Main

RA, the most biologically active derivative of retinol, is a potent morphogen shown to regulate lung morphogenesis [reviewed in Zachman(1) and Chytil(2)]. Its role as a lung morphogen is indicated by studies on RAR knockout mice. The mice with dual ablation of RAR-α and RAR-β2 genes showed agenesis of the left lung and hypoplasia of the right lung(3). In cultured fetal lungs, RA has been shown to both stimulate(4) and inhibit(5, 6) branching, depending upon the concentration of RA and the presence or absence of serum in culture media. There are conflicting data about the effects of RA on surfactant protein expression(5–7). The effects appear to vary with the length of exposure and the concentrations of RA(5–7).

RA, like glucocorticoids, regulates the expression of target genes through nuclear receptors and has been shown to antagonize the effects of glucocorticoids on gene expression and organogenesis. For example, RA antagonizes the glucocorticoid-stimulated fatty acid synthase gene expression in cultures of fetal rat lung explants(8). RA administration to postnatal rats prevents the low number of alveoli produced by treatment with DEX(9). The antagonism is not limited to the lung, but&& also is seen in other tissues. For example, RA stimulates the expression of phosphoenolpyruvate carboxykinase in adipocytes, whereas glucocorticoids inhibit the expression(10).

In the present study, we investigated whether glucocorticoids and RA have mutually antagonistic effects on lung morphology, and the expression of genes related to epithelial growth and differentiation in cultured embryonic rat lungs. In a companion study(24), we reported that early embryonic rat lungs treated with DEX show features of both distorted and accelerated lung maturation. The treated lungs have a distorted pattern of branching, tubular dilatation, and overall retarded growth. Several biochemical and morphologic features of advanced lung maturation are also observed: 1) the epithelial cells lining the distal tubules (future respiratory airways) are cuboidal/flattened; 2) the cuboidal cells often contain lamellar bodies and abundant glycogen; 3) rudimentary septa and large airspaces are present; 4) the mesenchymal tissue is considerably attenuated; 5) a more mature distribution of SP-C expressing cells is present; and 6) the expression of genes related to epithelial differentiation (SP-A, SP-B, SP-C, and CC10) and growth (KGF) increases. This report shows that RA antagonizes several of these effects of DEX on the early embryonic lung.

METHODS

The details of the methods and sources of materials used in the present study are presented in Oshika et al.(24). The information specific to this study is presented below.

Timed-pregnant rats of the Sprague-Dawley strain were obtained from Hilltop(Scottdale, PA). Lung primordia were dissected from d 14 and 15 rat embryos, placed on the upper surface of polycarbonate isopore membranes in Millicell-PCF assemblies(4), and cultured at 37 °C in 5% CO2/95% air for 4 d. Dulbecco's modified Eagle's medium containing 10% fetal bovine serum, streptomycin/penicillin G (0.1 mg/mL), gentamicin (10μg/mL), and Fungizone (2.5 μg/mL) was used as culture medium. Stock solutions of RA (Sigma Chemical Co., St. Louis, MO) and DEX were prepared in ethanol, and aliquots of the solutions were added to the cultures. Control cultures contained the same volume of ethanol (0.01-0.02%) as was used for the addition of DEX and RA. Lungs from the same litters were divided into control and treated groups. The number of branches in d 14 and 15 preculture explants were 7-10 and 30-40, respectively. The concentrations of RA(10-7-10-5 M) and DEX (10-9-10-6 M) used in this study are in the range of those used by other investigators for studies on lung development(4–7, 9, 11–14). Cultures were fed daily with a complete change of the medium, and explants were photographed daily. Cultured lungs from the treated or control groups were pooled (three lungs per pool) for RNA extraction with Trizol reagent (Life Technologies, Inc., Gaithersburg, MD). Data are expressed as mean ± SD. A t test was used to compare the two groups. A p values of <0.05 was considered significant.

RESULTS

The effects of DEX on lung morphology and gene expression are described in a companion study(24) and are discussed here to show the antagonism in the actions of RA and DEX. To facilitate description of the results, the tubules in a fetal lung explant are divided into zones based on their location, the future structures derived from them, and their susceptibility to DEX [Fig. 1B in Oshika et al.(24)]. Two major zones, I and II, contain proximal tubules (future conducting airways) and distal tubules/buds (future respiratory airways), respectively. Most of the proximal tubules of zone I are dilated (zone IB) after treatment with DEX, except those destined to form the trachea and main stem bronchi (zone IA) [Fig. 1B in Oshika et al.(24)].

Figure 1
figure 1

Effects of RA and DEX on branching morphogenesis of d 14 fetal lungs. Day 14 fetal lungs were cultured for 4 d in the presence of 10-5 M RA (panel RA), varying concentrations of DEX(D-1, 1 nM; D-10, 10 nM; D-50, 50 nM; and D-100, 100 nM), and combinations of the same concentrations of RA and DEX. A control lung (medium + ethanol) is shown in panel C. RA inhibited branching and caused dilatation of zone II tubules/buds (open arrowheads) and of zone IA tubules (solid arrows). DEX distorted branching and caused dilatation of zone IB tubules (solid arrowheads) and formation of thin and closely packed zone II tubules/buds(open arrow). RA prevented these effects of DEX. The RA-induced dilatation of zone IA tubules was not affected by co-treatment with DEX.

Full size image

Effect of RA and DEX on branching. Figure 1 shows branching morphogenesis of d 14 fetal lungs cultured for 4 d in the presence of RA, varying concentrations of DEX (1-100 nM), and combinations of RA (10-5 M) and the same varying concentrations of DEX. RA and DEX had distinct effects on lung branching and morphology. RA treatment at 10-5 M inhibited branching and caused dilatation of the tubules/buds of zones II and IA. As a result of treatment with DEX, the tubules of zone IB were dilated, and those of zone II became thin and closely packed at the periphery of the explant. The overall pattern of branching was also distorted. These effects of DEX were prevented by RA in culture. In contrast, DEX had no effect on the dilatation of zone IA tubules induced by RA (Fig. 1). We next examined the effects of varying concentrations of RA(10-7, 10-6, and 10-5 M) and/or DEX (10 and 100 nM) on d 14 lung explants in culture for 4 d. RA induced a concentration-dependent dilatation of the distal tubules (zone II) and an inhibition of branching (Fig. 2). The dilatation of zone IB tubules seen at 10 nM and 100 nM DEX was prevented by RA at 10-6 M. RA transformed the distorted pattern of branching induced by DEX to a pattern similar to that seen in control explants (for example, compare C with the explants treated with 10 nM DEX + 10-7 M RA, 10 nM DEX + 10-6 M RA, or 100 nM DEX + 10-6 M RA). The dilatation of zone IA tubules at 10-5 M RA and its persistence in the presence of DEX (shown in Fig. 1) were also seen (Fig. 2).

Figure 2
figure 2

Effects of different combinations of DEX and RA concentrations on branching morphogenesis. Day 14 fetal rat lungs were cultured for 4 d in the presence of different concentrations of DEX(D-10, 10 nM; D-100, 100 nM) and RA(R-10-7, 10-7 M;R-10-6, 10-6 M;R-105, 10-5 M), and their combinations. A control lung (medium + ethanol) is shown in panel C. RA at 10-6 and 10-5 M reduced the number of distal tubules/buds (zone II) and caused their dilatation (open arrowhead). Thin and closely packed zone II tubules/buds (open arrow) were seen in DEX-treated explants. The zone IB tubules were dilated (solid arrowhead) at both concentrations of DEX (10 and 100 nM). RA at 10-6 M prevented the dilatation induced by DEX (10 and 100 nM). RA transformed the distorted branching pattern induced by DEX to a pattern similar to that seen in control explants (for example, compare C with explants treated with 10 nM DEX + 10-7 M RA, 10 nM DEX + 10-6 M RA, or 100 nM DEX + 10-6 M RA). At a higher concentrations of RA (10-5 M), the zone IA tubules were highly dilated (arrows), and their growth was distorted. These effects of RA persisted even in the presence of DEX.

Full size image

Day 15 explants were also cultured in the presence of RA and DEX to determine the possible influence of the maturational state of preculture explants. The d 15 lungs, like those of d 14, are in the pseudoglandular stage of development(15). Lung explants were cultured for 4 d with RA (10-5 M), varying concentrations of DEX (1, 10, 100, and 1000 nM), or combinations of RA with the same concentrations DEX. The explants showed a much stronger dilatation of zone IB tubules after treatment with DEX(Fig. 3, compare with Fig. 1). Consequently, the zone occupied by the distal tubules (zone II) became even more narrow compared with d 14 explants after DEX treatment (compare Figs. 1 and 3). Furthermore, the antagonistic effects of RA (10-5 M) were seen only in combination with lower concentrations of DEX (1 nM) (Fig. 3). The stronger effects of 10 nM or higher concentrations of DEX were partially prevented by RA (Fig. 3). Thus, the magnitude of responses of d 14 and 15 explants to RA and DEX is quite different; the d 15 explants were more responsive to DEX, whereas RA had limited antagonistic effects.

Figure 3
figure 3

Effects of RA and DEX on branching morphogenesis of d 15 fetal lungs. Day 15 fetal lungs were cultured for 4 d in the presence of 10-5 M RA (panel RA), varying concentrations of DEX(D-1, 1 nM; D-10, 10 nM; D-100, 100 nM,D-1000, 1000 nM), and combinations of the same concentrations of RA and DEX. A control lung (medium + ethanol) is shown in panel C. RA reduced the number of distal tubules/buds (zone II, open arrows) and caused their dilatation (compare panels RA and C). The more proximal tubules(zone IA) were dilated and distorted by RA (arrows). DEX caused dilatation of Zone IB tubules (solid arrowheads), and as a result, the zone II tubules (open arrows) were thin and closely packed. The antagonistic effects of RA (10-5 M) were only seen in combination with lower concentrations of DEX (1 nM) (compare D-1 and D-1 + RA). The stronger effects observed at 10 nM or higher concentrations of DEX were partially prevented by RA. Thus, the magnitude of responses of the d 14 and 15 explants to RA and DEX was quite different; the d 15 lungs were more responsive to DEX and showed limited antagonistic effects of RA.

Full size image

Microscopic examination of lung explants treated with RA and DEX. For light microscopy, tissue sections were prepared from d 14 lung explants after 4 d in culture with 100 nM DEX, 10-5 M RA, or combinations of the same concentrations of RA and DEX. The DEX-treated explants (Fig. 4) showed that: 1) the dilated zone IB tubules often extended to the periphery of the explant; 2) the epithelium of dilated tubules mainly consisted of pseudostratified cells and, occasionally, of columnar/cuboidal cells (panel F);3) the zone II tubules were thin-walled and generally lined by cuboidal/flattened epithelial cells (panel J); and 4) rudimentary septa, large airspaces, reduced cellularity, and attenuated mesenchymal tissue were present. The zone IA tubules were distended after treatment with RA(panels C and G) and RA + DEX (panels D and H), but not with DEX alone. The zone II tubules of RA-treated explants were dilated and lined mainly by columnar epithelial cells(panel K). RA prevented the DEX-induced dilatation of zone IB tubules (panels D and H) and stimulated a pattern of terminal tubules and buds (zone II) similar to that seen in the control explant (compare panels I and L). The attenuation of mesenchymal tissue, seen after DEX treatment (panel B), was prevented by a co-treatment with RA (panel D).

Figure 4
figure 4

Light microscopy of cultured explants. Day 14 fetal lungs were cultured for 4 d in the presence of 100 nM DEX, 10-5 M RA, and their combination (DEX + RA). Sections of control explants (medium + ethanol) are shown in panels A, E, and I. The magnification in panels E-H and I-L is five times that in A-D. DEX treatment induced dilatation of tubules of zone IB(solid arrowheads). The dilated tubules often extended to the periphery of the explants. Both pseudostratified (open arrowheads) and columnar/cuboidal epithelium (thick solid arrows) can be seen, sometimes within the same proximal tubule (panel F). The cuboidal/flattened epithelium is also seen in distal tubules (thick solid arrows) of DEX-treated explants (panel J). Rudimentary septa, large airspaces, reduced cellularity, and attenuated mesenchymal tissue were all seen in DEX-treated explants. In RA-treated explants, zone IA tubules were dilated (thin arrow). The lumenal surface of the epithelium of the proximal (open arrow) and distal tubules was smooth, and the lining epithelial cells were mainly columnar. RA also caused dilatation of zone II tubules/buds (panel K, labeled X). In DEX + RA-treated explants, the distal tubules and their epithelia appeared morphologically similar to those in control explants. However, the RA-induced dilatation of the zone IA tubules persisted in explants co-treated with DEX (thin arrows).

Full size image

Effects of RA and DEX on DNA content of fetal lung explants. For these analyses, d 14 fetal lung explants were cultured for 4 d in the presence of DEX (100 nM), RA (10-5 M), and a combination of the same concentrations of DEX and RA. DNA content of cultured explants is shown in Figure 5. DEX-treated lung explants had significantly lower amounts of DNA compared with control or RA-treated explants. The DEX-induced decrease was prevented by co-treatment with RA. Labeling index(LI) of the epithelial cells of zone II were next determined. The results were: control, 44.7%; DEX, 11%; RA, 67.8%; and RA + DEX, 65.6%. Thus, the DEX-induced inhibition of cell proliferation was prevented by co-treatment with RA.

Figure 5
figure 5

Effects of RA and DEX on DNA content of fetal lung explants. Day 14 fetal lung explants were kept in culture for 4 d with DEX(100 nM), RA (10-5 M), and a combination of same concentrations of DEX and RA. Control lungs were cultured in medium + ethanol. The values are Mean± SD (the number (N) is shown). *p < 0.05, when DEX-treated group is compared with control, RA, or RA + DEX group.

Full size image

Effects of RA and DEX on gene expression. For these analyses, d 14 and 15 lung explants were cultured for 4 d with DEX (100 nM), RA(10-5 M), or with a combination of same concentrations of DEX and RA. We examined the expression several genes related to the growth (KGF and HGF) and differentiation (surfactant protein SP-A, SP-B, and SP-C and Clara cell secretory protein CC10) of epithelial cells. The results of Northern blot analysis and of corresponding densitometric quantitation are shown in Figure 6,A and B, respectively. DEX treatment of d 14 and 15 explants increased the levels of KGF, SP-A, SP-B, SP-C, and CC10 transcripts. However, the increase was more remarkable in d 15 explants. The DEX-induced increased expression of SP-A, SP-B, and CC10 transcripts was inhibited by a co-treatment with RA in d 14 explants. The inhibition was either less remarkable (SP-A and SP-B) or was not seen (CC10) in d 15 explants. RA suppressed mRNA expression of SP-C in both d 14 and 15 lungs. However, the DEX-stimulated SP-C mRNA expression was not affected by RA in the lungs of both gestation. The mRNA expression of KGF stimulated by DEX, either remained unchanged (d 14 explants), or was slightly increased (1.7-fold in d 15 explants) by RA.

Figure 6
figure 6

Effects of DEX and RA on mRNA expression of KGF, SP-A, SP-B, SP-C, CC10, and actin in lung explants. Lung explants were obtained from d 14 and 15 fetuses, and cultured for 4 d with DEX (100 nM), RA (10-5 M), or their combination. Control (C) explants were cultured in medium + ethanol. Results of Northern blot analysis and densitometric quantitation are shown in A and B, respectively. DEX increased the levels of KGF, SP-A, SP-B, SP-C, and CC10 transcripts in explants of both gestational days. However, the stimulation was stronger in d 15 explants. RA prevented the DEX-induced increased expression of SP-A, SP-B, and CC10 transcripts in d 14 explants. The inhibition by RA was either less remarkable (SP-A and SP-B) or not seen (CC10) in d 15 explants. RA suppressed mRNA expression of SP-C in both d 14 and 15 explants. The DEX-stimulated expression of KGF transcripts either remained unchanged (d 14 explants) or was slightly increased (1.7-fold in d 15 explants) by RA.

Full size image

Effects of RA and varying concentration of DEX on gene expression. Lungs from d 14 fetuses were cultured for 4 d with varying concentrations of DEX (1, 10, 50, and 100 nM), RA (10-5 M) alone, and combinations of the same concentrations of DEX and RA. Northern blot analysis was performed to monitor mRNA expression of SP-A, SP-B, SP-C, CC10, KGF, HGF, and actin. The results of Northern blot analysis and of corresponding densitometric quantitation are shown in Figure 7,A and B, respectively. DEX, at higher concentrations, stimulated the expression of SP-A, SP-B, SP-C, CC10, and KGF. The DEX-induced increases of SP-A, SP-B, and CC10 transcripts were inhibited by RA. The expression of HGF transcripts was decreased by DEX in a concentration-dependent manner and was further decreased by RA.

Figure 7
figure 7

Effects of RA and DEX on mRNA expression of KGF, HGF, SP-A, SP-B, SP-C, CC10, and actin in cultured fetal lungs. Lungs from d 14 fetuses were cultured for 4 d in the presence of DEX (1-100 nM, indicated as DX-1 to DX-100), RA (10-5 M), or combinations of the same concentrations of RA and DEX. Control explants were cultured in medium (lane C) or medium + ethanol (C/Et). Results of Northern blot analysis and densitometric quantitation are shown in A and B, respectively. DEX increased mRNA expression of SP-A, SP-B, SP-C, CC10, and KGF. The DEX-stimulated expression of all mRNAs (except SP-C and KGF) was inhibited by RA. HGF mRNA expression was inhibited by DEX, and the inhibition was further augmented by co-treatment with RA.

Full size image

Effects of RA and DEX on the distribution of CC10 and SP-C transcripts. 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 the type II cells, the latter weakly express CC10 transcripts(16). To establish the type of epithelial cells that appear in branching tubules in response to RA and DEX, we examined the distribution of CC10 and SP-C transcripts by in situ hybridization. Fetal lungs (d 14) were cultured for 4 d in the presence of DEX (100 nM) and/or RA (10-5 M). The distribution of CC10 transcripts was distinct after each treatment (Fig. 8). In both control and DEX-treated explants, CC10 mRNA was strongly expressed in the proximal tubules, and weakly expressed in the distal tubules. In DEX treated explants (Fig. 8), the dilated tubules had a strong, but variable, expression of CC10 transcripts, suggesting that these tubules in part represent prospective conducting airways. In RA-treated explants, most epithelial tubules expressed CC10 transcripts. However, CC10 expression varied within the same tubules, with a stronger expression seen in the proximal region of the tubules. Most of the tubules in RA + DEX-treated explants, including those with a high expression of SP-C transcripts (see below,Fig. 9), had uniform expression of CC10 transcripts. The distribution of SP-C transcripts is shown in Figure 9. To show possible colocalization of SP-C and CC10 transcripts in the same tubules, serial sections were used for in situ hybridization shown in Figures 8 and 9. In control lungs, SP-C transcripts were not detected in the tubules of zone I, which highly expressed CC10 transcripts (compare Figs. 8 and 9). The same tubules of zone II (prospective respiratory airways), which had a weak expression of CC10 (Fig. 8), strongly expressed SP-C transcripts (Fig. 9). In DEX-treated explant, the tubules with a strong CC10 expression (Fig. 8) lacked SP-C transcripts (Fig. 9). In contrast to the acinar distribution of epithelial cells expressing SP-C transcripts in control explants, their distribution in DEX-treated explants was patchy; individual or clusters of cells expressing SP-C mRNA were seen. The patchy distribution of SP-C transcripts reflects a more mature stage of lung development. In RA-treated explants, CC10 and SP-C transcripts were seen both in the same as well as in separate tubules (compare Figure 8 and 9). Furthermore, the dilated tubules of zone II seen in RA-treated explants had a strong expression of SP-C transcripts. In explants treated with DEX + RA, the DEX-induced patchy distribution of SP-C transcripts was not seen. Instead, an acinar distribution, similar to that seen in control explant, was observed.

Figure 8
figure 8

Distribution of CC10 mRNA by in situ hybridization. Day 14 lungs were cultured for 4 d in the presence of 10-5 M RA, 100 nM DEX, or a combination of the same concentrations of RA and DEX. CC10 is a marker of adult nonciliated bronchial cells (Clara cells) and is weakly expressed in rat alveolar type II cells. In control explant, CC10 transcripts were strongly expressed in the tubules of zone I(prospective conducting airways), whereas those of zone II (prospective respiratory airways) had a weak expression. In DEX-treated explants, the dilated tubules of zone I (zone 1B) showed a strong expression of CC10 transcripts, suggesting that these tubules in large part represent prospective conducting airways. In RA-treated explants, most epithelial tubules expressed CC10 transcripts. However, CC10 expression varied within the same tubules, with a stronger expression seen in the proximal region of the tubules. Most of the tubules in RA + DEX-treated explants, including those with a high SP-C mRNA expression (see below, Fig. 9), had uniform expression of CC10 transcripts. Left panels, bright field;right panels; dark field.

Full size image
Figure 9
figure 9

Distribution of SP-C mRNA by in situ hybridization. Day 14 lungs were cultured for 4 d in the presence of 10-5 M RA, 100 nM DEX, or a combination of the same concentrations of RA and DEX. In situ hybridization was performed on sections that were cut serially to those presented in Figure 8. SP-C, a specific marker of alveolar type II cells, is expressed by epithelial cells of the distal tubules (zone II, destined to form the respiratory airways). In contrast to the acinar distribution of epithelial cells expressing SP-C mRNA in distal tubules of control explants, their distribution in DEX-treated explants was patchy; individual cells or clusters of cells expressing SP-C were seen. In RA-treated explants, the epithelium of dilated tubules of zone II had a strong expression of SP-C transcripts. The distribution of SP-C transcripts in control and DEX + RA treated explants was similar. Left panels, bright field; right panels, dark field.

Full size image

DISCUSSION

In this study, we have used early embryonic rat lungs in culture to show the effects of RA alone and in combination with DEX on branching morphogenesis and gene expression. We observed that the treatment of lung explants with 10-6 and 10-5 M RA decreased branching and caused dilatation of the zone II tubules. Similar findings were reported by other investigators(5, 6). However, a key finding reported by Cardoso et al.(5) that RA treatment of lung explants caused proximalization was not observed in this study. According to these investigators, RA treatment of d 13 lung explants suppressed the expression of SP-C (a marker of distal epithelium) in distal tubules and caused proximalization, a developmental pattern that favored growth of proximal airways (zone I) and suppressed formation of distal epithelial tubules/buds(zone II). In the present study, however, the expression of SP-C was high in the zone II tubules of RA-treated explants, suggesting a lack of proximalization. Cardoso et al.(5) further reported that the proximalization is only seen if lungs were exposed to RA early in development (d 13), which may explains the lack of proximalization in d 14 lung explants used in this study. The tubular dilatation in zone IA(tubules destined to become trachea and main bronchi) seen after treatment of lung explants with 10-5 M RA is a new finding. The increased susceptibility of these tubules to RA may be related to the associated high expression of RAR-β(3).

Other investigators have reported profound, but contradictory, effects of RA on surfactant proteins expression in cultured fetal lungs(5–7). A short-term treatment (4 h) of fetal rat lung explants with 10-10 to 10-5 M RA resulted in an increase of SP-A mRNA(6), whereas longer treatments (5-6 d) of fetal rat (10-5 M RA) or human (3 × 10-6 M RA) lungs in culture decreased SP-A mRNA(5, 7). In the present study, the effects of RA on SP-A and SP-B (and CC10) expression were difficult to assess because of very low levels of their transcripts in both treated and control explants after 4 d of culture. The levels of SP-C mRNA, in contrast, were high, and were suppressed by RA treatment of d 14 and 15 lungs. Both increases and decreases in the levels of SP-C mRNA have been reported in RA-treated explants(5–7). The contrasting effects of RA may be due to differences in the developmental state of preculture lung explants, the concentration of RA, and the length of RA exposure.

In the present study, we noted major effects of RA in conjunction with DEX. RA prevented many of the effects of DEX on lung morphology and gene expression(24). Thus, RA antagonized the DEX-induced: 1) dilatation of proximal tubules and formation of large airspace; 2) increased levels of SP-A, SP-B, and CC10 mRNAs; 3) a more mature distribution of SP-C transcripts; 4) reduced lung growth and mesenchymal tissue; and 5) inhibition of epithelial cell proliferation in the zone II tubules. Strong antagonistic effects of RA were seen on genes expressed in epithelial cells (SP-A, SP-B, and CC10). In contrast, the effects of DEX on the expression of HGF and KGF, both products of mesenchymal cells and potent mitogens of alveolar epithelial type II cells(16, 17), was potentiated by RA. It is worth noting that the antagonistic effects of RA on lung morphology and gene expression were observed at high concentrations of RA (10-6 and 10-5 M). These concentrations have been used by others in studies on lung development in vitro(4–7). Lower concentrations of RA, either alone or in combination with DEX, had no significant morphologic effects, but were not tested in our study of gene expression. This aspect needs to be examined further in view of a recent report, showing that short-term exposures (4 h) of fetal lung explants to low concentrations of RA (10-10 to 10-5 M RA) stimulated expression of surfactant proteins(6). Nevertheless, others have reported antagonistic effects of RA in other aspects of lung development. Thus, RA (5 × 10-6 to 5 × 10-4 M) antagonized the stimulatory effects of DEX (100 nM) on the activity of fatty acid synthase and the fatty acid synthase gene transcription in cultured fetal lungs(9). During postnatal alveolarization in the rat, RA treatment not only counteracts the DEX-induced inhibition of alveolar formation, but further increases the number of alveoli(8). In the adult rat, DEX administration diminishes the concentrations of retinol-binding protein, whereas treatment with RA increases the lung concentrations of this mRNA(18, 19). RA and DEX have antagonistic effects on epidermal growth factor receptor expression in fetal rat lung cells; RA enhances receptor synthesis, whereas DEX inhibits it(20).

Some antagonistic effects may result from the glucocorticoid-induced alterations of retinoid metabolism(1, 2). Glucocorticoid administration to pregnant rats reduces mRNA expression of retinol binding protein(18, 19, 21), and accelerates the depletion of retinyl esters from the lung(22). The depletion normally occurs close to birth(22). Thus, glucocorticoids may promote lung maturation by inhibiting retinoid action through effects on retinoid metabolism. Some of the RA- and DEX-induced effects on gene expression and morphology are dependent on the gestational age of the lung at explanting. For example, the DEX-stimulated expression of genes related to epithelial cell differentiation(SP-A, SP-C, and CC10) and airway dilatation were more pronounced in d 15 lungs than in d 14 lungs. Furthermore, RA greatly suppressed DEX-induced SP-A, SP-B, and CC10 mRNA expression in d 14 lungs, but had a less remarkable (SP-A and SP-B) or no suppression (CC10) in d 15 lungs. These observations show a trend toward RA having effects early in gestation that are lost as development proceeds, whereas the effects of DEX become progressively more potent. Developmental changes in signal transduction and in the expression of receptors and growth factors may contribute to altered responses of fetal lungs to DEX and RA.

In a companion study(24), we postulated that KGF mediates some of the effects of DEX on lung maturation. The basis of this hypothesis is the similarity of some morphologic and biochemical effects induced by DEX and KGF. Thus, both agents distort branching, cause airway dilatation, and stimulate expression of one or more of the surfactant proteins in fetal lung explants in culture(16, 24). In the present study, DEX and DEX + RA induced contrasting effects on explant morphology and gene expression, despite a high expression of KGF. A plausible explanation of this finding may be that RA, either directly, or indirectly through the production of inhibitory substances (for example, transforming growth factor-β)(23), inhibits KGF activity and/or alters its distribution or of its receptors. Further studies are required to distinguish the effects of DEX and RA that are through interaction with the regulatory elements of the affected genes from those mediated by, for example, KGF(16, 24), epidermal growth factor(20), and transforming growth factor-β(23).

In summary, the results of this study show that RA antagonizes the DEX-induced effects on lung development and on the expression of genes related to epithelial cell differentiation (surfactant proteins and CC10). We postulate that a balanced action of endogenous retinoids and glucocorticoids is essential for normal lung development.