung liquid clearance before and after birth. Sem. Perinat. 12(2): 124-133 Bland, R.D. 1991. Fetal lung liquid and its removal near birth.In: The lung: Scientific Foundations (Crystal, R.G., J.B West et al. eds) pp 1677-1685, Raven Press Ltd, New York Bland, R.D., T.N Hansen, C M . Haberkern, M.A. Bressack, T.A. Hazinski, J.U. Raj and R.B. Golberg. 1982. Lung fluid balance in lambs before and after birth. J. Appl. Physiol. 53(4):992-1004 192 Bland, R.D., D.D. McMillan, M.A. Bressack and L. Dong. 1980. Clearance of liquid from lungs of newborn rabbits. J. Appl. Physiol. 49:171-177 Boucher, R.C., P.A.Bromberg and J.T.Gatzy. 1980. Airway transepithelial electric potential in vivo: species and regional differences. J. Appl. Physiol. 48(1): 169-176 Boucher, R.C. and J.T. Gatzy. 1982. Regional effects of autonomic agents on ion transport across excised canine airways. J. Appl. Physiol. 52:893-901 Boucher, R . C , M.J. Stutts and J.T.Gatzy. 1981. Regional differences in bioelectric properties and ion transport across excised canine fetal neonatal airways. J. Appl. Physiol. 51:706-719 Braude, S., C. Coe, D. Royston and P.J. Barnes. 1984. Histamine increases lung permeability by an H2-receptor mechanism. Lancet 2(8399):372-374 Braunstein G., C. Labat, S. Brunelleschi, J. Benveniste, J. Marsac and C. Brink. 1988. Evidence that the histamine sensitivity and responsiveness of guinea-pig isolated trachea are modulated by epithelial prostaglandin E 2 production. Br. J. Pharmacol. 95:300-308 Brown, M.J., R.E 01ver,C.A. Ramsden, L.B. Strang and D.V Walters. 1981. Effects of adrenaline and of spontaneous labour on lung liquid secretion and absorption in the fetal lamb. J. Physiol. 313:13P-14P Brown, M.J., R.E 01ver,C.A. Ramsden, L.B. Strang and D.V Walters. 1983. Effects of adrenaline and of spontaneous labour on the secretion and absorption of lung liquid in the fetal lamb. J. Physiol. 344:137-152 Burton, R.F. 1975. Ringers solutions and physiological salines, pp 140-143. Dorset Press, Dorchester. Bustos, R., G. Ballejo, G. Guissi, R. Rosas and J.C. Issa. 1978. Inhibition of fetal lung maturation in pregnant rabbits. J. Perinat. Med. 6:240-245 Butcher, P.A., L.W. Steele, M.R. Ward and R.E. Olver. 1989. Transport of sodium into apical membrane vesicles prepared from fetal sheep alveolar type II cells. Biochim. Biophys. Acta 980:50-55 Bylund, D.B. 1988. Subtypes of a2-adrenoreceptor: Pharmacological and molecular biology evidence converge. Trends Pharmacol. Sci. 9:356-361 193 Carlton, D.P., J.J. Cummings, D.L.Chapman, F.R. Poulain and R.D. Bland. 1992. Ion transport regulation of lung liquid secretion in foetal lambs. J. Dev. Physiol. 17:99-107 Carlton, D.P., J.J. Cummings, F.R. Poulain and R.D. Bland. 1992. Increased pulmonary vascular pressure does not alter lung liquid secretion in fetal sheep. J. Appl. Physiol. 72(2):650-655 Carter, Biter, Zfass and Makhlouf. 1978. Inhibition of VIP-stimulated intestinal secretion and cyclic AMP production by somatostatin in the rat. Gastroenterology 74;726-730 Cassin, S. 1984. Effect of indomethacin on fetal lung liquid formation. Can. J. Physiol. Pharmacol. 62:157-159 Cassin, S. 1987. Role of prostaglandins, thromboxane and leukotrienes in the control of pulmonary circulation in the fetus and newborn. Semin. Perinatol. 11:53-63 Cassin, S. and A . M . Perks. 1982. Studies of factors which stimulate lung fluid secretion in fetal goats. J. Dev. Physiol. 4:311-325 Cassin, S. and A . M . Perks. 1993. Amiloride inhibits arginine vasopressin-induced decrease in fetal lung liquid secretion. J. Appl. Physiol. 75:1925-1929 Cassin, S., G. Gause and A.M. Perks. 1986. The effects of bumetanide and furosemide on lung liquid secretion in fetal sheep. Proc. Soc. Exp. Biol. Med. 181:427-431 Cassin, S., M . Tod, J. Philips, J. Frisinger, J. Jordan and C. Gibbs. 1981. Effects of prostaglandin D 2 on perinatal circulation. Am. J. Physiol. 240 (Heart Circ. Physiol. 9):H755-H760 Cassin, S., T. Tyler, C. Leffler and R. Wallis. 1979. Pulmonary and systemic vascular responses of perinatal goats to prostaglandins E, and E 2 . Am. J. Physiol. 240 (Heart Circ. Physiol. 9):H828-H832 Challis, J.R.G., S.R. Dilley, J.S. Robinson and G.D. Thoburn. 1976. Prostaglandins in the circulation of the fetal lamb. Prostaglandins 11:1041-1049 Chapman, D.L., J.H. Widdicombe and R.D. Bland. 1990. Developmental differences in rabbit lung epithelial cell Na7K + ATPase. Am. J. Physiol. 259 (Lung Cell Mol. Physiol. 3):L481-L487 Chapman D.L., D.P. Carlton, D.W. Nielson, J.J. Cummings, F.R. Poulain and R.D. Bland. 1994. Changes in lung liquid during spontaneous labour in fetal sheep. J. Appl. Physiol. 76(2): 523-530 194 Chapman D.L., D.P. Carlton, J.J. Cummings, F.R. Poulain and R.D. Bland. 1991. Intrapulmonary terbutaline and aminophylline decrease lung liquid in fetal lambs. Ped. Res. 29:357-361 Chan, T.B., N. Eiser, D. Shelton and P.J. Rees. 1987. Histamine receptors and pulmonary epithelial permeability. Br. J. Dis. Chest. 81:260-267 Chand, N. and L. DeRoth. 1979. Dual histamine receptor mechanism in guinea pig lung. Pharmacology 19:185-190 Clyman, R.I. 1987. Ductus arteriosus: Current theories of perinatal and postnatal regulation. Semin. Perinatol. 11:64-71 Clyman, R.I., F. Mauray, C. Roman, A .M. Rudolf and M.A. Heymann. 1980. Circulating prostaglandin E 2 and patent ductus arteriosus in fetal and neonatal lambs. J. Pediatr. 97:455-461 Cohen, W.R., G.J. Piasecki and B.T. Jackson. 1982. Plasma catecholamines during hypoxemia in fetal sheep. Am. J. Physiol. 243 (Regul. Integr. Comp. Physiol. 12):R520-R525 Collins, M.H., J. Kleinerman, A.C. Moessinger, A.H. collins, L.S. James and W.A. Blanc. Morphometric analysis of the growth of the normal fetal guinea pig lung. Anat. Record 216:381-391 Comline, R.S. and M . Silver. 1961. The release of adrenaline and noradrenaline from the adrenal glands of the fetal sheep. J. Physiol. 156:424-444 Cotton, C U , R.C.Boucher and J.T.Gatzy. 1988a. Bioelectric properties and ion transport across excised canine fetal and neonatal airways. J. Appl. Physiol. 65(6):2367-2375 Cotton, C U , R.C.Boucher and J.T.Gatzy. 1988b. Paths of ion transport across canine fetal tracheal epithelium. J. Appl. Physiol. 65(6):2376-2382 Cotton, C U , E.E.Lawson, R.C.Boucher and J.T.Gatzy. 1983. Bioelectric properties and ion transport of airways excised from adult and fetal sheep. J. Appl. Physiol. 55(5): 1542-1549. Crump, R.G., G.R. Askew, S.E Wert, J.B. Lingrel and H. Joiner. 1995. In situ localization of sodium-potassium ATPase mRNA in developing mouse lung epithelium. Am. J. Physiol. (Lung cell. Mol. Physiol. 13):L299-L308 Cummings, J.J. 1995. Pulmonary vasodilator drugs decrease lung liquid production in fetal sheep. J. Appl. Physiol. 79(4): 1212-1218 195 Cummings, J.J.,D.P. Carlton, F.R. Poulain, J.U. Raj and R.D. Bland. 1993. Hypoproteinemia slows lung liquid clearance in young lambs. J. Appl. Physiol. 74(1): 153-160 Cuthbert, M.F. 1973. Prostaglandins and respiratory smooth muscle. In: The prostaglandins: Pharmacological and Therapeutic Advances. (Cuthbert, M.F., ed.) J.B. Lippincott C , Philadelphia, pp 253-286 Cuthbert, A.W., J.M. Edwardson, J.A. Ceves and S.A. Wilson. 1979. Amiloride as membrane probe. In: Amiloride and Epithelial Transport. (Cuthbert, A.W., G.M. Fanelli and A. Scribiane eds.), Baltimore MD. Urban and Scharzenberg, ppl 13-123 Dawes, G.S., J.C. Mott, J.G. Widdicombe and D.G. Wyatt. 1953. Changes in the lungs of the newborn lamb. J. Physiol. 121:141-162 Demello, D.E., J.D. Murphy, M.J. Aronovitz, P. Davies and L . M . Reid. 1987. Effects of indomethacin in utero on the pulmonary vasulature of the newborn guinea pig. Pediatr. Res. 22:693-697 Dickson, K.A., J.E. Maloney and P.J. Berger. 1986. Decline inlung liquid volume before labor in fetal lambs. J. Appl. Physiol. 61(6):2266-2272 Dobbs, L.G. , R.F. Gonzalez, L.A. Marinari, E.J. Mescher and S. Hagwood. 1986. The role of calcium in the secretion of surfactant by rat alveolar type II cells. Biochim. Biophys. Acta 877:305-313 Dobbins, J.W., K. Dharmsathphorn, L. Racusen and H.J. Binder. 1981. The effects of somatostatin and enkephalin on ion transport in the intestine. Annal. NY. Acad. Sci. 372:594-612 Doe and Perks. 1998. a-Adrenoreceptor influences on lung liquid movements by in vitro lungs from fetal guinea pigs. J. Appl. Physiol. 84(2):746-753 Douglas, W.W. 1980. Histamine and 5-hydroxytryptamine (serotonin) and their antagonists. In: The Pharmacological Basis of Therapeutics, 6th Edition (Gilman, A.G. , L.S. Goodman and A. Gilman eds.), MacMillan Publishing Co., Inc., NY, pp 609-646 Egan, E.A., R.E. Olver and L.B. Strang. 1975. Changes in non-electrrolyte permeability of alveoli and the absorption of lung liquid at the start of breathing in the lamb. J. Physiol. 244:161-179 Engel, G. 1981. Subclasses of beta-adrenoreceptors: a quantitative estimation of beta,- and beta2-adrenoreceptors in guinea pig and human lung. Postgrad. Med. J. 57(Suppl. l):77-83 196 Enhorning G., D. Chamberlain, G. Contreras, R. Burgoyne and B. Robertson. 1977. Isoxsuprine-induced release of pulmonary surfactant in the rabbit fetus. Am. J. Obst.Gynecol. 129:197-202 Euler, U.S. von. 1954. Adrenaline and noradrenaline. Distribution and action. Pharmacol. Rev. 6; 15-22 Fewell, J.E., A. A. Hislop, J.A. Kitterman and P. Johnson. 1983. Effect of tracheostomy on lung development in fetal lambs. J. Appl. Physiol. 55:1103-1108 Foskett, J.K., H.A. Bern, T.E. Machen and M . Conner. 1983. Chloride cells and the hormonal control of teleost fish osmoregulation. J. Exp. Biol. 106: 255-281 Foskett, J.K. and G.M. Hubbard. 1981. Hormonal control of chloride secretion by teleost opercular membrane. Annal. NY. Acad. Sc. 372: 643 Gardey-Levassort, C , M . Richard, de Lauture, G. Thiroux and G. Olive. 1981. Catecholamines in developing rat lung. Life Sc. 28:331-337 Garrad Nelson, P. and A.M. Perks. 1996a. Effects of lung expansion on lung liquid production in vitro by lungs from fetal guinea-pigs. I. Basic studies and effects of amiloride and propranolol. Reprod. Fertil. Dev. 8:335-346 Garrad Nelson, P. and A . M . Perks. 1996b. Effects of lung expansion on lung liquid production in vitro by lungs from fetal guinea-pigs. II. Evidence for the generation of an inhibitory factor. Reprod. Fertil. Dev. 8:335-346 Giannopoulos, G. 1980. Identification and ontogeny of P-adrenergic receptors in fetal rabbit lung. Biochem. Biophys. Res. Commun. 95(l):388-394 Goodman, B.E. and D. Wangensteen. 1982. Alveolar epithelium permeability to small solutes; developmental changes. J. Appl. Physiol. 57:3-8 Hanrahan, J.W., J.Meredith, J.E.Phillips and D.Brandys. 1983. Methods for the study of transport and control in insect hindgut. In: Methods For The Study Of Insect Epithelia. (Bradley, T., and T.Miller eds.). pp 19-67. New York. Springer-Verlag. Harris, R.A., K.G. Baimbridge, M.A. Bridges and J.E. Phillips. 1991. Effects of various secretagogues of [Ca2+]j in cultured human nasal epithelial cells. Can. J. Physiol. Pharmacol. 69:1211-1216 197 Harris, R.A. and J. W. Hanrahan. 1993. Histamine stimulates a biphasic calcium response in the human tracheal epithelial cell line CF/T43. Am. J. Physiol. 265 (Cell Physiol. 34):C781-C791 Harris, R.A. and J. W. Hanrahan. 1993. Effects of EGTA on calcium signaling in aiway epithelial cells. Am. J. Physiol. 267 (Cell Physiol. 36):C1426-C1434 Hasan, S.U., D.M. Olson, A. Rigaux, A. Bano, J. Pankovich and G.T. Connors. 1996. Umbilical arterial blood flow and plasma prostaglandin E 2 concentrations during arousal and breathing movements in fetal sheep. Pediatr. Res. 40(5):723-731 Higuchi, M . , Y. Murata, Y. Miyake, J. Hesser, J. Tyner, K.A. Keegan and M . Porto. 1987. Effects of norepinephrine on lung liquid flow rate in the chronically catheterized fetal lamb. Am. J. Obst. Gynec. 157:986-990 Hooper, S.B., M.J. Wallace and R. Harding. 1993. Amiloride blocks the inhibition of fetal lung liquid secretion caused by AVP but not by asphyxia. J. Appl. Physiol. 74(1): 111-115 Hume, R., D. Cossar, R. Kelly, M. Giles, A. Hallas, M. Gourlay and J. Bell. 1992. Prostaglandins PGE 2 and PGF 2 a in human fetal lung: Immunohistochemistry and release from organ culture. Exp. Lung Res. 18:259-273 Humphreys, P.W., I.C.S. Normand, E.O.R. Reynolds and L.B. Strang. 1967. Pulmonary lymph flow and uptake of liquid from the lungs of the lamb at the start of breathing. J. Physiol. 193:1-29 Iversen, L.L. 1967. The uptake and storage of noradrenaline in sympathetic nerves. University Press, Cambridge Jelinek, J. and A. Jensen. 1991. Catecholamine concentrations in plasma and organs of the fetal guinea pig during normoxemia, hypoxemia, and asphyxia. J. Dev. Physiol. 15:145-152 Jones, C.T. 1980. Circulating catecholamines in the fetus, their origin, actions and significance. In: Biogenic amines in development. (Parvez, H. and S. Parvez eds.), Elveiser/North Holland Biomedical press, pp 63-86 Jones, C.T. and R.O. Robinson. 1975. Plasma catecholamines in foetal and adult sheep. J. Physiol. 248:15-33 Kercsmar, C M . , Y. Chung and P.B. Davis. 1991. Receptor-mediated cAMP production in adult and infant ferret tracheal epithelium. Ped. Res. 30:75-82 198 Kindler, P.M., D.C. Chuang and A.M. Perks. 1993. Fluid production by in vitro lungs from near-term fetal guinea pigs: effects of Cortisol and aldosterone. Acta Endocrinologica 129:169-177 Kindler, P.M., S. Ziabakhsh and A . M . Perks. 1992. Effects of cAMP,its analogues, and forskolin on lung fluid production by in vitro lung preparations from guinea pigs. Can J. Physiol. Pharmacol. 70(3):330-337 Kitterman, J.A., P.L. Ballard, J.A. Clements, E.J. Mescher and W.H. Tooley. 1979. Tracheal fluid in fetal lambs: Spontaneous decrease prior to birth. J. Appl. Physiol. 47(5):985-989 Kitterman, J.A., G.C. Liggins, J.A. Clements, G. Campos, C H . Lee and P.L. Ballard. 1981. Inhibitors of prostaglandin synthesis, tracheal fluid, and surfactant in fetal lambs. J. Appl. Physiol. 51:1562-1567 Kitterman, J.A., G.C. Liggins, J.E. Fewell and C H . Lee. 1982. Prostaglandin E 2 decreases tracheal fluid production in fetal sheep. Proceedings of the 9th International Symposium on Fetal Breathing and Other Measurements. London, Ontario (abstract) Knowles, M.R., G.F.Murray, J.T.Gatzy and R.C.Boucher. 1982. Effect of amiloride on ion transport across human bronchi in vitro (Abstract). Am. Rev. Respir.Dis. 125: 242 Krochmal, E.M. , S.T. Ballard, J.R. Yankaskas, R.C. Boucher and J.T. Gatzy. 1993. Volume and ion transport by fetal rat alveolar and tracheal epithelia in submersion culture. Am. J. Physiol. 256 (Renal Fluid Electrolyte Physiol. 25):F397-F407 Krochmal-Mokrzan, E.M., P.M. Barker and J.T. Gatzy. 1993. Effects of hormones on potential difference and liquid balance across explants from proximal and distal fetal lung. J. Physiol. 463:647-665 Kunst, A., B. Draeger and J. Ziegerhorn. 1983. UV methods with hexokinase and G-6-P dehydrogenase. In: Methods of Enzymatic Analysis (Bergmeyer, H.U., J. Bergmeyer and M Grabl eds.). Deerfield Beach, FL. Verlag Chemie International, pp 163-172 Lachmann, B., G. Grossmann, R. Nilsson and B. Robertson. 1979. Lung mechanics during spontaneous ventilation in premature and full-term rabbit neonates. Resp. Physiol. 38:283-302 Lagercrantz, H and P. Bistoletti. 1973. Catecholamine release in the newborn infant at birth. Pediat. Res. 11:889-893 Lagercrantz, H and T.A. Slotkin. 1986. The "stress" of being born. Sci. Amer. 254:100-107 199 Latifpour, J. and D.B. Bylund. 1983. Characterization of adrenergic receptor binding in rat lung: Physiological regulation. J. Pharmacol. Exp. Therap. 224(1):186-192 Latifpour, J. S.B. Jones and D.B. Bylund. 1982. Characterization of [3H]yohimbine binding to putative alpha-2 adrenergic receptors in neonatal rat lung. J. Pharmacol. Exp. Therap. 223(3):606-611 Lawson, E.F., E.R. Brown, J.S. Torday, D.L. Mandansky and H.W. Taeusch. 1978. The effect of epinephrine on tracheal fluid flow and surfactant efflux in fetal sheep. Am. Rev. Resp. Dis. 118:1023-1026 Lechner, A.J. and N. Banchero. 1982. Advanced pulmonary development in newborn guinea pigs (Cavia Porcellus). Am. J. Anat. 163:235-246 Leffler, C.W., J.R.Hessler and R.S. Green. 1984. Mechanism of stimulation of pulmonary prostacyclin synthesis at birth. Prostaglandins 28:877-887 Leffler, C.W., J.R.Hessler and N.A. Terragno. 1980. Ventilation-induced released of prostaglandinlike material from fetal lungs. Am. J. Physiol. 238 (Heart Circ. Physiol. 7):H282-H286 Leffler, C.W., T.L. Tyler and S. Cassin. 1978. Effect of indomethacin on pulmonary vascular response to ventilation of fetal goats. Am. J. Physiol. 234(4): H346-351 Leitch, I.M., A.L.A. Boura and R.G. King. 1992. Pharmacological evaluation of the histamine HI and 5-HT blocking properties of 2-AL(carboxamidinonormianserin) (FCC5): in vitro studies. J. Pharm. Pharmacol. 44:315-320 Lewis, A.B., W.N. Evans and W. Sischo. 1982. Plasma catecholamine responses to hypoxemia in fetal lambs. Biol. Neonate 41:115-122 Liedtke, C M . 1986. Interaction of epinephrine with isolated rabbit tracheal epithelial cells. Am. J. Physiol. 251 (Cell Physiol. 20):C209-C215 Liedtke, C M . and B. Tandler. 1984. Physiological responsiveness of isolated rabbit tracheal epithelial cells. Am. J. Physiol. 247 (Cell Physiol. 16):C441-C449 Lines, A., S.B. Hooper and R. Harding. 1997. Lung liquid production rates and volumes do not decrease before labour in healthy fetal sheep. J. Appl. Physiol. 82(3):927-932 200 Liu, K.H. and C U . Chiou. 1981. Continuous, simultaneous and instant display of aqueous humour dynamics with micro-spectrophotometer and sensitive drop-counter. Exp. Eye Res. 32:583-592 Lock, J.E., P.M. Olley and F. Coceani. 1980. Direct pulmonary vasular responses to prostaglandins to the conscious newborn lamb. Am. J. Physiol. 238 (Heart Circ. Physiol. 7):H631-H638 Marchini, G., H. Lagercrantz and K. Uvnas-Moberg. 1990. Plasma gastrin and somatostatin in newborn infants and their relationship to catecholamines. J. Dev. Physiol. 14: 147-156 Martins, A.N., A. Ramirez and T.F. Doyle. 1975. Comparison of radio-iodinated serum albumen and blue dextran as indicators to measure rate of formation of cerebrospinal fluid. Exp. Neurol. 47:249-256 Marunaka, Y., H. Tohda, N. Nagiwara and H. O'Brodovich. 1992. Cytosolic Ca2+-induced modulation of ion selectivity and amiloride sensitivity of a cation channel and beta agonist action of fetal lung epithelium. Biochem. Biophys. Res. Commun. 187:648-656 Matalon, S., M.L. Bauer, D.J. Benos, T.R. Kleyman, C. Lin, E.J. Cragoe, Jr., and H. O'Brodovich. 1993. Fetal lung epithelial cells contain two populations of amiloride-sensitive Na + channels. Am. J. Physiol. 264 (Lung Cell Mol. Physiol. 8):L357-364 Matalon, S., R.J. Bridges and D.J. Benos. 1991. Amiloride-inhibitable Na+ conductive pathways in alveolar type II pneumocytes. Am. J. Physiol. 260 (Lung Cell Mol. Physiol. 4):L90-L96 McCray, P.B. Jr. and J.D. Bettencourt. 1993. Prostaglandins mediate secretion of lung liquid in the human fetal distal pulmonary epithelium. J. Dev. Physiol. 19:29-36 McCray, P.B. Jr., J.D. Bettencourt and J. Bastacky. 1992a. Secretion of lung fluid by the developing fetal rat alvoelar epithelium in organ culture. Am. J. Resp. Cell. Mol. Biol. 6:609-616 McCray, P.B. Jr., J.D. Bettencourt and J. Bastacky. 1992b. Developing bronchopulmonary epithelium of the human fetus secretes fluid. Am. J. Physiol. 262 (Lung Cell. Mol. Physiol. 6):L270-L-279 McCray, P.B.Jr. and M.J. Welch. 1991. Developing fetal alveolar epithelial cells secrete fluid in primary culture. Am. J. Physiol. 260(4):L494-L500 Mcintosh, C.H.S., R. Arnold, E. Bothe, M. Becker, J. Koberling and W. Creutzfeldt. 1978. Somatostatin: extraction and radioimmunoassay in different species. Gut 19:655-663 201 Mcintosh, C.H.S, Y.N. Kwok, C. Tang and J.C. Brown. 1987. The use of monoclonal antibodies in radioimmunoassays for gastrointestinal hormones. J. Clin. Immun. 10:79-84 Moessinger, A.C. , R. Harding, T.M. Adamson, M. Singh and G. Kim. 1990. Role of lung liquid volume in growth and maturation of the fetal sheep lung. J. Clin. Invest. 86:1270-1277 Moran, A., C. Asher, E.J. Cragoe Jr., and H. Garty. 1988. Conductive sodium pathways with low affinity to amiloride in LLC-PK1 cells and other epithelial cells. J. Biol. Chem. 263:19586-19596 Nakahari, T. and Y. Marunaka. 1995. Regulation of whole cell currents by cytosolic cAMP, Ca 2 + , and CI" in rat fetal lung distal epithelium. Am. J. Physiol. 269 (Cell Physiol. 38):C156-C162 Nielson, D.W. 1988. Changes in the pulmonary alveolar subphase in anaesthetized rabbits. Pediatr. Res. 23:418-422 Nielson, D.W. and M.B.Lewis. 1990. Effects of amiloride on alveolar epithelial PD and fluid composition in rabbits. Am. J. Physiol. 258 (Lung Cell.Mol.Physiol. 2):L215-L219 Normand I.S.C., E.O.R. Reynolds and L.B. Strang. 1970. Passage of macromolecules between alveolar and interstitial space in fetal and newly ventilated lungs of the lamb. J. Physiol. 210:151-164 Normand I.S.C., E.O.R. Reynolds and L.B. Strang. 1971. Permeability of lung capillaries and alveoli to non-electrolytes in the foetal lamb. J. Physiol. 219:303-330 O'Brodovich, H. 1991. Epithelial ion transport in the fetal and perinatal lung. Am. J. Physiol. 261(30):C555-C564 O'Brodovich, H., H.V. Hannam, M. Seear and J.B.M. Mullen. 1990. Amiloride impairs lung water clearance in newborn pigs. J. Appl. Physiol. 68(4): 1758-1762 O'Brodovich, H., H.V. Hannam and B. Rafii. 1991. Sodium channel but neither Na + -H + nor Na-glucose symport inhibitors slow neonatal lung water clearance. Am. J. Resp. Cell Mol. Biol. 5:377-384 O'Brodovich, H., B. Rafii and M. Post. 1990. Bioelectric properties of fetal alveolar epithelial monolayers. Am. J. Physiol. 258(2):L201-L206 O'Donnell, S.R., N. Saar and L.J. Wood. 1978. The density of adrenergic nerves at various levels in the guinea-pig lung. Clin. Exp. Pharmacol. Physiol. 5:325-332 202 Olley, P.M., J. Heaton and F. Coceani. 1978. Circulatory effects of a continuous infusion of E-type prostaglandins in the newborn lamb. Advan. Prostaglandin Thromboxane Res. 4:239-247 Olver, R.E. 1981. Of labour and the lungs. Archives of Disease in Childhood. 56:659-662 Olver, R.E., C A . Ramsden and L.B. Strang. 1981. Adrenaline-induced changes in net lung liquid volume flow across the pulmonary epithelium of the fetal lamb: evidence for active sodium transport. J. Physiol. 319:38P-39P Olver, R.E., C A . Ramsden, L.B. Strang and D.V. Walters. 1986. The role of amiloride-blockable sodium transport in adrenaline-induced lung liquid reabsorption in the fetal lamb. J. Physiol. 376:321-340 Olver, R.E., C A . Ramsden and D.V. Walters. 1987. The effect of dibutyryl cyclic AMP on lung liquid volume flow across the pulmonary epithelium of the fetal sheep. J. Physiol. 391:61P Olver, R.E. and Robinson. 1986. Sodium and Chloride transport by the tracheal epithelium of fetal, newborn and adult sheep. J. Physiol. 375:377-390 Olver, R.E., E.E. Schneeberger and D.V. Walters. 1981. Epithelial solute permeability, ion transport and tight junction morphology in the developing lung of the fetal lamb. J. Physiol. 315:395-412 Olver, R.E. and L.B. Strang. 1974. Ion fluxes across the pulmonary epithelium and the secretion of lung liquid in the fetal lamb. J. Physiol. 241:327-357 Orser, B.A., M . Bertlik, L. Ferdoko and H. O'Brodovich. 1991. Cation selective channel in alveolar type II epithelium. Biochim. Biophys. Acta 1094:19-26 Ott, I., M.J. Lohse, K.N. Klotz, I. Vogt-Moykopf and U. Schwabe. 1992. A23187 releases histamine in human lung fragments. Int. Arch. Allerg. Immunol. 98(l):50-56 Palmer, G.C. 1971. Characteristics of the hormonal induced cyclic adenosine 3 ^'-monophosphate response in the rat and guinea pig lung in vitro. Biochim. Biophys. Acta. 252:561-566 Palmer, S.M., G.K. Oakes, R.W. Lam, T.H. Oddie. C.J.Hobel and D.A. Fisher. 1984. Catecholamine physiology in the ovine fetus. I. Gestational age variation in basal plasma concentration. Am. J. Obstet. Gynecol. 149:420-425 203 Padbury, J.F.,R.W. Lam, C.J. Hobel and D.A. Fisher. 1983. Identification and partial purification of phenylethanolamine N-methyl transferase in the developing ovine lung. Ped. Res. 17:362-367 Perks, A .M. and S. Cassin. 1982. The effect of arginine vasopressin and other factors on the production of lung fluid in fetal goats. Chest 81 S:63S-65S Perks, A . M . and S. Cassin. 1985a. The rate of production of lung liquid in fetal goats, and the effect of expansion of the lungs. J. Dev. Physiol. 7:149-160 Perks, A . M . and S. Cassin. 1985b. The effects of arginine vasopressin on lung liquid secretion in chronic fetal sheep. In: The physiological development of the fetus and newborn Academic Press Lond. pp 253-257 Perks, A .M. and S. Cassin. 1989. The effects of vasopressin and epinephrine on lung liquid production in fetal goats. Can. J. Physiol. Pharmacol. 67:491-498 Perks, A .M. , J.J. Dore, R. Dyer, J. Thorn, J.K. Marshall, T. Ruiz, B.A. Woods, E. Vanderhorst and S. Ziabakhsh. 1990. Fluid production by in vitro lungs from fetal guinea pigs. Can. J. Physiol. Pharm. 68:505-513 Perks, A .M. , P.M. Kindler, J. Marshall, B. Woods, M. Craddock and I. Vonder Muhll. 1993a. Lung liquid production by in vitro lungs from fetal guinea pigs: effects of arginine vasopressin and arginine vasotocin. J. Dev. Physiol. 19:203-212 Perks, A .M. , Y.N. Kwok, C.H.S. Mcintosh, T. Ruiz and P.M. Kindler. 1992. Changes in somatostatin-like immunoreactivity in lungs from perinatal guinea pigs and the effects of somatostatin-14 on lung liquid production. J. Dev. Physiol. 18:151-159 Perks, A .M. , T. Ruiz and E. Vanderhorst. 1991. Lung liquid production by in vitro lungs from fetal guinea pigs: Studies with metabolic inhibitors. Can. J. Physiol. Pharm. 69(9): 1247-1256 Perks, A .M. , T. Ruiz, B. Chua, I. Vonder Muhll, P.M. Kindler and W. Blair. 1993b. Reabsorption of fetal lung liquid produced by 2,4-dinitrophenol in in vitro lungs from fetal guinea pigs. Can. J. Physiol. Pharm. 71(1): 1-11 Perks, A .M. , J. Thorn, T. Ruiz, E. Vanderhorst and B.Woods. 1989. The effects of iodoacetate, bumetanide, adrenaline and ANF on fluid production by in vitro lungs from fetal guinea pigs. Proceedings of the International Symposium on Fetal and Neonatal Physiology, 122 (CI 1). Karolinska Institute, Stockholm. 204 Piper, P. and J. Vane. 1971. The release of prostaglandins from lungs and other tissues. Annal. NY. Acad. Sci. 180:363-385 Ramsden, C.A., M . Markiewicz, D.V. Walters, G. Gabella, K.A. Parker, P.M. Barker and H.L. Neil. 1992. Lung liquid flow across the epithelium of the artificially perfused lung of fetal and postnatal sheep. J. Physiol. 448:579-597 Rao, A.K. and G.R. Cott. 1989. Bioelectric properties of fetal rat pulmonary epithelial cells in monolayer culture. Am. Rev. Resp. Dis. 139:A130 Rao, A.K. and G.R. Cott. 1991. Ontogeny of ion transport across fetal pulmonary epithelial cells in monolayer culture. Am. J. Physiol. 261(5):L178-L187 Robinson, C , C C . Hardy and S.T. Holgate. 1985. Pulmonary synthesis, release and metabolism of prostaglandins. J. Allergy Clin. Immunol. 76:265-271 Robinson, C. and S.T. Holgate. 1986. The synthesis, release and effects of prostaglandins in the lung. Asthma: Clinical Pharmacology and Therapeutic Progress, pp 213-225 Russo, R.M., R.L. Lubman and E.D. Crandall. 1992. Evidence for amiloride-sensitive sodium channels in alveolar epithelial cells. Am. J. Physiol. 262 (Lung Cell Mol. Physiol. 6):L405-L411 Said, S.L, S. Kitamura and C. Vriem. 1972. Prostaglandins: Release from the lung during mechanical ventilation at large tidal volumes. J. Clin. Invest. 51(3):83a Saito, H., S. Saito, T. Sano, E. Hosoi and H. Saito. 1983. Fetal and maternal plasma levels of immunoreactive somatostatin at delivery: Evidence for its increase in the umbilical artery and its arterion-venous gradient in the feto-placental circulation. J. Clin. Endocrinol. Metab. 56(3): 567-571 Siegl, P.K.S. and R. Orzechowski. 1981. Alpha- and beta-adrenergically mediated responses in isolated lung strips of guinea pigs. Res. Commun. Chem. Pathol. Pharmacol. 32(3):459-470 Shaw, A . M . , , L.W. Steele, P.A.Butcher,M.R. Ward and R.E. Olver. 1990. Sodium-proton exchange across the apical membrane of the alveolar type II cell of the fetal sheep. Biochim. Biophys. Acta 1028:9-13 Shaw, J.O. and K . M . Mosher. 1975. The current status of prostaglandins and the lungs. Chest 68:75-80 205 Smedira, N., L. Gates, R. Hastings, C. Jayr, T. Sakuma, J.F. Pittet and M.A. Matthay. 1991. Alveolar and lung liquid clearance in anesthetized rabbits. J. Appl. Physiol. 70:1827-1835 Song, Guo-Wei, B. Sun, T. Curstedt, G. Grossmann and B. Robertson. 1992. Effect of amiloride and surfactant on lung liquid in newborn rabbits. Resp. Physiol. 88:233-246 Sosenko, I.R.S. and L. Frank. 1987. Lung development in the fetal guinea pig lung: surfactant, morphology and premature viability. Pediatr. Res. 21:427-431 Steel, R.G.D. and J.H. Torrie. 1970. Principles and procedures of statistics, pp 481, McGraw-Hill, New York. Strang, L.B. 1991. Fetal lung liquid: Secretion and reabsorption. Physiol. Rev. 71(4):991-1016 Tauber, A.L, M . Kaliner, D.J. Stechchulte and K. Frank Austen. 1973. Immunologic release of histamine and slow-reacting substance of anaphylaxis from human lung. V. Effects of prostaglandins on release of histamine. J. Immunol. I l l :27-32 Tchepichev, S., J. Ueda, C. Canessa, B.C. Rossier and H. O'Brodovich. 1995. Lung epithelial Na channel subunits are differentially regulated during development and by steroids. Am. J. Physiol. 269 (Cell Physiol. 38):C805-C812 Tessier, G.J., G.D. Lester, M.X. Langham and S. Cassin. 1996. Ion transport properties of fetal sheep alveolar epithelial cells in monolayer culture. Am. J. Physiol. 270 (Lung Cell. Mol. Physiol. 14):L1008-L 1016 Thorn J. and A . M . Perks. 1990. The effects of furosemide and bumetanide on lung liquid production by in vitro lungs from fetal guinea pigs. Can. J. Physiol. Pharm. 68(8): 1131-1135 Timann, M. , K. Kunzelmann, U. Frobe, L. Cabantchik, H.J. Lang, H.C. Englert and R. Greger. 1991. Different types of blockers of the intermediate-conductance outwardly rectifying chloride channel in epithelia. Pflugers Arch. 418:556 Tohda, H., J.K. Foskett, H. O'Brodovich and Y. Marunaka. 1994. CI" regulation of Ca2+ activated non selective cation channel in p-agonist treated fetal distal lung epithelium. Am. J. Physiol. 266 (Cell Physiol. 35):C104-C109 Wachtfogel, Y.T., R.A. DeLa Cadena and R.W. Coleman. 1993. Structural biology, cellular interactions and pathophysiology of the contact system. Thrombos. Res. 72:1-21 206 Wallace, M.J., S.B. Hooper and R. Harding. 1990. Regulation of lung liquid secretion by arginine vasopressin in fetal sheep. Am. J. Physiol. 258(27):R104-R111 Wallen, L.D., D.T. Murai, R.I. Clyman, C H . Lee, F.E. Mauray, P.L. Ballard and J.A. Kitterman. 1989. Meclofenamate does not affect lung development in fetal sheep. J. Dev. Physiol. 12:109-115 Walters, D.V. and R.E. Olver. 1978. The role of catecholamines in lung liquid absorption at birth. Pediatr. Res. 12:239-242 Walters, D.V., C A . Ramsden., M.J. Brown, R.E. Olver and L.B. Strang. 1982. Fetal lung liquid absorption during epinephrine infusion and spontaneous labor in the lamb. Chest 81(S):65S-66S Walters, D.V., C A . Ramsden and R.E. Olver. 1990. Dibutyryl cAMP induces gestation dependent absorption of fetal lung liquid. J. Appl. Physiol. 68(5):2054-2059 Walters, D.V., L.B. Strang and F. Geubelle. 1985. Physiology of the fetal and neonatal lung. Proc. Int. Symp. Physiol. Pathophysiol. Fet. Neonat. Lung. Brussels, pp 61-73 Welsh, M.J. 1986. Adrenergic regulation of ion transport by primary cultures of canine tracheal epithelium: cellular electrophysiology. J. Membr. Biol. 91:121-128 Welsh, M.J. 1987. Electrolyte transport in airway epithelia. Physiol. Rev. 67(4):1143-1184 Whitsett, J.A., A. Machulskis, A. Noguchi and J.A. Burdsell. 1981. Ontogeny of a,- and 0-adrenergic receptors in rat lung. Life Sci. 30:139-145 Wirtz, H.R.W. and L.G. Dobbs. 1990. Calcium mobilization and exocytosis after one mechanical stretch of lung epithelial cells. Science 250:1266-1269 Wlodek, M.E. , R. Harding and G.D. Thorburn. 1994. Effects of inhibition of prostaglandin synthesis on flow and composition of fetal urine, lung liquid, and swallowed fluid in sheep. Am. J. Obstet. Gynecol. 170:186-95 Woods, B.A., S. Doe and A . M . Perks. 1997. Effects of epinephrine on lung liquid production by in vitro lungs from fetal guinea pigs. Can. J. Physiol. Pharmacol. 75:772-780 Wyszogrodski, I., H.W. Taeusch and M.E. Avery. 1974. Isoxsuprine-induced alterations of pulmonary pressure-volume relationships in premature rabbits. Am. J. Obst. Gynecol. 119:1107-1111 207 Velvis, P. Moore and M.A. Heymann. 1991. Prostaglandin inhibition prevents the fall in pulmonary vascular resistance as a result of rythmic distention of the lungs in fetal lungs. Pediatr. Res. 30:62-68 Vjelstrup, N.G., C.A.R. Boyd and K.L. Dorrington. 1994. Effect of lung inflation on active and passive liquid clearance from in vivo rabbit lung. Am. J. Physiol. 267 (Lung Cell Mol. Physiol. 11):L482-L487 Yankaskas, J.R., C U . Cotton, M.R. Knowles, J.T. Gatzy and R.C Boucher. 1985. Culture of human nasal epithelial cells on collagen matrix supports. Am. Rev. Respir. Dis. 132:1281-1287 Yoshimura, K., M.L. Tod, K.G. Pier and L.J. Rubin. 1989. Effects of a thromboxane A 2 analogue and prostacyclin on lung fluid balance in newborn labs. Circ. Res. 65:1409-1416 Zeitlin, P.L., G.M. Loughlin and W.B. Guggino. 1988. Ion transport in cultured fetal and adult rabbit tracheal epithelia. Am. J. Physiol. 254 (Cell Physiol. 23):C691-C698 208 Appendices Appendix A: Effects of various agents on transepithelial PD across the fetal guinea pig lung. PD (mV) Agent Concn, M n Baseline Treatment Change % Change Dopamine 5 x 10"5M 2 -10.2 ± 0 . 2 -8.2 ± 2 . 0 2.0 ± 1.8 20.4 ± 18.2 Isoproterenol 10"5M 1 -8.0 -3.4 4.6 57.2 PGI2 10"6M 1 -10.0 -10.0 0.0 0.0 Calcimycin 5 x 10- 5M 1 -12.3 -0.9 11.4 92.7 Forskolin 10"6M 1 -9.8 -2.8 7.0 71.8 Ethanol 0.5% 2 -6.6 ± 1.4 -5.5 ± 2 . 1 1.1 ± 0 . 7 18.8 ± 13.8 All agents were applied added to the outer saline. Values are means ± SEM; PD, potential difference (lumen negative); n, number of animals. 209 Appendix B: Effects of various ion transport antagonists on baseline PD across the fetal guinea pig lung. PD (mV) Concn, M n Baseline Treatment Change % Change NPPB 5 x 10"5 3 -7.3 ±0.3 -6.7 ± 0 . 7 0.6 ±0.5 8.8 ±7 . 6 Benzamil 10"4 2 -7.9 ± 1.0 -7.9 ± 1.0 0.0 0.0 Bumetanide lO"5 2 -10.2 ± 1.1 -5.6 ±2.1 4.6 ±3.1 42.5 ± 26.4 SQ22536 io- 5 1 -7.8 -4.6 .3.2 41.0 Anthracene lO"3 1 -4.9 -3.3 1.6 33.0 Loperamide 5 x 10"5 1 -9.7 -9.7 0.0 0.0 Verapamil IO"5 1 -10.8 -10.8 0.0 0.0 Benzamil was added directly into the lung lumen. All other agents were added to the outer saline. Values are means ± SEM; PD, potential difference (lumen negative); n, number of animals. 210 Appendix C: Effects of various ion transport antagonists, on norepinephrine-induced inhibition of transepithelial PD across the fetal guinea pig lung. PD (mV) Antagonist M [NE] M n Baseline Treatment Change % Change Benzamil IO"5 IO"6 1 -7.0 -3.4 3.6 51.7 Bumetanide IO"5 10"6 2 -5.9 ± 2 . 4 -2.7 ± 0 . 7 3.2 ± 1.7 50.9 ± 8 . 1 Verapamil 10'5 10"6 1 -9.3 -2.7 6.6 71.0 Values are means ± SEM. PD, potential difference (lumen negative); NE, norepinephrine; n, number of animals. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088977/manifest

Comment

Related Items