Hormonal control of the surfactant system in fetal lung.

ANNUAL REVIEWS

Further

Quick links to online content Annu. Rev. Physiol. Copyright ©

1991

1991. 53:415-40

by Annual Reviews Inc. All rights reserved

Annu. Rev. Physiol. 1991.53:415-440. Downloaded from www.annualreviews.org Access provided by Technische Universiteit Eindhoven on 01/25/15. For personal use only.

HORMONAL CONTROL OF THE SURFACTANT SYSTEM IN FETAL LUNG Carole R. Mendelson and Vijayakumar Boggaram Departments of Biochemistry and Obstetrics-Gynecology, The Cecil H. and Ida Green Center for Reproductive Biology Sciences, University of Texas Southwestern Medical Center at Dallas, 5323 Harry Hines Boulevard, Dallas, Texas 75235 KEY WORDS:

developmental regulation, pulmonary, gene expression, glucocorticoids, multifactorial

INTRODUCTION Pulmonary surfactant, a lipoprotein enriched in dipalmitoylphosphatidylchol ine (DPPC)

(21), is synthesized by the type II cells of the lung alveolus where

surfactant glycerophospholipids and proteins function to reduce surface ten sion at the alveolar air-liquid interface. Surfactant contains several lung specific proteins (84) that appear to act in concert with DPPC to reduce alveolar surface tension. The fetal lung acquires the capacity for surfactant synthesis relatively late in gestation; augmented surfactant synthesis and secretion are initiated after completion of 85-90% of gestation in all mam malian species thus far studied. In the human fetus, type II cells are first identifiable in the terminal sacs at

20-22 weeks of gestation; how

ever, secretion of surfactant into the amniotic fluid is detectable only after

30-32 weeks of gestation. A consequence of surfactant deficiency in pre maturely born infants is the respiratory distress syndrome (RDS), the lead ing cause of neonatal morbidity and mortality in developed countries

(5). Al

though RDS is primarily associated with prematurity, term infants of diabet ic mothers also manifest an increased incidence of RDS. On the other hand, premature infants of mothers afflicted with chronic or pregnancy-induced 415

0066-4278/9110315-0415$02.00


416

MENDELSON & BOGGARAM

hypertension commonly manifest a decreased propensity to develop RDS (106). The cellular mechanisms involved in the initiation of surfactant synthesis by the fetal lung have not been defined; however, an expanding list of hormones and factors have been found to regulate in vivo and in vitro the synthesis and secretion of surfactant components by the fetal type II cell. In this chapter, we only briefly consider the regulation of surfactant glycero phospholipid synthesis in fetal lung tissue, since this topic has been reviewed extensively (6,37,89). Rather, we focus primarily on recent studies concern ing the effects of hormones and factors on the regulation of surfactant protein synthesis and gene expression in fetal lung. REGULATION OF SURFACTANT GLYCEROPHOSPHOLIPID SYNTHESIS IN FETAL LUNG Developmental Regulation Pulmonary surfactant is a unique lipoprotein because it is comprised of a high proportion of DPPC, the major surface-active component, which accounts for >50% of surfactant glycerophospholipids. In addition, the surfactant of most adult mammals, including man, contains a relatively large proportion of phosphatidylglycerol (PG), which comprises -10% of surfactant composition (21). Both of these glycerophospholipid species are normally present only in trace amounts in other tissues. In most mammalian species, the surfactant that is synthesized initially by fetal lung tissue contains only small amounts of PG, although relatively large amounts of another acidic glycerophospholipid, phosphatidylinositol (PI), are present. With advancing gestation, the relative amount of PG in fetal surfactant increases, while the relative amount of PI declines (42). The reciprocal changes in these acidic glycerophospholipid species result from their synthesis from a common precursor, cytidine di phosphodiacylglycerol (CDP-diacylglycerol). It has been suggested that the decrease in PI synthesis in fetal lung tissue with advancing gestation results from the decreased availability of circulating myo-inositol. In studies with fetal rabbit lung tissue in vitro, it has been observed that synthesis of surfactant PI relative to PG is increased as the concentration of myo-inositol in the culture medium is increased (61). In addition, the PI synthase reaction is reversible in lung tissue and can be shifted in the direction of CDP diacylglycerol by an increase in the levels of cytidine monophosphate (14), which is formed in association with augmented phosphatidylcholine (PC) biosynthesis that occurs in fetal lung tissue during the latter part of gestation (85). Thus the shift in a surfactant enriched in PI to one enriched in PG also may be due to an increase in the amount of CDP-diacylglycerol available for PG synthesis. The role of PG in surfactant function has not been defined, although its

REGULATION OF SURFACTANT SYNTHESIS

417


presence in increased amounts in pulmonary surfactant is correlated with enhanced fetal lung maturity (42). The finding that inositol supplementation of drinking water of adult rabbits resulted in the synthesis of lung surfactant containing only minor amounts of PG (0.3%) and an increased proportion of PI (8.S%) with unaltered surface-active properties suggests that PG does not play an essential role in surfactant function (12). It should be noted that pulmonary surfactant of adult Rhesus monkeys contains relatively small amounts of PG (-1% of surfactant lipid composition) (26).

Regulation by Hormones GLUCOCORTICOIDS The fundamental discovery by Liggins (S7) that ad ministration of synthetic glucocorticoids to fetal lambs resulted in accelerated lung maturation, followed by the findings that glucocorticoid treatment of fetal rabbits enhanced surfactant activity and accelerated the appearance of alveolar type II cells (54, 111), led to numerous studies that supported the concept that glucocorticoids are important in the regulation of surfactant glycerophospholipid synthesis in fetal lung tissue (see 6 for review). The results of the first clinical trial published by Liggins & Howie in 1972 provided evidence that administration of synthetic glucocorticoids to women in preterm labor prior to 34 weeks gestation caused a significant decrease in the incidence of RDS in their premature newborns (S8). The results of numerous in vivo and in vitro studies utilizing fetal lung tissues of a number of species suggest that surfactant glycerophospholipid synthesis by fetal lung is, in fact, subject to multifactorial control and that, in addition to glucocorticoids, prolactin, thyroid hormones, estrogens, andro gens, growth factors, insulin, catecholamines acting through ,B-adrenergic receptors and cAMP are important in its regulation. Since this topic has been reviewed extensively (6,37,89),in this section we focus on some aspects of our own research and that of others on the multifactorial regulation of surfactant glycerophospholipid synthesis by fetal lung tissue. MULTIFACTORIAL REGULATION We and others have utilized fetal lung in organ culture as a model system for study of the regulation of surfactant synthesis because the preservation of tissue architecture and appropriate cellular interactions appear to be essential for initiation and maintenance of type II cell differentiation (62, lOS). Lung explants from midtrimester human abortuses (6S, 99), or from 19-21 day gestational age fetal rabbits (104) differentiate and develop the capacity to synthesize surfactant after several days of organ culture in serum-free medium. Before the start of culture, the ductular epithelial cells of these tissues are columnar in form, contain abun dant cytoplasmic glycogen, and have no lamellar bodies. Within four days of organ culture in serum-free defined medium, the epithelium lining the pre alveolar ducts is comprised of differentiated type II cells that contain numer-


418


ous lamellar bodies (65, 99, 104). These morphologic changes are associated with a marked increase in phosphatidate phosphohydrolase (PAPase) activity, an increased rate of PC and DPPC synthesis (99, 104), and induction of transcription of the gene encoding the major surfactant protein, surfactant protein A (SP-A) (15), with associated increases in the levels of SP-A mRNA and protein (15, 64, 102). Lamellar bodies isolated from human fetal lung explants maintained in organ culture for eight days have a glycerophospholip id composition similar to that of surfactant produced by the fetal lung at 36 to 38 weeks of gestation (101). The mechanisms that underly this phenomenon of in vitro differentiation are not known, although it is thought to result from the removal of the tissue from an inhibitory factor(s) that is present in vivo (99). A role for mesenchyme-derived factors in type II cell differentiation has been proposed by Smith (95), who identified a glucocorticoid-induced lung fibroblast-derived protein (-8 kd), fibroblast-pneumonocyte factor (FPF), that stimulated PC synthesis by type II cells. It has been further suggested that the stimulatory effect of FPF on type II cell PC synthesis is mediated by an effect to increase cellular levels of cAMP (97). Studies from several laboratories using cultured fetal lung tissues from a variety of species suggest that surfactant glycerophospholipid synthesis by the fetal lung is under multifactorial control and that glucocorticoids in concert with a number of other hormones and factors play a major role in its regulation. In studies using fetal rabbit lung explants maintained in organ culture in serum-free medium, we observed that cortisol (10-7) alone stimu lated PC synthesis when lung tissues from 21-28 day gestational age fetal rabbits were utilized; however, a stimulatory effect of cortisol on lung ex plants from 19 day gestational age fetuses was observed only when fetal calf serum (10%) also was present in the culture medium (66, 104). These findings suggest that prior to day 21 of gestation in the rabbit, the capacity of the fetal lung to respond to glucocorticoids with increased PC synthesis is dependent upon exposure to a factor(s) present in fetal serum. In studies using lung explants from 18 day gestational age fetal rats, Gross & Wilson (39) found that dexamethasone and the phosphodiesterase inhibitor, theophylline, each caused an approximate two and one-half fold increase in the rate of DPPC synthesis; whereas, a greater than sixfold increase in DPPC synthesis was found when the two agents were added together. A supra additive effect on DPPC synthesis also was observed when lung explants were cultured in the presence of dexamethasone and Bt2cAMP (39). In studies using rat (39), rabbit (8), and human (38) fetal lung in vitro, additive or synergistic effects of glucocorticoids and triiodothyronine (T3) on PC synthesis have been reported. An interactive effect of glucocorticoids and thyroid hormones on fetal lung maturation also was suggested by in vivo studies of Hitchcock (47), who found that intra-amniotic administration of thyroxine to fetal rats caused an enhanced rate of appearance of differentiated


419


type II cells; this effect was greatly reduced when the mothers were adre

nalectomized or treated with an inhibitor of glucocorticoid synthesis. By contrast, the stimulatory effects of maternal glucocorticoid treatment on fatty acid synthesis and glycogenolysis in fetal lung tissues were found to be antagonized by simultaneous administration of T3 (83). In human fetal lung in vitro, we found that cortisol, in combination with prolactin, insulin, or with prolactin + insulin, enhanced PC synthesis by two to threefold as compared to that of explants maintained in control medium, or in medium that contained either prolactin, cortisol, or insulin alone (65). The synthesis of DPPC in the human fetal lung explants was stimulated by these hormonal combinations in a similar manner; therefore, the ratio of DPPC to PC was relatively unaffected by hormone treatment. Upon morphologic examination of the cultured tissue, it was found that the stimulatory effects of these hormonal treatments on PC and DPPC synthesis were accompanied by a striking increase in the amount of secreted surfactant material within lumina of the prealveolar ducts (65). A stimulatory effect of prolactin on PC and DPPC synthesis also was reported in studies with fetal rat lung in vitro (67). On the other hand, other researchers (38) have been unable to find an effect of prolactin added in the absence or presence of glucocorticoids on PC synthesis by human fetal lung in vitro. A role for prolactin in lung maturation is suggested by the presence of prolactin receptors in fetal lung tissues (2, 10, 49, 92) and the findings that the marked increase in prolactin levels in human fetal plasma (3, 120) precedes the increase in the lecithin to sphingomyelin (US) ratio in amniotic fluid (43), an index of fetal lung surfactant synthesis. Also, significant negative correlations were found between the concentrations of prolactin in cord plasma and the incidence of RDS in premature newborns (36, 41, 43, 98). As discussed above, the surfactant that is synthesized initially in fetal lung tissue is enriched in DPPC and PI and contains small amounts of PG. As gestation proceeds, the relative amount of PI in surfactant declines, whereas the relative amount of PG increases. Thus the ratio of PG to PI in human amniotic fluid increases from 0.04 at 35 weeks of gestation to 1.75 at term (42, 76). In studies to evaluate the hormonal regulation of surfactant glycerophospholipid composition, we observed that lamellar body lipid phos phorus content and relative rates of synthesis of lamellar body PG and PI also are subject to multihormonal regulation. Incubation of human fetal lung explants in medium that contained cortisol together with insulin, prolactin, or both hormones, resulted in a twofold increase in the amount of lamellar body lipid phosphorus isolated from the explants, as compared to tissues main tained in control medium or with either of the hormones alone (101). Further more, these hormones had profound effects on the relative rates of synthesis of lamellar body PG and PI in the human fetal lung in vitro. In lung explants maintained for seven days in control medium, the relative rate of synthesis of


420


lamellar body PG to PI (PG/PI) was 0.4; whereas, the PG/PI of lamellar bodies isolated from insulin + cortisol + prolactin-treated fetal lung explants was increased to 1. 6 (10 I). These findings suggest that surfactant synthesis by the human fetal lung is under multihormonal control; glucocorticoids act in concert with prolactin and insulin to stimulate surfactant DPPC synthesis and to alter the relative rates of synthesis of PG and PI, which result in the accumulation of increased numbers of lamellar bodies enriched in PG with reduced PI content. As noted above, term infants of women with certain forms of diabetes have an increased incidence of RDS (87). Although the LIS ratio in amniotic fluid obtained from diabetic women between 35 and 37 weeks of gestation is frequently normal or even elevated, the levels of PG may be greatly reduced or absent (22). It has been suggested that the fetal hyperinsulinemia associated with maternal diabetes (72) exerts an antagonistic effect on fetal lung matura tion. In studies using fetal rabbit lung cells, Smith & colleagues (96) observed that insulin antagonized the stimulatory effect of cortisol on DPPC synthesis. By contrast, our own studies using human fetal lung in vitro fail to support an inhibitory role of insulin either on surfactant PC and DPPC synthesis or on the synthesis of surfactant PG. As noted above, we observed that in fetal lung explants incubated with insulin + cortisol, lamellar body phospholipid syn thesis was significantly increased as compared to that observed in human fetal lung tissues maintained in control medium or in medium that contained either insulin or cortisol alone (65, 101). In addition, the PG to PI ratio of lamellar bodies isolated from human fetal lung explants incubated with insulin + cortisol + prolactin (PG/PI 1.6) was found to be greater than that of explants incubated with cortisol + prolactin (PG/PI 1.0); the lamellar body PG/PI of insulin + cortisol-treated fetal lung explants (PG/PI 1.4) was similar to that of tissues treated with cortisol alone. Thus our studies using human fetal lung in vitro provide no evidence for an inhibitory role of insulin on surfactant glycerophospholipid synthesis. On the other hand, the recent observations that SP-A levels in amniotic fluid of diabetic mothers are significantly reduced as compared to gestation matched non-diabetic women (50, 100) and that insulin causes a dose-dependent inhibition of SP-A synthe sis in human fetal lung in vitro (103), may help to explain the increased incidence of RDS in newborn infants of diabetic mothers (see below). =

=

=

REGULATION OF SURFACTANT PROTEIN SYNTHESIS AND GENE EXPRESSION IN FETAL LUNG TISSUE Surfactant Proteins-Properties Since the properties of the surfactant proteins and their genes are described in detail in other chapters of this volume (see Haagsman & van Golde; Hawgood & Shiffer) we only will provide a brief review of this topic. In recent years,


421

several surfactant-associated proteins have been isolated and characterized (84). The major protein associated with pulmonary surfactant, SP-A (Mr 29-36,000), is a glycoprotein modified by N-linked oligosaccharide side chains containing sialic acid residues (69). A variety of possible functions have been ascribed to SP-A, including a role coupled with calcium and the hydrophobic surfactant proteins, SP-B and SP-C, in the transformation of the secreted lamellar body into tubular myelin (44, 52) and in the reduction of alveolar surface tension (45). In addition, SP-A may mediate endocytosis and reutilization of secreted surfactant components through binding to specific high-affinity receptors on type II cells (55, 90) and, thereby, may act in a negative-feedback manner to regulate surfactant synthesis and secretion (24, 86). The primary structures of human (31, 113), dog (11), rabbit (16), and rat (91) SP-A, determined by sequencing of complementary DNA (cDNA) clones, are found to be highly conserved and are comprised of 247-248 amino acids. SP-A can be subdivided into two distinct domains; the amino-terminal third of the protein is collagen-like (1 13), while the carboxy-terminal two thirds has properties of a lectin (25). Surfactant also contains several extremely hydrophobic polypeptides (Mr 5-18,000) or proteolipids that remain associated with the glycerophospholip ids during organic solvent extraction. Two proteolipids, termed SP-B (35, 45, 48, 121) and SP-C (28, 34, 112), which have been isolated and characterized, are derived from two different precursor molecules by proteolytic cleavage at both amino- and carboxy-termini. The proteolipid derived from the SP-B precursor (Mr 40-42,000) has an apparent molecular weight of 18,000 in the non-reduced and 7,000 in the reduced form; whereas the proteolipid derived from the SP-C precursor (Mr 22,000) has an apparent molecular weight of 10,000 in the non-reduced and 5,000 in the reduced form. SP-C contains a unique polyvaline sequence (34) and two palmitic acids covalently linked to cysteine residues at the amino-terminus of the mature protein (23) that contribute to its extremely hydrophobic properties. These low molecular weight hydrophobic proteins markedly enhance the surface tension-lowering properties of surfactant glycerophospholipids (71, 107, 115, 122). The im portance of these hydrophobic polypeptides is suggested by the findings of certain clinical trials, in which surfactant replacement therapy using bovine surfactant extracts containing these proteins was found to be more efficacious in the prevention and treatment of RDS in prematurely born infants than are synthetic phospholipid mixtures (70). Recently, another structurally-unique surfactant-associated protein, termed SP-D (Mr 43,000; previously designated as CP4), has been identified (79). The complete primary sequence of SP-D has not as yet been determined. It is apparent, however, that, like SP-A, SP-D is a glycoprotein comprised of a collagen-like domain containing hydroxyproline residues that is secreted from


�

�

�

�

�

422

MENDELSON

&

BOGGARAM

type II cells as a multimeric complex held together in part by disulfide bonds

(79). Also, like SP-A, SP-D has calcium-dependent carbohydrate binding properties (78).


Developmental Regulation in vivo Expression of the SP-A gene is developmentally-regulated in fetal lung tissue. SP-A mRNA is undetectable in lung tissues of human abortuses at 16-20 weeks of gestation (7, 103). Differentiated type II cells containing few lamellar bodies can be observed in human fetal lung tissue as early as 22 weeks gestation; however, active secretion of surfactant occurs only after 30 weeks at which time SP-A can be detected in the amniotic fluid (50, 53, 56, 63,� 94, 100). The levels of SP-A in amniotic fluid continue to increase throughout the remainder of gestation in association with an increase in the levels of surfactant glycerophospholipids. No significant differences were observed in amniotic fluid levels of SP-A on the basis of fetal sex (100). This is of interest in light of the increased risk of RDS in male as compared to female newborns (68) and apparent delay in lung maturation in males as reflected by a reduced LIS ratio and DPPC content of amniotic fluid (108). In the rat, levels of SP-A mRNA and protein, which are first detectable on day 18 of gestation, increase markedly through day 21 to approximately 50% of adult levels, decline moderately during the first week of life, and then increase to adult levels by day 28 (93). No sex differences in the levels of SP-A or its mRNA were observed in fetal or adult lung tissues. In the rabbit, SP-A gene expression is initiated in fetal lung tissue several days prior to the time when augmented surfactant glycerophospholipid syn thesis occurs. SP-A gene transcription and mRNA levels are first detectable on day 24 of the 31 day gestation period (Figure l) (15). SP-A gene transcrip tion attains maximum levels by day 28 of gestation and then decreases slightly in the neonate; the levels of SP-A mRNA reach a maximum by day 30-31 and then decline somewhat after birth (16, 64). Parallel changes in the levels of immunoreactive SP-A in the fetal rabbit lung tissue also are observed. Im munoreactive protein, which is first detectable on day 24 of gestation, reaches peak levels by day 30, and declines modestly after birth (102). During gestation there are changes in the apparent molecular weight of the protein, which appear to result from changes in glycosylation state. In lung tissue homogenates from 24-28 day gestational age fetal rabbits, immunoreactive SP-A is present as a 29,000 Mr species, pI :5 5.6; whereas, in lung homogen ates from 30 day fetal rabbits, neonates and adults, the major immunoreactive species is the fully glycosylated, 29-36,000 Mr form of the protein (102). Only the fully glycosylated form of SP-A is detectable in lamellar bodies isolated from lung tissues of 28-31 day fetal rabbits, neonates and adults,

C,estat ional


211

Age (days)

24

26

•

• •

28

Neo prSP-A

"'-\\//

kb

.... 3.0 ..

Figure 1

....

2.0

Changes in SP-A gene transcription and mRNA levels in rabbit lung tissue during

development. Upper Panel: Transcriptional activity of the SP-A gene was assessed by transcrip tion run-on a�alysis using nuclei isolated from lungs of fetal rabbits of 21-28 days gestational age and from neonates. Lower Panel: SP-A mRNA was analyzed by Northern blotting of total RNA isolated from the same lung tissues using a homologous 32P-labeled SP-A eDNA insert. From Boggaram & Mendelson (15) with permission.

which suggests a role of posttranslational modification in transport of SP-A to the lamellar body (102). The time of initiation of SP-A gene expression in fetal lung tissue is correlated with the appearance of identifiable type II cells. It is uncertain, however, as to whether subsequent in vivo changes in SP-A mRNA levels are the result of an increase in SP-A gene expression per cell, or to increased numbers of type II cells, or to both. In human fetal lung tissue, initiation of expression of the genes encoding SP-B and SP-C occurs at a much earlier time in development than is the case for SP-A. mRNAs for SP-B and SP-C are detectable in human fetal lung as early as 13 weeks of gestation (60, 117) and continue to increase during development, so that by 24 weeks, the levels of SP-B and SP-C mRNA are

50% and 15%, respectively, of the adult levels (60). In other mammalian species, this discrepancy in timing of developmental expression of genes encoding the precursors of the surfactant proteolipids (SP-B and SP-C) and

SP-A is not as apparent. In rats, it was observed that SP-B mRNA was first detectable in fetal lung tissue on day 18 of gestation (as is the mRNA for SP-A); whereas, the mRNA for SP-C was readily detectable as early as day 17

424



(93). In contrast to SP-A mRNA, which only reaches adult levels by postnatal day 28, the mRNAs for SP-B and SP-C were found to attain adult levels by day 20 of gestation (93). In studies of developmental expression of the SP-B gene in fetal rabbit lung tissue, Xu et al (121) reported that the levels SP-B mRNA, which were first detected in tissues of 27 day fetal rabbits, were increased on day 30 and declined slightly after birth. The SP-A gene appears to be expressed exclusively in lung tissue (16,30). By use of immunocytochemistry, SP-A has been localized to the type II cell, to nonciliated bronchiolar epithelial cells or Clara cells, and to alveolar macrophages (110, 119). Macrophages do not synthesize SP-A, but avidly take up the protein within the lumen of the alveolus. The role of the Clara cell in SP-A synthesis and metabolism remains uncertain. The results of in situ hybridization studies using human lung tissues indicate that the SP-A gene is expressed only in the type II pneumonocyte (81); however, in 31-day gesta tional age fetal and adult rabbits, the SP-A gene also appears to be expressed in columnar bronchiolar epithelial cells of proximal and distal airways (4), albeit at lower levels than in type II cells. Through in situ hybridization, SP-B mRNA has been identified in type II as well as in bronchiolar epithelial cells (81). The identity of the cells in which the SP-B and SP-C genes are expressed in human fetal lung prior to the appearance of differentiated type II cells is, at present, uncertain. The recent finding that immunoreactive SP-B is associated with surfactant-like particles secreted by intestinal enterocytes (27) suggests that SP-B gene expression may not occur exclusively in lung tissue.

Developmental Regulation in vitro The spontaneous differentiation of fetal lung explants maintained in organ culture in serum-free medium is associated with a rapid induction of SP-A gene transcription and of the levels of SP-A mRNA and protein (15, 16, 64). In studies using fetal rat lung in organ culture, it was observed that the glucocorticoid receptor antagonist, RU 486, failed to block the spontaneous changes in morphology and increases in DPPC synthesis and SP-A mRNA accumulation (40). These findings suggest that spontaneous differentiation is not induced by the action of glucocorticoids retained within the cultured tissue (40). In recent studies, we found that human fetal lung tissue in organ culture produces large amounts of prostaglandin E2 (PGE2) (1). The prostaglandin synthesis inhibitor, indomethacin, was found to prevent the spontaneous induction of SP-A gene expression and markedly reduce cAMP formation by the cultured lung tissue (1). The finding that PGE2 can markedly increase cAMP formation by the cultured tissue and that either PGE2 or cAMP analogues could overcome the inhibitory effect of indomethacin on SP-A gene expression suggest that prostaglandins, acting through cAMP, may serve a



425

role in the spontaneous induction of SP-A gene expression in human fetal lung in vitro (1). As discussed above, SP-B and SP-C mRNAs are detectable in human fetal lung tissue as early as 13 weeks of gestation; the levels of SP-B mRNA increase in human fetal lung explants as a function of time in organ culture and reach adult levels after several days, whereas, the levels of SP-C rnRNA decline as compared to preculture values (60). These findings are indicative that the genes encoding SP-A, SP-B and SP-C are independently regulated in human fetal lung tissue.

Regulation by Hormones Although it is apparent that midgestation fetal lung explants differentiate spontaneously and develop the capacity to synthesize surfactant glycerophos pholipids and proteins when placed in culture in serum-free medium, there is abundant evidence that hGilUOnes and bioactive substances can modulate the rate of biochemical and morphological differentiation. EFFECTS OF cAMP ANALOGUES AND OF AGENTS THAT INCREASE THE CELLULAR LEVELS OF CAMP ON SP-A SYNTHESIS AND GENE EXPRESSION

Cyclic AMP greatly enhwlces SP-A synthesis and rnRNA levels, as well as morphologic development of fetal lung tissues maintained in organ culture (16, 64, 74, 116). In rabbit fetal lung in vitro, SP-A synthesis is augmented by cAMP analogues and by agents that increase the accumulation of cAMP, such as isobutylmethylxanthine, which inhibits phosphodiesterase activity, and fors kolin, which activates adenylyl cyclase directly (64). The cAMP induction of SP-A synthesis is associated with a rapid increase in SP-A rnRNA levels (16, 64) and a comparable increase in SP-A gene transcription (15). The stimula tory effects of cAMP on SP-A gene transcription and rnRNA levels are dependent upon ongoing protein synthesis, which suggests that a labile protein factor(s) mediates the stimulatory effects of \-AMP on SP-A gene expression (15). Cyclic AMP analogues also increase the levels of SP-A (74) and its rnRNA (74, 116) in human fetal lung in culture. An autoradiogram of an immunoblot of the levels of immunoreactive SP-A in human fetal lung explants after two to six days of organ culture in the absence or presence of dibutyryl cAMP (Bt2cAMP) is shown in Figure 2. Immunoreactivy SP-A was first detectable in control tissues on day four of incubation and was present in increased amounts on day six. The spontaneous induction of SP-A synthesis in the human fetal lung explants was associated with an induction of surfactant glycerophospholipid synthesis, me enlargement of I le prealveolar ducts, and me appearance of differentiated type II cells. Bt.tc \hlP treatment increased AND ON MORPHOLOGIC DEVELOPMENT OF FETAL LUNG

426


C

C

C kDa


-92

-66 -45

-31 -21

t- d2

-t

f-

d4

-t

t-

d6

-t

Days in Culture Figure 2 Effect of Bt2cAMP on the levels of immunoreactive SP-A in human fetal lung in culture. Lung explants from midtrimester human abortuses were maintained in organ culture for up to six days in control medium (C) or in medium containing Bt2cAMP (Bt2• I mM). Specific content of immunoreactive SP-A was analyzed in equal amounts of homogenate protein (40 /Lg) by immunoblotting using specific antibodies against human SP-A. SP-A is the major band at 35 kd. From Odom et al (74) with permission.

the levels of immunoreactive SP-A on days four and six of incubation. The cAMP induction of SP-A accumulation in the fetal lung explants was associ ated with comparable increases in the levels of SP-A mRNA (74) and of SP-A gene transcription (17). The stimulatory effects of Bt2cAMP on SP-A gene expression in human fetal lung in vitro are associated with pronounced effects on morphology (74). In human fetal lung tissues incubated for two days in Bt2cAMP-containing medium, a marked enlargement of the prealveolar ducts and decrease in the volume density of the interalveolar connective tissue was found, as compared to fetal lung tissues maintained for this period in control medium (Figure 3A,B). These differences between control and Bt2cAMP-treated tissues were no longer evident after four and six days of incubation because the alveolar lumenal volume density of control explants was increased at these time points (Figure 3B). In cAMP-treated fetal lung explants there was an enhanced rate of appearance of differentiated type II cells (Figure 3B) and an increased


80

OCOII/.01 Oe'rAMl'

40 30 20 10 O��-L--��--��

S

e �"'

60

;3� 40 t:I.g

!;at 20 0

Doy2

Day 4

Day 6

Figure 3 Effect of Bt2cAMP on morphological development uf human fetal lung in culture. Upper panel: Light micrographs (X340) of human fetal lung explants after two days of culture in control medium (C) or in medium containing Bt2cAMP. Lower panel: Morphometric analysis at the light and electron microscopic levels of human fetal lung explants incubated for two to six days in the absence or presence of Bt2cAMP. Data are the mean

± SEM (n

=

6).

*

=

P < 0.001.

Six explants from two independent experiments were analyzed for each treatment at each time point to determine the relative volume density of the alveolar lumen

(upper graph); each 300 grid intersections scored for each explant. The proportion of type II cells (lower graph) was determined by ultrastructural analysis; 300 epithelial cells were scored as type II cells (one or more lamellar bodies) or as undifferentiated cells (no

determination represents pooled data from

lamellar bodies) per treatment and time point in two independent experiments. From Odom et al

(74) with permission.


428


accumulation of secreted lamellar bodies and tubular myelin within the lumina of the prealveolar ducts (74). There is evidence that catecholamines, acting through l3-adrenergic recep tors and cAMP, may serve a role in the regulation of SP-A gene expression in fetal lung tissue during development. The l3z-adrenergic agonist, terbutaline, was found to increase the accumulation of immunoreactive SP-A in human fetal lung in vitro (74). I3-Adrenergic receptors have been identified in fetal lung tissues (32, 88, 114) and appear to be concentrated on type II cells (88). The specific content of such receptors, as well as the responsiveness of adenylyl cyclase to catecholamines, is increased in fetal rabbit lung tissue with advancing gestational age (9). Receptor number also is increased in response to cortisol treatment (20). The findings that norepinephrine levels in human fetal plasma increase markedly during late gestation (77) and that administration of l3-adrenergic agonists as tocolytic agents to women in preterm labor decrease the incidence of RDS in their premature infants (13, 18, 51) are further suggestive of the importance of the adrenergic system in fetal lung maturation and surfactant synthesis. EFFECTS OF GLUCOCORTICOIDS ON SP-A GENE EXPRESSION AND ON DEVELOPMENT OF FETAL LUNG Glucocorticoids have complex actions on SP-A gene expression in fetal lung tissues that may be species-specific and dependent upon the stage of development at which treatment is initiated. In lung explants from 21-day fetal rabbits, glucocorti coids have acute but transient effects to inhibit SP-A gene transcription and mRNA levels that are followed by a stimulatory action on SP-A gene expres sion (15). Administration of dexamethasone to fetal and neonatal rats has been reported to enhance SP-A synthesis and mRNA levels in a dose dependent manner (29, 80). No significant differences in responsiveness to dexamethasone treatment were observed as a function of postnatal age, although a trend toward decreased steroid responsiveness with increasing age was noted (29). In human fetal lung in vitro, glucocorticoids have been reported to exert both stimulatory and inhibitory effects on the levels of SP-A and its mRNA that are dose- and time-dependent (7, 17, 59, 73, 116). We examined in detail the effects of dexamethasone in various concentrations on SP-A gene expres sion and on morphologic development of human fetal lung in vitro. Dexa methasone was found to have differential effects on the levels of SP-A and its mRNA in human fetal lung tissue that are dose-dependent; at concentrations of 10-10 and 10-9 M, a stimulatory effect was observed, while at con centrations > 10-8 M, the glucocorticoid was markedly inhibitory (73). De xamethasone (10-7 M) also antagonized the action of Bt2cAMP to increase the levels of SP-A and its mRNA. In recent studies, we observed that the MORPHOLOGIC


C

429

Oexamethasooe 1(1 cr-1ctl

Hormonal effects on the surfactant protein B (SP-B) mRNA in cultured fetal rat lung.

Hormonal and developmental regulation of pulmonary surfactant synthesis in fetal lung.

Mature Surfactant Protein-B Expression by Immunohistochemistry as a Marker for Surfactant System Development in the Fetal Sheep Lung.

Fetal mouse lung in circumfusion system cultures.

Surfactant protein C: hormonal control of SP-C mRNA levels in vitro.

Effect of metopirone on the synthesis of lung surfactant in does and fetal rabbits.

[Induction of surfactant formation in the fetal lung using bromhexine metabolite VIII].

Posttranscriptional regulation of surfactant protein-A messenger RNA in human fetal lung in vitro by glucocorticoids.

Fetal lung development and the influence of glucocorticoids on pulmonary surfactant.

Prostaglandins regulate surfactant protein A (SP-A) gene expression in human fetal lung in vitro.

Control of lung surfactant by ventilation, adrenergic mediators, and prostaglandins in the rabbit.

The hormonal control of regeneration in plants.

The surfactant system of the adult lung: physiology and clinical perspectives.

Hormonal control of gluconeogenesis.

Interferon-gamma and synthesis of surfactant components by cultured human fetal lung.

Insulin regulation of messenger ribonucleic acid for the surfactant-associated proteins in human fetal lung in vitro.

Regulation of lung surfactant secretion.

Levels of surfactant-associated protein messenger ribonucleic acids in rabbit lung during perinatal development and after hormonal treatment.

Isolation and characterization of proteins associated with the lung surfactant system.

The hormonal control of wheat leaf unrolling.

Studies on the fate of pulmonary surfactant in the lung.

Hormonal regulation of the hypothalamic melanocortin system.

Hormonal control of gastrointestinal motility.

Hormonal control of gastrointestinal motility.