Main

Alveolar epithelial type II cells synthesize and secrete pulmonary surfactant, a complex of lipids and proteins (SP-A, SP-B, SP-C, and SP-D)(1–4). Pulmonary surfactant maintains alveolar stability and normal lung function, and its deficiency in the newborn often leads to respiratory distress syndrome. The respiratory distress syndrome is a major cause of infant mortality and morbidity(5), and is more common among prematurely delivered infants because of pulmonary immaturity.

A variety of circulating substances and the mesenchymal-epithelial interactions influence lung development(6). Circulating glucocorticoids have profound effects on fetal lung development, such as those seen in mice with ablated GR or ACTH genes(7, 8). The GR knockout animals develop atelectasis of the lung and die shortly after birth(7). The ACTH knockout animals have features of depressed lung maturation, which can be reversed by prenatal administration of glucocorticoids(8). Antenatal glucocorticoid administration in humans promotes fetal lung maturation and reduces the incidence of respiratory distress syndrome and mortality(9). Exogenous glucocorticoids are known to accelerate morphologic and biochemical maturation of the lung during late stages of gestation [reviewed in Ballard and Ballard(9) and Odom and Ballard(10)], but relatively fewer studies have examined the effects of glucocorticoids on the early embryonic lung. Jaskoll and co-workers(11–14) have reported that treatment of early embryonic mouse lung in vitro with a glucocorticoid agonist, TAC, enhanced morphodifferentiation, and precociously increased SP-A expression. They have also shown that glucocorticoids regulate the expression of several growth factors (tumor necrosis factor-α and transforming growth factors β2 and β3)(12–14), which may mediate some of the effects of glucocorticoids on morphology and differentiation of embryonic and fetal lungs.

Glucocorticoids have the potential of directly influencing the transcription of genes related to epithelial growth and differentiation during lung development. They may also act through paracrine mediators. Fibroblast-pneumonocyte factor, a paracrine mediator produced by fetal lung fibroblasts in response to cortisol, stimulates the synthesis of pulmonary surfactant by alveolar type II cells(15). Other secretory products of mesenchymal cells, such as KGF and HGF, have been implicated in the development of a variety of organs and may play similar roles during lung development(16–19). HGF, KGF, and their respective receptors, MET and KGFR, are highly expressed in fetal and adult lungs(16, 17, 20–22). Both growth factors are potent mitogens of lung epithelial cells(16, 20, 23–25) and may participate, as paracrine mediators, in the action of glucocorticoids on epithelial cells. KGF, in particular, has potent effects on lung development. Pulmonary targeted expression of KGF or of dominant-negative KGFR in transgenic mice resulted in abnormal lung development(26, 27). In cultures of fetal rat lung explants, KGF inhibited branching and induced cystic development(21, 28). KGF mediates the action of steroids on the development of other organs(18, 19) and may play a similar role during the glucocorticoid-induced maturation of the lung.

In this study, we investigated whether DEX, a synthetic glucocorticoid, stimulates maturation of the early embryonic rat lung and regulates the expression of KGF and HGF, the potential paracrine mediators of glucocorticoid action. DEX treatment of fetal rat lung explants was initiated during the early pseudoglandular stage of development and was shown to induce features of both distorted and accelerated maturation. The adverse effects of the treatment include dilatation of the proximal tubules and an abnormal branching pattern. A feature of the accelerated maturation was the increased expression of several genes related to epithelial differentiation (SP-A, SP-B, SP-C, and CC10) and growth (KGF, KGFR, and MET). In a companion study(51), we describe several antagonistic effects of retinoic acid on DEX-induced distorted and accelerated lung development in vitro.

METHODS

Materials. The materials and their suppliers were as follows: DEX, streptomycin, gentamicin, triethanolamine, acetic anhydride, heparin, and RNase from Sigma Chemical Co. (St. Louis, MO); OCT compound from Miles(Elkhart, IN); glutaraldehyde and paraformaldehyde from Electron Microscopy Science (Ft. Washington, PA); Millicell-PCF assemblies with polycarbonate isopore membranes from Millipore (Bedford, MA); penicillin G, amphotericin B, DMEM, fetal bovine serum, Trizol reagent, and Fungizone from Life Technologies, Inc. (Gaithersburg, MD); proteinase K and random-primed DNA labeling kit from Boehringer Mannheim (Indianapolis, IN); GeneScreen Plus nylon membrane and [33P]UTP from NEN Research Products (Boston, MA);[3H]thymidine from ICN (Irvine, CA); Riboprobe Gemini Core System II transcription kit from Promega (Madison, WI).

Preparation of organ cultures. Timed-pregnant rats of the Sprague-Dawley strain were obtained from Hilltop (Scottdale, PA). Lungs were dissected from d 14 embryos under a stereomicroscope (Olympus SZ40, Tokyo, Japan). At this gestational age, the lung is in the pseudoglandular stage of development and has 7-10 branches. Lungs from same litters were divided into control and experimental groups, and cultured on polycarbonate isopore membranes in Millicell-PCF assemblies. Culture medium consisted of DMEM with 10% fetal bovine serum, streptomycin/penicillin G (0.1 mg/mL), gentamicin (10μg/mL), and Fungizone (2.5 μg/mL). A stock solution of DEX was prepared in ethanol, and aliquots of the solution were added to explant cultures. Control cultures contained the same volume of ethanol (0.01-0.02%) as was used in DEX-treated cultures. Cultures were maintained at 37 °C in 5% CO2/95% air for 4 d and fed daily with a complete change of the medium. Lung explants were photographed daily under a stereomicroscope. DEX was used at 10-9 to 10-6 M concentrations. Similar concentrations have been previously used to examine the effects of DEX on lung development in vitro(29–32).

Hybridization probes. Probes (CC10, SP-A, SP-C, HGF, MET, KGF, and KGFR) for Northern blot analysis were obtained as previously described(21). SP-B probe was generated by reverse transcriptase-PCR using primers (sense, 482-502; antisense, 790-810) based on the published nucleotide sequence of rat SP-B(33).32 P-Labeled probes were prepared using a random primed labeling kit. Sense and antisense RNA probes for in situ hybridization were prepared from CC10 cDNA linearized with SmaI or ThaI, respectively. SP-C cDNA (427 bp), with flanking SP6 and T3 RNA polymerase promoter sequences, was prepared by reverse transcriptase-PCR(21). 33P-Labeled antisense and sense riboprobes were synthesized using SP6 and T3 RNA polymerases and a Riboprobe Gemini Core System II transcription kit.

Northern blot analysis The protocols for the isolation of RNA and Northern blot analysis were previously described(20, 21). Briefly, cultured lungs with attached membranes were removed and frozen in liquid nitrogen. Several pools of three lungs each were used for RNA extraction with Trizol reagent. RNA was electrophoresed in 1% agarose gel containing 0.66 M formaldehyde. Gels were photographed under UV light and treated with 0.05 N NaOH, before RNA transfer to Gene Screen Plus with a Transvac Blotting System (Hoefer Scientific, San Francisco, CA). Blots were exposed to UV light in a UV Stratalinker(Stratagene, San Diego, CA) and sequentially hybridized with different probes as previously reported(20, 21). Before hybridization with a new probe, the blot was stripped of radioactivity by boiling in 0.1 × SSC, 1% SDS. Densitometric analyses of the autoradiograms were done on a BioImage analytical scanning densitometer(Millipore/BioImage, Bedford, MA) or on an AlphaEase digital imaging and analytical system (α Innotech, San Leandro, CA).

In situ hybridization. In situ hybridization was done as previously described(20, 21). Briefly, lungs were removed from polycarbonate membranes after 4 d of culture, placed in OCT compound, and frozen. Cryosections (10 μm thick) were cut, placed on aminopropyltriethoxysilane-coated slides, and stored at -80 °C. Before hybridization, tissue sections were fixed in PFA for 20 min, washed in PBS, and treated sequentially with proteinase K (1 μg/mL) and 0.25% acetic anhydride in 0.1 M triethanolamine, pH 8. Hybridization, washes, and emulsion autoradiography were done as described by McLaughlin and Margolskee(34).

DNA determination. Two or three lungs were pooled after 4 d of culture and homogenized in 1 × SSC (150 mM NaCl, 15 mM sodium citrate), 5 mM EDTA, 0.2% SDS. DNA concentrations were determined as described by Labarca and Paigen(35).

[3H]Thymidine nuclear incorporation and autoradiography. [3H]Thymidine (35 μCi/mL) was added to cultures 4 h before harvesting of explants on d 4. Explants were then washed in PBS, fixed in 4% PFA for 20-30 min, and mounted in OCT compound. Cryosections (8 μm) were cut, placed on slides, and fixed in PFA for 20 min. Slides were washed extensively with PBS, dehydrated, and coated with a photographic emulsion (NTB2). The emulsion was developed after 1-2 wk of exposure at 4 °C and counterstained with hematoxylin.

Light and electron microscopy. After 4 d of culture, fetal lung explants were fixed in 2% glutaraldehyde in cacodylate buffer and Epon-embedded. Tissue sections were stained with toluidine blue. Ultrathin sections were processed for electron microscopy and viewed and photographed under an electron microscope (Philips CM12).

RESULTS

Effects of DEX on branching morphogenesis. The morphologic effects of increasing concentrations of DEX (0, 1, 10, 50, 100, and 200 nM) are shown in Figure 1A. To simplify description of the results, an explant is divided into two major zones (Fig. 1B). Zone I consists of the proximal tubules, which form future conducting airways and zone II consists of the distal tubules and buds, which are destined to become respiratory airways. DEX treatment caused dilatation of some tubules of zone I (zone IB), whereas the more proximal tubules of zone I, which represent prospective trachea and main stem bronchi (zone IA), were not affected. Because of tubular dilatation in zone IB, the distal tubules/buds of zone II were thin and closely packed at the periphery of the explant and in the region between the dilated tubules. The overall branching pattern was distorted in DEX-treated explants. Figure 2 shows the development of lung explants during 1-4 d of culture in the presence or absence of DEX (100 nM). The DEX-induced tubular dilatation was apparent only after 3 d of culture.

Figure 1
figure 1

(A) Effect of varying concentrations of DEX on lung branching morphogenesis. Lungs from d 14 rat fetuses were cultured for 4 d in the presence of DEX (1-200 nM, labeled as DX1-DX200) or its absence (DMEM + ethanol, labeled as Cont.) DEX treatment caused dilatation of some proximal tubules (arrows). The most proximal tubules, which represent prospective trachea and main stem bronchi, were not affected. As a result of the dilatation of proximal tubules, the distal tubules and buds were thinner and closely packed at the periphery of the explant (arrowheads) and in the region between the dilated tubules.(B) To further simplify the description of the results, control(a) and DEX-treated (c) explants were divided into zones. Control explants (a) were divided into two major zones (shown in b). Zone I consists of the proximal tubules, which form future conducting airways; and zone II consists of the distal tubules and buds, which are destined to become respiratory airways. In explants treated with 100 nM DEX (c), zone I was further subdivided into two zones based on the susceptibility to DEX (shown in d): zone 1A (filled in black), the most proximal tubules of zone I, which represent prospective trachea and main stem bronchi, were not affected by DEX; and zone IB (filled in white), the more peripherally located proximal tubules were dilated as a result of DEX treatment. Because of tubular dilatation in zone IB, the tubules of zone II(filled in gray) were thin and closely packed at the periphery of the explant and in the region between the dilated tubules.

Full size image
Figure 2
figure 2

Branching morphogenesis of lungs in culture for 1-4 d. Day 14 fetal rat lungs were cultured from 1 to 4 d in the absence(CON) and presence of 100 nM DEX. DEX treatment caused dilatation of some proximal tubules (arrows) after 3-4 d of culture. The most proximal tubules, which represent prospective trachea and main stem bronchi, were not affected. As a result of tubular dilatation, branching was distorted, and the distal tubules and buds became thin and closely packed at the periphery of the explant (arrowhead) and in the region between the dilated tubules.

Full size image

Light and electron microscopy. A light microscopic examination of the control explant revealed that the epithelium lining the proximal tubules was pseudostratified and had a smooth luminal surface (Fig. 3C). Some of the mesenchymal cells were concentrically arranged to form rudimentary cartilage (Fig. 3C). A large number of distal tubules with narrow lumens were present (Fig. 3,E and G). The epithelial cells lining the distal tubules were columnar (Fig. 3,E and G). Loosely arranged mesenchymal cells surrounded the distal tubules (Fig. 3,E and G). In DEX (100 nM)-treated explants, the epithelium lining the proximal tubules was pseudostratified and had an uneven luminal surface due to protruding epithelial cells (Fig. 3D). Rudimentary cartilage was also present (Fig. 3D). The dilated tubules often extended to the periphery (Fig. 3F). Rudimentary septa (Fig. 3H), large airspace, attenuated mesenchymal tissue, and reduced overall cellularity were present in treated explants (Fig. 3,F and H). The epithelial lining of the distal tubules consisted of cuboidal or flattened cells (Fig. 3H).

Figure 3
figure 3

Light microscopic examination of DEX-treated and control lung explants. Day 14 fetal rat lungs were cultured for 4 d in the presence (B, D, F, and H) or absence (A, C, E, and G) of 100 nM DEX. Tissue sections were stained with toluidine blue. In control explants, epithelium lining the proximal tubules was pseudostratified, and had a smooth luminal surface (C, black arrow). Some of the mesenchymal cells were concentrically arranged to form rudimentary cartilage (C, red arrow). A large number of distal tubules with narrow lumens was present (E and G). The epithelial cells lining the distal tubules were columnar (G, arrows). Loosely arranged mesenchymal cells surrounded the distal tubules (E and G). In DEX-treated explants, the epithelium lining the proximal tubules was pseudostratified and had uneven luminal surface due to protruding epithelial cells (D, black arrow). Rudimentary cartilage was also present (D, red arrow). The dilated tubules often extended to the periphery (F). Rudimentary septa (H, black arrow), large airspace, attenuated mesenchymal tissue, and overall reduced cellularity were all present in treated explants (F and H). The epithelial lining of the distal tubules consisted of cuboidal (H, red arrows) or flattened (H, arrowhead) cells. Bar: A and B, 500 μm; C-F, 100 μm; G and H, 50 μm.

Full size image

An electron microscopic examination of control explants revealed that the epithelial cells of distal tubules were columnar with occasional lamellar bodies (Fig. 4A). In contrast, the distal epithelium of DEX-treated explants often consisted of cuboidal cells containing lamellar bodies and abundant glycogen (Fig. 4B).

Figure 4
figure 4

Electron microscopy of DEX-treated and control lung explants. Day 14 fetal rat lungs were cultured for 4 d in the presence(B) or absence (A) of 100 nM DEX. In control explants, the epithelial cells lining the distal tubules were columnar and infrequently contained lamellar bodies (arrow) and glycogen(arrowhead). In DEX-treated explants, the epithelial cells lining the distal tubules were generally cuboidal and often contained lamellar bodies(white arrow) and abundant glycogen (black arrow). Magnification: A, ×1600; B, ×1950.

Full size image

Thymidine incorporation and emulsion autoradiography. Nuclear incorporation of thymidine was examined in tissue sections of control and DEX-treated explants. Fewer labeled cells were present in DEX-treated lungs compared with control lungs (Fig. 5,A and B). Labeling was more extensive in the peripheral regions of both control and DEX-treated explants. Considerable variability existed in the labeling of various distal tubules. We determined a labeling index by counting a minimum of 400 epithelial cells of the distal tubules. The labeling index (%) in control and DEX-treated explants was 51 ± 6.4 and 12 ± 1.3 (values are mean± SEM; n = 2), respectively. In addition, DEX-treated lungs had less cellularity compared with control lungs (Fig. 5,A and B). The reduced cellularity was further reflected in DNA content of the explants (μg/lung): DEX, 17.3 ± 1.69; control, 25.4 ± 2.17(values are mean ± SD; n = 5; p < 0.001).

Figure 5
figure 5

Thymidine incorporation and emulsion autoradiography. Day 14 fetal rat lungs were cultured for 4 d with (B) or without(A) 100 nM DEX. Four hours before harvesting, [3H]thymidine was added to explant cultures. Tissue sections of cultured explants were subjected to emulsion autoradiography, followed by hematoxylin staining. Labeling was more extensive in the peripheral regions of control and DEX-treated explants. The number of labeled epithelial cells in distal tubules was about 4-fold higher in control compared with DEX-treated explants.

Full size image

Northern blot analysis. Expression of several genes related to epithelial cell differentiation (SP-A, SP-B, SP-C, and CC10) and growth (KGF, HGF, KGFR, and MET) was examined in explants treated with or without 100 nM DEX for 4 d. Of those examined, HGF and KGF genes are expressed in stromal cells, whereas the remaining are expressed in alveolar and/or bronchial epithelial cells. The results of Northern blot analysis and corresponding densitometric quantitation are shown in Figures 6 and 7, respectively. DEX markedly increased the levels of SP-A, SP-B, SP-C, CC10, KGFR, and MET transcripts (Figs. 6A and 7). Large increases were noted in the expression of SP-A (15-fold), SP-B(40-fold), and CC10 (8-fold) mRNAs. KGF mRNA increased (4-fold), whereas HGF mRNA decreased (2-fold) in explants treated with DEX (Figs. 6B and 7). DEX did not affect the levels of actin mRNA (Figs. 6B and 7).

Figure 6
figure 6

Northern blot analysis. Lungs were cultured for 4 d in the presence (D) or absence (C) of 100 nM DEX. Total RNA was extracted from four pools of three lungs each (treated and control groups), and electrophoresed in 1% agarose-formaldehyde gel. RNA was then transferred to GeneScreen Plus nylon membrane. The same blot was sequentially hybridized with different probes. (A) Genes related to epithelial differentiation (SP-A, SP-B, SP-C, and CC10). (B) Genes related to epithelial growth (KGF, KGFR, HGF, and MET) and actin. The portion of the gel(GEL) showing bands of 28 and 18 S ribosomal RNA is included in A.

Full size image
Figure 7
figure 7

Densitometric quantitation of the blot shown in Figure 6. Relative scan values are mean ± SD(n = 4). Differences between control (C) and DEX-treated groups are statistically significant (p < 0.05). The levels of SP-A, SP-B, SP-C, CC10, KGF, KGFR, and MET transcripts were markedly increased by DEX. Large increases were noted in the expression of SP-A (15-fold), SP-B(40-fold), CC10 (8-fold), and KGF (4-fold) mRNAs. DEX reduced the levels of HGF mRNA (2-fold).

Full size image

In situ hybridization. The distribution of CC10 and SP-C transcripts was examined by in situ hybridization to differentiate prospective bronchial and alveolar epithelial cells in branching tubules. CC10 and SP-C are often used as markers of Clara cells and type II cells, respectively. In the rat, although the expression of SP-C mRNA is specific to type II cells, the latter also weakly express CC10 transcripts(21). In both control and treated explants (Fig. 8,A-D), CC10 mRNA was strongly expressed in the proximal tubules and weakly expressed in the distal tubules. In DEX treated explants (Fig. 8,C and D), the dilated tubules had a strong but variable expression of CC10 transcripts, suggesting that these tubules in part represent future conducting airways. The distribution of SP-C transcripts is shown in Figure 9. To show possible co-localization of SP-C and CC10 transcripts in the same tubules, in situ hybridization was done on serially cut sections (shown in Figs. 8 and 9). In control lungs, SP-C transcripts were not present in the epithelium of proximal tubules, which, instead, had a strong expression of CC10 transcripts (compare Figs. 8 and 9). The same distal tubules with weaker expression of CC10 transcripts (Fig. 8,A and B), strongly expressed SP-C transcripts (Fig. 9,A and B). In DEX-treated explants, many of the tubules with a strong CC10 mRNA expression (Fig. 8,C and D) lacked SP-C transcripts (Fig. 9,C and D). In contrast to the acinar distribution of cells expressing SP-C transcripts in control lungs, their distribution in DEX-treated lungs was patchy; SP-C transcripts were found in individual cells or clusters of cells. The patchy distribution of SP-C transcripts resembles that seen in more mature lungs.

Figure 8
figure 8

Distribution of CC10 transcripts by in situ hybridization. Lungs from d 14 fetuses were cultured for 4 d with (A and B) or without (C and D) 100 nM DEX. Tissue sections were subjected to in situ hybridization with a33 P-labeled CC10 RNA probe. In both the control and treated explants, CC10 mRNA was strongly expressed in the proximal tubules and weakly expressed in the distal tubules. In DEX-treated explants (C and D), the dilated tubules had a strong, but variable expression of CC10 transcripts. A an C, bright field; B and D, dark field.

Full size image
Figure 9
figure 9

Distribution of SP-C mRNA by in situ hybridization. Lungs from d 14 fetuses were cultured for 4 d with (A and B) and without 100 nM DEX (C and D). The sections were subjected to in situ hybridization with a33 P-labeled SP-C RNA probe. To demonstrate possible co-localization of SP-C and CC10 transcripts in the same tubules, serially cut sections were used for in situ hybridization (see also Fig. 8). In control lungs, SP-C transcripts were not present in the epithelium of proximal tubules, which, in contrast, strongly expressed CC10 transcripts (compare with Fig. 8). The same distal tubules with a weak expression of CC10 transcripts (see Fig. 8,A and B) strongly expressed SP-C transcripts (A and B). In DEX-treated explants, many of the tubules with a strong CC10 mRNA expression (see Fig. 8,C and D) lacked SP-C transcripts (C and D). In contrast to the acinar distribution of cells expressing SP-C transcripts in control lungs, their distribution in DEX-treated lungs was patchy; individual cells or clusters of cells expressing SP-C transcripts were seen. A and C, bright field; B and D, dark field.

Full size image

DISCUSSION

In this study, DEX treatment of fetal lung explants was initiated during the early pseudoglandular stage of lung development, which allowed monitoring of airway branching and of subsequent lung maturation. Lungs treated with 10 nM or higher concentrations of DEX showed morphologic features of both distorted and accelerated maturation. DEX-treated explants showed dilatation of proximal tubules and distorted branching. The relative position of a tubule within the branching tree determined its fate to undergo dilatation. Although the dilatation occurred mostly in proximal tubules, the most proximal tubules(destined to form main bronchi and trachea) remained unaffected by DEX.

In this study, we also observed several biochemical and morphologic features of accelerated maturation in explants treated with DEX: 1) the epithelial cells lining the distal tubules were often cuboidal or flattened; 2) the cuboidal cells often contained lamellar bodies and abundant glycogen; 3) rudimentary septa and large airspace were present; 4) mesenchymal cells were less in number and were compressed between adjacent epithelial tubules; 5) a more mature SP-C expression was seen, with individual and clusters of cells expressing SP-C mRNA; and 6) the expression of genes related to epithelial growth (KGF, KGFR, and MET) and differentiation (SP-A, SP-B, SP-C, and CC10) was precociously increased. These results show that DEX treatment of lungs during an early pseudoglandular stage accelerates the acquisition of several features of accelerated maturation that normally accompany late stages of development.

Some features of distorted and accelerated maturation described above have been previously reported by others. TAC, a more potent glucocorticoid agonist than DEX, has been previously used in cultures of fetal rat lungs (gestational d 15). Massoud et al.(36) reported that TAC at 10-8 M and 10-5 M distorted branching pattern and caused dilatation of bronchial tubules at 10-5 M. They also noted the formation of saccules and septa, and a decrease in the number of mesenchymal cells induced by TAC. TAC treatment of fetal mouse lungs at 10-8 M resulted in denser explants with blunter and coarser secondary and tertiary branching, and increases in SP-A expression(12). In addition, the number of epithelial branches and generation number were increased(13). TAC administration to rhesus macaques during the mid-pseudoglandular phase of lung development induced retradation of septal growth(37). Kaufmann(8) observed significant dose-dependent decreases in epithelial cell proliferation in fetal mouse lungs after transplacental exposure to DEX on d 17 of gestation. In the present study, DEX treatment of fetal rat lung explants in vitro also caused a decrease in the proliferation of distal epithelial cells. In agreement with the results of the present study, glucocorticoid-induced stimulation of surfactant protein expression was shown by others. In vitro exposure of fetal rat lungs to DEX (1-200 nM) increased the levels of SP-A, SP-B, and SP-C mRNAs(29, 30). The increased expression of surfactant proteins was also observed after glucocorticoid exposure in vivo(39, 40). Shellhase and Shannon(40) found that DEX administration to pregnant rats 1 or 3 d before delivery on d 17 of gestation during the pseudoglandular period of lung development resulted in precocious appearance of SP-A and SP-B mRNAs and an increased level of SP-C mRNA. Increased expression of SP-B and SP-C mRNAs after glucocorticoid treatment of lung explants from other species, including humans, has also been reported(9, 41). Thus, surfactant protein expression is clearly inducible by glucocorticoids during the early and late pseudoglandular stage of lung development. The treatment accelerated the progressive increases that normally occur in the expression of these genes with advancing gestational age. Consequently, the transcripts for SP-A, SP-B, and CC10, which normally surge during the canalicular stage of rat lung development, are precociously increased after DEX treatment.

Putative cis-acting response elements for GR binding have been identified in the regulatory region of CC10 gene(42), suggesting that the gene is inducible by glucocorticoids. Indeed, glucocorticoids stimulate the expression of CC10 transcripts in the developing lung both in vitro (this study) and in vivo(43, 44). In this study, we observed a 9-10-fold increase of CC10 transcripts induced by DEX in cultured lung explants. Maternal administration of betamethasone on gestational d 21-22 resulted in a 2-3-fold increase of CC10 mRNA in the lungs of newborn rats(43). Glucocorticoids also stimulated CC10 gene expression in the adult lung(44). The glucocorticoid-induced CC10 expression is significant in view of the suggested role of CC10 in host defense against infection(43, 45). The protein has both phospholipase A2 inhibitory and antiinflammatory activities(45) and may thus be important in preparing the neonatal lung to cope with infection. The ability to fight infection may be further enhanced by a glucocorticoid-stimulated expression of CC10.

In the present study, we have shown that DEX treatment caused dilatation of proximal tubules and altered gene expression in both epithelial and mesenchymal cells. These effects of DEX may result from: 1) a direct action of DEX on epithelial and mesenchymal cells and/or 2) an indirect action mediated by a paracrine factor(s) produced in response to DEX. The mode of action must be dictated by the spatial and temporal distribution of GR in the developing lung. Several investigators have studied the distribution of GR in fetal lungs. The binding sites for DEX increase during late stages of gestation(46). In the rat, the increase was 103% between gestational d 16 and 22(46). Beer et al.(47) have used autoradiographic localization of [3H]DEX in the early embryonic mouse lungs (d 14-16) and reported that the more proximal branches, which later differentiate into bronchi and bronchioles, showed very little binding, whereas the surrounding mesenchymal cells had intense binding. The distal epithelium, which is destined to become alveolar ducts and alveoli, and the surrounding mesenchyme both had intense DEX-binding(47). Jaskoll et al.(14) reported similar, but not identical, findings using in situ GR immunolocalization in mouse lung explants treated with TAC. In contrast to the findings of Beer et al.(47), these investigators observed only infrequent staining for GR in the distal epithelial cells of mouse embryonic lungs. Other evidence suggests that glucocorticoid treatment may further alter the activity and distribution of GR in the developing lung(48). Thus, the GR protein decreased in epithelial cells and increased in fibroblasts after glucocorticoid treatment of fetal lungs(48). Collectively, these findings suggest that significant mesenchymal-epithelial interactions occur during lung development in response to glucocorticoids. Accordingly, the DEX-induced dilatation of proximal tubules may be mediated by a paracrine factor(s) produced by surrounding mesenchymal cells. The growth factors of mesenchymal origin, such as HGF and KGF, may mediate the effects of glucocorticoids on lung epithelial cells. Both growth factors are potent mitogens for adult alveolar epithelial type II cells(16, 20, 23–25), but only KGF appears to influence fetal lung development. When added to fetal lung explants in culture, KGF inhibited branching, increased the expression of surfactant protein C, and induced airway dilatation and formation of fluid-filled cysts(21, 28). Pulmonary targeted expression of KGF and dominant-negative KGFR both caused abnormal lung development in transgenic mice(27). In contrast, HGF, or its antibody, do not affect lung maturation in vitro(21). In the present study, we have shown that DEX differentially regulates KGF and HGF; DEX increases the expression of KGF (and KGFR) transcripts but decreases those of HGF. The negative regulation of the HGF gene by DEX has been previously shown in studies with cultured human lung fibroblasts(49). As shown by our studies with fetal lung explants, the KGF gene is responsive to steroids also in the prostate and seminal vesicles(18, 19). In these organs, KGF appears to mediate epithelial cell proliferation stimulated by steroids(18, 19). The steroid regulation of the KGF gene occurs despite the absence of a consensus glucocorticoid-response-element sequence in the regulatory domain of the gene(19).

Several observations support a mediatory role of KGF in the DEX-induced distorted and accelerated lung maturation: 1) KGF and KGFR mRNA expression is increased by DEX (this study); 2) KGF has pronounced effects on pulmonary development and expression of surfactant proteins(21, 26–28, 50); 3) KGF induces cystic development of the lung both in vivo and in vitro(21, 26, 28); and 4) KGF promotes fluid accumulation via a CFTR-independent chloride transport(28). Thus, KGF may mediate the DEX-induced dilatation of the proximal tubules through its ability to induce formation of fluid-filled cysts during lung development in vitro and in vivo(21, 28). Additionally, because KGF has been shown to increase the expression of one or more of the surfactant proteins both in explant cultures and cultured type II cells(21, 50), the growth factor may mediate, at least in part, the DEX-stimulated expression of surfactant proteins.