Altered Epithelial Cell Proportions in the Fetal Lung of Glucocorticoid Receptor Null Mice Timothy J. Cole, Nicola M. Solomon, Rosemary Van Driel, Julie A. Monk, Daniel Bird, Samantha J. Richardson, Rodney J. Dilley, and Stuart B. Hooper Department of Biochemistry and Molecular Biology, University of Melbourne, Parkville, Victoria; Murdoch Children’s Research Institute, Royal Children’s Hospital, Parkville, Victoria; Baker Heart Research Institute, Prahran, Victoria; and Department of Physiology, Monash University, Clayton, Victoria, Australia Glucocorticoids provide important signals for maturation of the fetal lung and antenatal glucocorticoids are used to reduce the respiratory insufficiency suffered by preterm infants. To further understand the role of glucocorticoids in fetal lung maturation, we have analyzed mice with a targeted null mutation for the glucocorticoid receptor (GR) gene, which severely retards lung development. The lungs of fetal GR-null mice have increased lung weight and DNA content, are condensed and hypercellular, with reduced septal thinning leading to a 6-fold increase in the airway to capillary diffusion distance. In fetal GR-null mice, mRNA levels of the type II epithelial cell surfactant protein genes A and C were reduced by ⵑ 50%. Analysis of epithelial cell types by electron microscopy revealed that the proportions of type II cells were increased by ⵑ 30%, whereas the proportions of type-I cells were markedly reduced (by ⵑ 50%). Similarly, we found a 50% reduction in mRNA levels for T1␣ and aquaporin-5, two type I cell–specific markers, and a 20% reduction in aquaporin-1 mRNA levels. This demonstrates that during murine embryonic development, receptor-mediated glucocorticoid signaling facilitates the differentiation of epithelial cells into type I cells, but is not obligatory for type II cell differentiation. The development and embryonic growth of the mammalian lung is a complex and highly organized process involving a combination of intrinsic growth and differentiation factors, and concerted actions from circulating factors and hormones (1). Glucocorticoids are one of a number of circulating hormones that also include retinoids, thyroid hormone, and other cAMP-mediated factors that are important during the final stages of lung maturation. These hormones play important roles during late gestation, particularly in the differentiation and development of terminal alveoli and the promotion of lung surfactant production. Synthetic glucocorticoids (particularly betamethasone or dexamethasone) are widely used antenatally to reduce the severity of the respiratory insufficiency suffered by preterm infants, and act to accelerate fetal lung maturation and increase lung surfactant production (2). Antenatal glucocorticoid treatment has had a major benefit in reducing the incidence of neonatal respiratory distress syndrome (RDS) and intraventricular hemorrhage, leading to decreased neonatal mortality. However, its use remains controversial, particularly the administration of multiple doses, due to the reported side effects of glucocorticoids on lung and body growth and development of the central nervous system (Received in original form June 17, 2003 and in revised form October 20, 2003) Address correspondence to: Dr. Timothy J. Cole, Department of Biochemistry and Molecular Biology, University of Melbourne, Parkville, 3010, Victoria, Australia. E-mail: tjcole@unimelb.edu.au Abbreviations: corticotrophin-releasing hormone, CRH; epithelial cell, EC; epithelial sodium channel, ENaC; glucocorticoid receptor, GR; mineralocorticoid receptor, MR; post coitum, p.c.; surfactant protein, SP; transmission electron microscopy, TEM; wild type, WT. Am. J. Respir. Cell Mol. Biol. Vol. 30, pp. 613–619, 2004 Originally Published in Press as DOI: 10.1165/rcmb.2003-0236OC on October 24, 2003 Internet address: www.atsjournals.org (3, 4). Prenatal glucocorticoid treatment also has the potential to alter fetal programming, leading to the onset of diseases such as hypertension when the offspring develop into adults (5). However, it is well established that endogenous glucocorticoids increase in an exponential-like manner just before birth and participate in the regulation of biochemical and cytoarchitectural changes in the developing fetal lung (1), yet little is known of the underlying molecular and cellular events that underpin these glucocorticoidmediated effects. The majority of glucocorticoid actions are mediated via the intracellular glucocorticoid receptor (GR) (6). The GR is a ligandactivated transcriptional regulator and is a member of the steroid receptor family, a subgroup of the nuclear receptor superfamily (7). After steroid binding, the GR translocates to the nucleus, dimerizes, and binds to glucocorticoid-response DNA binding sites upstream of specific target genes (6). Glucocorticoids via GR are able to both activate and repress target gene transcription. Previous studies, in which fetal sheep or rabbits were treated with synthetic glucocorticoids such as betamethasone and dexamethasone, indicated that glucocorticoids enhanced lung development and type II epithelial cell (EC) differentiation (8). Numerous biochemical studies in preterm rabbits, rats, mice, and sheep have shown that glucocorticoids, in part, regulate the production of lung surfactant, particularly in promoting expression of the type II EC-specific surfactant protein (SP)-A, -B, and -C genes (1). Glucocorticoids also influence clearance of lung liquid and promote tissue remodeling before birth, thereby greatly improving lung tissue compliance and gas diffusion after birth (9–11). To investigate the role of glucocorticoid action via GR signaling during embryonic development, the GR gene was ablated using gene targeting in mice (12), which causes a number of phenotypic effects. The most striking observation was seen in the lung, where GR-null mice at birth displayed severe lung atelectasis with little to no inflation of lung tissue. On a C57Bl6/ 129 sv genetic background, ⬎ 90% of GR-null mice die at birth from respiratory dysfunction, whereas all GR-null mice die at birth on a 129 sv isogenic genetic background (12). An identical phenotype of perinatal death and lung dysfunction has been described in another GR-null mouse constructed by gene-targeted deletion of exon 2 and the proximal promoter of the GR gene (13). An almost identical lung phenotype is seen in corticotrophinreleasing hormone (CRH)-deficient mice (of CRH-deficient mothers), where there is a clear impairment in glucocorticoid production (14). CRH-null mice have delayed induction of SP-A and SP-B with reduced pulmonary septal thinning and airway formation. Interestingly, mice with a gene-targeted point mutation in GR that prevents dimerization and DNA binding develop normally, with no overt lung defect, indicating that the actions of glucocorticoids in the developing lung may be mediated by DNA-binding independent actions (15). A recent study using 614 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 30 2004 GR-null mice with deletions of exon 2 and the proximal promoter have found a marked reduction in the growth factor midkine, 3 d before birth, that may contribute to the immature lung phenotype detected at birth (16). To clarify, in more detail, the role of glucocorticoids during fetal lung development, we have further analyzed the lungs of isogenic 129 sv GR-null mice generated previously with a series of histologic and molecular studies (12). Using both light and electron microscopy, we show dramatic morphologic differences in the development of the lung, with lung maturation of GRnull mice severely retarded at birth. Airspace formation in lungs of GR-null mice is severely reduced compared with wild-type (WT) controls. Pulmonary surfactant is present in the airway of GR-null lung tissue, indicating apparent normal surfactant production and secretion. Lung tissue expansion is markedly reduced, and upon investigation using electron microscopy, the lungs of Day 18.5 GR-null mice resemble those of their WT littermates at Day 16.5 of pregnancy. Finally, careful analysis of epithelial cell types using electron microscopy and cell-specific markers reveals a marked reduction in type I ECs at birth, which would significantly contribute to the respiratory dysfunction due to a severe reduction in the surface area for gas exchange. In contrast, the proportion of type II ECs was increased in GRnull mice, indicating that corticosteroid action via GR is unnecessary for type II cell differentiation. A minimum of 100 ECs were classified for each animal (from six GR-null and six WT littermate controls) and the number of nuclear profiles of each type counted (19, 20). Identification of ECs depended on visualization of the epithelial cell basement membrane, with all ECs localized to the luminal surface of this membrane. ECs for each genotype were counted in a blinded analysis and categorized as one of four phenotypes—stem cells, type I ECs, type II ECs, and intermediate ECs. Undifferentiated (or stem) epithelial cells were rounded in shape and contained abundant cytoplasmic glycogen; they did not contain lamellar bodies or have any evidence of a cytoplasmic extension (see below). Type I ECs had flattened cytoplasmic extensions, flattened nuclei, little perinuclear cytoplasm and few cytoplasmic organelles. Cytoplasmic extensions are defined as peripheral regions of cytoplasm that extend along the luminal surface of the epithelial cell basement membrane, with both apical and basolateral membranes running in parallel to each other and separated by ⬍ 0.5 ␮m. Type II ECs were rounded in shape with a rounded nucleus and had microvilli on their apical surface and abundant cytoplasmic organelles, including lamellar bodies. The intermediate cells were a heterogeneous group that displayed characteristics of both type I and type II ECs. Their classification depended on the presence of a flattened nucleus and marked cytoplasmic extensions, but they also contained lamellar bodies and usually had microvilli on their apical surface (19). Previous studies have shown that in sheep, humans, and rats, the nuclear diameters of type-I and type-II ECs are similar and, therefore, the chances of counting a nuclear profile of each type are equal using EM (21–23). Materials and Methods Lung tissue was dissected from mouse embryos and fixed using the same procedures as described previously (24). Portions of lung tissue were embedded in paraffin, cut at 5 ␮m, and stained with hematoxylin and eosin. Stereologic measurements of lung tissue fraction and volume and luminal fraction and volume, were made for three GR-null and three WT littermate controls using standard stereologic techniques and grids (25). Lung sections were viewed using light microscopy (⫻40) and projected onto a screen; the final magnification was ⫻720, which was determined by simultaneous projection of a 0.1-mm graticule. Mice GR-null mice were generated by gene targeting as described previously (12) and produced by mating GR-heterozygous (⫹/⫺) mice. These mice were generated on an isogenic 129 sv genetic background and displayed a severe phenotype that resulted in 100% perinatal death. To study the lung abnormalities of GR-null mice in more detail, pregnant heterozygous female mice were killed with CO2 at Day 18.5 post coitum (p.c.), 0.5 d before term. Pups were removed by cesarean section, weighed, and killed by decapitation. A small piece of tail was used as a source of genomic DNA for genotyping at the GR locus by PCR as described previously (17). Lung Weight and Determination of DNA Content Total lung tissue was removed intact from Day 18.5 p.c. fetal mice, blotted to remove excess fluid, and weighed. For determination of DNA content, lung tissue was digested overnight by proteinase K (0.5 mg/ml) and then genomic DNA isolated after salt precipitation of protein as described previously (17). DNA concentrations were determined by absorbance at 260 nm. Hormone Assays Corticosterone levels in plasma from fetal mice were determined by a corticosterone radioimmunoassay (RIA) using {1, 2, 6, 7- 3H}corticosterone (Amersham, Little Chalfont, UK) and a corticosterone antibody (B3–163; Endocrine Sciences, Calabas, CA) as described previously (17). Plasma levels of total thyroxine (T4) was determined with a commercial RIA kit (Diagnostic Products Corp., Los Angeles, CA) using 125 I-labeled T4 as a tracer as described previously (18). Transmission Electron Microscopy For transmission electron microscopy (TEM), lung tissue (left major lobe) was sliced into 2- to 4-mm cubes and immersion fixed in 4% p-formaldehyde, 2% glutaraldehyde, 4% sucrose in Hepes-buffered saline, pH 7.4 for 3–4 h at room temperature. The tissue was then postfixed overnight at 4⬚C in 1.5% OsO4 in veronal acetate, rinsed 3 ⫻ 5 min at 4⬚C in Tris/maleate pH 5.2, stained en bloc for 90 min on ice in 1.5% uranyl acetate in Tris/maleate (pH 7.2), and dehydrated in cold acetone followed by a final two changes of dry acetone at room temperature. Fixed tissue blocks were embedded and polymerized in Epon overnight at 65⬚C. Ultrathin tissue sections (70 nm) were stained with uranyl acetate and lead citrate and were viewed using a Transmission Electron Microscope (TEM; JEOL100; JEOL Ltd., Akishima, Japan); random micrographs were taken and used for analysis. Light Microscopy and Morphometric Tissue Analysis Isolation of RNA and Northern Blot Analysis Total RNA was prepared from isolated mouse lung by homogenization in TRIzol reagent (Gibco/BRL, Rockville, MD), a guanidinium isothiocyanate/phenol-based homogenization solution. After chloroform extraction, RNA was precipitated from the aqueous phase with isopropanol, washed in 70% ethanol and redissolved in nuclease-free sterile water. Total RNA (10 ␮g) was separated in 1.2% agarose/2.2 M formaldehyde gels and blotted onto Genescreen Plus nylon membranes for Northern blot analysis by hybridization with 32P-labeled RNA probes. RNA probes were derived from pBluescript or pBluescribe (Stratagene, La Jolla, CA) plasmids containing either a 474-nucleotide mouse SP-A cDNA fragment, a 1,550-nucleotide mouse SP-B cDNA fragment, a 683-nucleotide mouse SP-C cDNA fragment, a 575-nucleotide rat ENaC␣ cDNA fragment, a 520-nucleotide rat ENaC␤ cDNA fragment, a 484nucleotide rat ENaC␥ cDNA fragment, a 310-nucleotide mouse aquaporin-1 (AQP1) cDNA fragment, a 240-nucleotide mouse aquaporin-5 (AQP5) cDNA fragment, or a 751-nucleotide mouse T1␣ cDNA fragment; T1␣ is a type-I AEC-specific marker (26). Each plasmid was linearized to generate a 32P-labeled antisense riboprobe using a riboprobe kit (Promega, Madison, WI). All filters were hybridized at 68⬚C in Ultrahyb buffer (Ambion, Austin, TX) and washed at 75⬚C in 0.1⫻ SSC, 0.1% SDS three times for 15 min each. Filters were rehybridized with a 32P-labeled 110-nucleotide cDNA fragment of the mouse cytochrome oxidase mRNA to control for differences in total RNA loading on agarose gels. Statistical Analysis Hormone levels, tissue and body weights, DNA contents, percent tissue and airspace volumes as well as differences in AEC proportions were analyzed by one-way ANOVA followed by a Tukey’s post hoc test, with statistical significance set at P ⬍ 0.05. For all Northern blot analyses, RNA levels were quantitated by densitometry analysis on a Typhoon Phosphoimager from Molecular Dynamics (Sunnyvale, CA) using ImageQuant software, and standardized to the expression of the housekeeping gene cytochrome oxidase (cytoxi). Cole, Solomon, Van Driel, et al.: Alveolar ECs in GR-Null Mice Results Circulating Hormone Levels As previously observed for C57Bl6/129 sv fetal GR-null mice (12), we detected in plasma samples from isogenic 129 sv fetal GR-null mice a significant increase in circulating levels of plasma corticosterone, presumably caused by a defect in normal glucocorticoiddriven negative feedback via the hypothalamus-pituitary-adrenal axis (Table 1). We found no difference in the levels of circulating thyroxine (T4), another systemic hormone important in embryonic lung development (Table 1). 615 TABLE 1. Lung weight, DNA content, and hormone analysis at Day 18.5 pc for GR-null and WT fetal mice Genotype n WT GR-Null P Value Total weight, g Lung wet weight, mg/g body weight Lung DNA content, ␮g/g body weight Corticosterone, ng/ml Thyroxine (total), nmol/liter 6 1.12 ⫾ 0.14 1.15 ⫹ 0.16 N.S. 5 30.8 ⫾ 1.3 38.7 ⫾ 2.5 ⬍ 0.01 5 5 6 56.2 ⫾ 3.5 109 ⫾ 3.8 9.22 ⫾ 0.3 89.5 ⫾ 5.4 349 ⫾ 7.4 8.82 ⫾ 0.10 ⬍ 0.02 ⬍ 0.01 N.S. Values are shown as ⫾ SEM. Fetal Body Weights, Lung Weight, and Total Lung DNA Content Although there was no difference in overall fetal body weight, lung wet weight and DNA content were significantly higher in the lungs of GR-null fetal mice, compared with WT mice, which is indicative of significant hypercellularity (Table 1). This is consistent with the histologic findings (see below) and is similar to results found in the CRH-null mouse (14). Histologic Analysis of the Developing Lung in Fetal GR-Null Mice Light microscopy. At Day 18.5 pc, histologic analysis of the lung from GR-null mice by light microscopy showed a more condensed lung architecture with a lack of normal airspace formation and thickened interalveolar septa (Figures 1A and 1B). The percentage of tissue and airspace for WT, heterozygous, and the GR-null fetal lung was determined by morphometric analysis and clearly showed that the lungs from GR-null mice have a significantly more condensed tissue morphology compared with WT mice, with heterozygous mice having an intermediate level of tissue expansion (Figure 1C). Electron microscopy. At Day 16.5 pc, 3 d before birth, there was little difference in the ultrastructure of lung from WT and GR-null mice; the epithelium of the terminal airways was predominantly comprised of undifferentiated epithelial cells containing large quantities of glycogen (data not shown). However, by Day 18.5 pc, many of these cells had differentiated (see below) in the lung of WT fetal mice, whereas the lungs from GR-null fetal mice remained similar in appearance to that of a Day 16.5 pc WT lung (Figures 2C and 2D). To further assess the hypercellularity and thicker interairway distances in the lungs of GR-null fetal mice, we measured the minimum distances between airways and the nearest adjacent capillaries. Using elec- tron microscopy, multiple fields of view were assessed from the lungs of WT (n ⫽ 6) and GR-null fetal mice (n ⫽ 6) at Day 18.5 pc. The lungs of GR-null fetal mice had significantly thicker air/blood gas diffusion barriers, with multiple cellular compartments between the airways and adjacent capillaries; there was no evidence of epithelial cell and endothelial cell basement membrane fusion in GR-null mice. As a result, the measured airway to capillary diffusion distance was increased 6-fold in GR-null mice. It increased from 0.68 ⫾ 0.14 ␮m in WT mice to 4.02 ⫾ 1.00 ␮m in GR-null mice (P ⬍ 0.05; see Figure 3A). Epithelial Cell Type Proportions in GR-Null Mice The proportions of epithelial cell phenotypes in the terminal airways of GR-null fetal mice were compared with those of WT fetal mice at Day 18.5 pc (n ⫽ 5 for each group; Figure 3B). Cells were categorized as either type I, type II, stem cells, or an intermediate cell type, based upon their ultrastructural appearance using electron microscopy (see Materials and Methods for details). There were significantly more (80% increase, P ⬍ 0.05) nuclear profiles counted per unit area of tissue from GRnull fetal mice than were counted in the lungs of WT fetal mice (Figure 3B). This was largely due to a 50% increase (P ⬍ 0.05) in type II cells and a 2-fold increase in undifferentiated EC nuclear profiles (P ⬍ 0.05). On the other hand, the number of type I epithelial cell nuclear profiles per unit area was reduced by 40% in GR-null mice compared with WT fetal mice. Expressed as a proportion of the total number of ECs counted, the proportion of type I ECs in the lungs of WT fetal mice (49.6 ⫾ 2.9%) was greater than in GR-null fetal mice (20.0 ⫾ 5.1%; P ⬍ 0.05). In contrast, the proportion of type II ECs in Figure 1. Histologic analysis of lung tissue from Day 18.5 pc fetal GR-null and WT littermate mice by light microscopy. (A ) Lung tissue of Day 18.5 pc WT fetal mice, ⫻200 magnification. (B ) Lung tissue of Day 18.5 pc GR-null fetal mice, ⫻200 magnification. (C ) Morphometric analysis for percentage (⫾ SD) of tissue (solid bars) and airspace (striped bars) in sections of Day 18.5 pc fetal lung from WT (⫹/⫹), heterozygous (⫹/⫺), and GR-null (⫺/⫺) mice. Asterisks indicate statistical significance. 616 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 30 2004 Figure 2. Ultrastructural analysis of lung tissue from Day 18.5 pc fetal GR-null (⫺/⫺) and WT (⫹/⫹) littermate mice by transmission electron microscopy. (A ) Pulmonary surfactant in the airway of Day 18.5 pc fetal WT littermate mice. (B ) Pulmonary surfactant in the airway of Day 18.5 pc GR-null mice. (C ) Lung from Day 18.5 pc WT fetal mice. (D) Lung from Day 18.5 pc GR-null fetal mice. RBC, red blood cell; E, endothelial cell; I, type I epithelial cell (EC); II, type II EC; S, undifferentiated EC; arrows, type I EC cytoplasmic extensions; asterisks, areas previously occupied by glycogen. Bar represents 1 ␮m. the lungs of GR-null fetal mice (44.8 ⫾ 3.9%) was significantly greater than the proportions of these cell types in the lungs of WT fetal mice (30.3 ⫾ 3.1%). A close analysis of these lungs by transmission electron microscopy revealed that the type II ECs in GR-null fetal mice not only contained multiple lamellar bodies (Figure 2D), but were also capable of releasing surfactant, as indicated by the presence of abundant surfactant material within the future airways (Figures 2A and 2B). Similarly, the proportion of undifferentiated epithelial stem cells was also significantly elevated in GR-null fetal mice compared with WT fetal mice (30.4 ⫾ 2.2% versus 13.6 ⫾ 2.5%; Figure 3). Consequently, we find clear and significant alterations in the proportions of EC phenotypes in the lungs of GR-null mice. In particular, it appears that EC differentiation is delayed, as indicated by the higher proportion of undifferentiated stem cells, and that differentiation into a mature type II cell phenotype is relatively unaffected by ablation of the GR. Expression of AEC-Specific Proteins in the Lungs of GR-Null Fetal Mice To further assess the differentiated type II epithelial cell phenotype in the lungs of GR-null fetal mice, we analyzed the expression of the type II cell–specific surfactant protein genes, SP-A, SP-B, and SP-C. Total RNA from Day 18.5 pc fetal lung tissue of WT and GR-null mice was analyzed by Northern blot analysis (Figure 4). We detected a 50% reduction in mRNA levels for SP-A and SP-C in the lungs of GR-null fetal mice, compared with WT littermate fetal mice, with no significant change in SPB mRNA levels. Taken together with the significant increase in the proportion of type II epithelial cells, this indicates that there is potentially a large reduction in SP-A and SP-C mRNA per type II AEC in the lungs of fetal GR-null mice. To further assess the reduction in type-I epithelial cell proportions we observed in the lungs of GR-null fetal mice, we used two well-defined type-I epithelial cell markers that are not influenced by glucocorticoids. These were a recently described type-I cell-specific marker called T1␣, a trans-membrane protein of currently unknown function (26), and aquaporin-5, a water channel believed to be important in the transcellular movement of water across the type I epithelial cell (27, 28). The steady-state mRNA levels of T1␣ and aquaporin-5 were measured by Northern blot analysis of total lung RNA from GR-null fetal mice versus WT controls (Figure 5). The levels of T1␣ mRNA were reduced from 0.4 ⫾ 0.05 arbitrary units in WT fetal mice to 0.23 ⫾ 0.07 arbitrary units in GR-null fetal mice, a reduction of ⵑ 50%. Similarly, the mRNA levels for AQP5 were reduced from 0.18 ⫾ 0.02 arbitrary units in WT fetal mice to 0.09 ⫾ 0.03 arbitrary units in GR-null fetal mice, a reduction of ⵑ 50%. The level of reduction in expression of T1␣ and AQP5 approximately corresponded to the reduction in the proportion of type I epithelial cells detected by electron microscopy. These results suggest that glucocorticoid signaling via GR either directly or indirectly plays an important role in differentiation of type I epithelial cells late in fetal development. Expression of Glucocorticoid-Regulated Genes in the Lungs of GR-Null Fetal Mice As glucocorticoids have been reported to play a role in fluid withdrawal at birth, we also measured the steady-state mRNA levels in the lungs of GR-null versus WT fetal mice for the epithelial sodium channel (ENaC) subunits ␣, ␤, and ␥ (Figure 4). We found little change in mRNA levels for the ENaC␣ and ENaC␤ subunits, but a marked reduction in mRNA levels for the ENaC␥ subunit in lung RNA from GR-null mice. ENaC␥ mRNA levels were reduced from 0.17 ⫾ 0.03 in WT fetal mice to 0.075 ⫾ 0.02 in GR-null fetal mice. In GR-null fetal mice, we also measured the steady-state mRNA levels for AQP1, which is an important water channel in endothelial cells and is a known glucocorticoid- Cole, Solomon, Van Driel, et al.: Alveolar ECs in GR-Null Mice 617 Figure 3. Airway to capillary distance and the proportions of the different epithelial cell types in the lung of WT (⫹/⫹) and GR-null (⫺/⫺) fetal mice at Day 18.5 pc. (A ) Comparisons of airway to capillary distance (␮m ⫾ SD) in the lungs of Day 18.5 pc WT (⫹/⫹) and GR-null (⫺/⫺) fetal mice. (B ) Epithelial cell types were identified using the criteria described in Materials and Methods and were determined from representative electron micrographs of lung tissue from six WT and six GRnull fetal mice. Open bars, type I; striped bars, type II; hatched bars, undifferentiated EC; solid bars, intermediate. Average numbers are shown ⫾ SD. Asterisks indicate statistical significance. induced gene in the developing lung (29) (Figure 5). Levels of AQP1 mRNA were reduced in GR-null fetal lung to ⵑ 30% of WT littermates. Abrogated induction of AQP1 in the developing lung may also contribute to respiratory dysfunction via a reduction in fluid withdrawal at birth. Discussion This study has further investigated some key aspects of the proposed physiologic role of glucocorticoids in embryonic lung development. GR-null mice at birth have a severe respiratory defect that has been recently reproduced in a second GR-null mouse line that lacks the GR promoter and entire second exon (13). This severe respiratory defect is almost identical to the lung defect of CRH-null mice that lack production of glucocoticoid hormones (14). GR-null mice have a reduced capacity to activate key gluconeogenic enzyme genes in the liver and impaired negative feedback in the hypothalamus-pituitary-adrenal (HPA) axis, resulting in markedly elevated plasma ACTH and corticosterone levels (12, 30). The adrenal glands of fetal GR-null mice are enlarged with significant hypertrophy of the adrenal cortex, and recent studies show abnormalities in adrenal medullary formation and function (12, 31). Thymic T cells were found to be resistant to dexamethasone-induced apoptosis, and glucocorticoids have been shown in a number of independent GR-null mouse lines not to be essential for normal thymic T cell development in both fetal and adult mice (13, 32–34). The present study addressed the role of glucocorticoids in the late embryonic development of the mouse lung by investigating in more detail the respiratory defect in GR-null mice. Pharmacologic doses of synthetic glucocorticoids have clearly been shown to induce lung surfactant production and to accelerate Figure 4. Northern blot analysis of the surfactant protein genes SP-A, SP-B, and SP-C; and ENaC subunit ␣, ␤, and ␥ mRNA levels in Day 18.5 pc fetal lung of WT and GR-null mice. (A ) Detection of SP-A, SP-B, SP-C mRNA, and ENaC subunit mRNAs in three WT (⫹/⫹) and three GR-null (⫺/⫺) fetal lung samples. Samples (10 ␮g/lane) were loaded onto a 1.2% agarose gel. RNA were separated by electrophoresis, transferred to a nylon membrane, and hybridized with riboprobes as described in Materials and Methods. Filters were rehybridized for cytochrome oxidase (cytoxi) expression to control for loading. (B ) Quantitation of SP-A, SP-B, SP-C, ENaC␣, ENaC␤, and ENaC␥ subunit mRNA levels relative to levels of cytochrome oxidase mRNA. Autoradiographs were scanned and bands quantified using a scanning densitometer from Molecular Dynamics. Asterisks indicate statistical significance. lung maturation and increase lung compliance (2). The lung defect in GR-null mice demonstrates the important physiologic role of endogenous glucocorticoids just before birth. We have found clear evidence of abrogated remodeling of lung cellular architecture, which is very similar to changes observed following surgical removal of the fetal adrenal glands (35). There is a lack of septal thinning and an alteration in normal epithelial cell differentiation, primarily causing a reduction in the number of differentiated type I epithelial cells. This defect of incomplete epithelial cell differentiation complements the findings from another GR-null mouse line where there is a lack of increased expression in the lung of midkine, an important cell growth factor, 3–4 d before birth (16). Surprisingly, we demonstrate that differentiation into the type II EC phenotype is accentuated in the absence of GR-mediated 618 AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 30 2004 Figure 5. Northern blot analysis of T1␣, aquaporin-5, and aquaporin-1 mRNA in Day 18.5 pc fetal lung of WT (⫹/⫹) and GR-null (⫺/⫺) mice. Total RNA samples (7.5 ␮g/lane) were loaded onto a 1.2% agarose gel and separated by electrophoresis, then transferred to a nylon membrane. Filters were hybridized with riboprobes as described in Materials and Methods and rehybridized for cytochrome oxidase (cytoxi) expression to control for loading. Washed filters were also exposed to a phosphoimaging plate (Molecular Dynamics) and mRNA levels quantitated using ImageQuant software (Molecular Dynamics). Asterisks indicate statistical significance. glucocorticoid signaling and provide evidence that pulmonary surfactant can be produced and secreted onto the lung lumen in GRnull mice. This finding is consistent with a previous study in fetal sheep that demonstrated that fetal hypophysectomy, which abolishes the preparturient increase in circulating fetal cortisol concentrations, greatly increases the number of type II and reduces the number of type I ECs; this effect was abolished by the reinfusion of cortisol (36). The mechanisms for these changes in EC proportions are unknown, although an alteration in lung compliance and expansion may be involved (see below). Nevertheless, our findings, in combination with those of Crone and coworkers (36), clearly indicate that differentiation into the type II EC phenotype is not dependent upon the action of glucocorticoids. Although expression of the surfactant protein genes in type II ECs was found to be reduced, it was not abolished in GR-null mice and this probably reflects the continued action of other signals such as those activating the cAMP pathway (1). The critical requirement for cAMP signaling for the SP-D gene is shown in cAMP response element–binding protein-null mice, where there is a severe effect on its full induction at birth (37). It is also possible that glucocorticoids could signal via the mineralocorticoid receptor (MR), sometimes referred to as the type I GR, but recent studies indicate that MR is not present in alveolar ECs either in the adult lung or at birth (38). Finally, non–receptor-mediated rapid nongenomic actions of glucocorticoids have been reported in immune cells and the cardiovascular system, but have not been described so far in the developing respiratory system. Therefore, from our results we conclude that glucocorticoid signaling via GR is not critical for respiratory surfactant production, although surfactant composition and phosphatidylcholine content need to be determined in the GR-null lung to fully assess the normal functioning of secreted pulmonary surfactant (39). The secretion of lung liquid plays a critical role in the development of the fetal lung, whereas the withdrawal of this liquid from the future airways at birth is an important process allowing the transition to air breathing after birth (40). Membrane chan- nels that establish transmembranous ionic gradients and promote the unidirectional movement of water, such as the ENaC and the aquaporins, are very important in the lung and a number of the components of these channels have been reported as targets for glucocoticoid induction in the lung before birth (11, 27). We find evidence for reduced expression of ENaC␥ and aquaporin-1 in the lung of GR-null mice before birth. All three ENaC subunit genes are rapidly induced by dexamethasone in vitro in the type II epithelial A549 cell line (11), but in the lung of GR-null mice expression of the ENaC␣ and ␤ subunit is relatively unaltered compared with the lungs of WT mice. These results are similar to an initial study on the amilioridesensitive sodium channel in the lung of GR-null mice (11), yet further analysis of channel activities, unidirectional ion fluxes, and fluid flow rates in GR-null mice are required to more clearly define an overall defect in lung fluid flux across the epithelium. It is well established that the differentiated state of ECs is closely regulated by the degree of mechanical strain imposed on them (41, 42), which is predominantly determined by the degree of lung expansion in vivo (40). Indeed, alterations in fetal lung expansion, caused by changes in lung luminal volume, have a profound effect on the differentiated state of ECs (43). Increases in fetal lung expansion induces type II to type I EC transdifferentiation via an intermediate cell type and can reduce the proportions of type II cells from ⵑ 30% to ⬍ 2% within 10 d (19). On the other hand, reductions in fetal lung expansion promote an increase in type II EC proportions, most probably via transdifferentiation of type I into type II ECs (20); similar observations have been made in vitro (41). Thus, it is possible that a reduction in fetal lung expansion is responsible for the changes in epithelial cell proportions we observed in GR-null fetal mice. This contention is consistent with our finding of altered tissue structure in the lungs of GR-null fetal mice, as reductions in lung expansion induce similar structural changes in the lungs of fetal sheep (43). Such a reduction in lung expansion could have resulted from a defect in the lung liquid secretory mechanism, due to a reduction in the expression of ion channels as outlined above, resulting in reduced liquid secretion into the future airways. Alternatively, the altered lung structure in GRnull fetal mice, particularly the hypercellularity, greater tissue volumes, and thicker terminal airway walls, may contribute to a reduction in lung expansion due to a reduction in lung tissue compliance. Fetal lung expansion is primarily determined by the ability of the fetal upper airway, particularly the glottis, to retain liquid within the future airways, which maintains an internal distending pressure on the lungs of 1–2 mm Hg at rest (40). Consequently, assuming that adductor activity of the glottis is not different between WT and GR-null fetal mice, the degree of lung expansion will primarily be determined by the compliance of lung tissue; a reduction in tissue compliance will reduce the degree of lung expansion for any given distending pressure. The finding that type I ECs predominate in the lung of WT fetal mice at Day 18.5 pc (just before birth) is consistent with the findings of other studies in which it has been shown that type I ECs predominate in the lung before birth (19, 20, 23); this is thought to result from the higher degree of basal lung expansion in the fetus, compared with the newborn (19). The higher proportion of undifferentiated epithelial stem cells in GR-null fetal mice indicates that abolition of signaling via the GR may delay EC differentiation into both phenotypes. The mechanisms involved are unknown, but it is clear that ultimately, differentiation into the type I or type II cell phenotype is not absolutely dependent upon GR signaling. Recent evidence indicates that both type I and type II ECs are capable of transdifferentiation, and that the mechanical strain experienced by the cells is a critical factor regulating the pathways that ultimately determine the phenotype Cole, Solomon, Van Driel, et al.: Alveolar ECs in GR-Null Mice of each EC (20, 41, 42). Thus, it is possible that corticosteroids may facilitate the activation of pathways leading to the differentiation of undifferentiated epithelial stem cells, but determination of the mature phenotype is dependent upon other factors. In summary, analysis of the lung in the GR-null fetal mouse indicates that glucocorticoid signling via GR is not essential for surfactant production. Analysis of epithelial cell types using electron microscopy and cell-specific markers reveals a marked reduction in differentiated type I AECs before birth, indicating that GR signaling either directly or indirectly mediates differentiation of ECs into this phenotype. The absence of GR-mediated glucocorticoid signaling leads to a profound alteration to the development of the terminal respiratory units of the lung, which results in severe respiratory dysfunction at birth. Acknowledgments: The authors thank Prof. Richard Harding for helpful discussions. S.J.R. is supported by an Australian Research Fellowship from the Australian Research Council (ARC) of Australia. T.J.C. is the Biochemistry Fund Fellow at the University of Melbourne, Australia. S.B.H. is a Senior Research Fellow of the National Health and Medical Research Council (NH&MRC) of Australia. This work was supported by grants from the ARC and NH&MRC of Australia. References 1. Mendelson, C. R. 2000. Role of transcription factors in fetal lung development and surfactant protein gene expression. Annu. Rev. Physiol. 62:875–915. 2. Lyons, C. A., and T. J. Garite. 2002. Corticosteroids and fetal pulmonary maturity. Clin. Obstet. Gynecol. 45:35–41. 3. Newnham, J. P., T. J. Moss, I. Nitsos, and D. M. Sloboda. 2002. Antenatal corticosteroids: the good, the bad and the unknown. Curr. Opin. Obstet. Gynecol. 14:607–612. 4. Hanson, M. 2002. Birth weight and the fetal origins of adult disease. Pediatr. Res. 52:473–474. 5. Dodic, M., V. Hantzis, J. Duncan, S. Rees, I. Koukoulas, K. Johnson, E. M. Wintour, and K. Moritz. 2002. Programming effects of short prenatal exposure to cortisol. FASEB J. 16:1017–1026. 6. McKenna, N. J., and B. W. O’Malley. 2002. Combinatorial control of gene expression by nuclear receptors and coregulators. Cell 108:465–474. 7. Robinson-Rechavi, M., H. E. Garcia, and V. Laudet. 2003. The nuclear receptor superfamily. J. Cell Sci. 116:585–586. 8. Snyder, J. M., H. F. Rodgers, J. A. O’Brien, N. Mahli, S. A. Magliato, and P. L. Durham. 1992. Glucocorticoid effects on rabbit fetal lung maturation in vivo: an ultrastructural morphometric study. Anat. Rec. 232:133–140. 9. Wallace, M. J., S. B. Hooper, and R. Harding. 1995. Effects of elevated fetal cortisol concentrations on the volume, secretion, and reabsorption of lung liquid. Am. J. Physiol. 269:R881–R887. 10. Wallace, M. J., S. B. Hooper, and R. Harding. 1996. Role of the adrenal glands in the maturation of lung liquid secretory mechanisms in fetal sheep. Am. J. Physiol. 270:R33–R40. 11. Itani, O. A., S. D. Auerbach, R. F. Husted, K. A. Volk, S. Ageloff, M. A. Knepper, J. B. Stokes, and C. P. Thomas. 2002. Glucocorticoid-stimulated lung epithelial Na(⫹) transport is associated with regulated ENaC and sgk1 expression. Am. J. Physiol. Lung Cell. Mol. Physiol. 282:L631–L641. 12. Cole, T. J., J. A. Blendy, A. P. Monaghan, K. Krieglstein, W. Schmid, A. Aguzzi, G. Fantuzzi, E. Hummler, K. Unsicker, and G. Schutz. 1995. Targeted disruption of the glucocorticoid receptor gene blocks adrenergic chromaffin cell development and severely retards lung maturation. Genes Dev. 9:1608–1621. 13. Brewer, J. A., O. Kanagawa, B. P. Sleckman, and L. J. Muglia. 2002. Thymocyte apoptosis induced by T cell activation is mediated by glucocorticoids in vivo. J. Immunol. 169:1837–1843. 14. Muglia, L. J., D. S. Bae, T. T. Brown, S. K. Vogt, J. G. Alvarez, M. E. Sunday, and J. A. Majzoub. 1999. Proliferation and differentiation defects during lung development in corticotropin-releasing hormone-deficient mice. Am. J. Respir. Cell Mol. Biol. 20:181–188. 15. Reichardt, H. M., K. H. Kaestner, J. Tuckermann, O. Kretz, O. Wessely, R. Bock, P. Gass, W. Schmid, P. Herrlich, P. Angel, and G. Schutz. 1998. DNA binding of the glucocorticoid receptor is not essential for survival. Cell 93:531–541. 16. Kaplan, F., J. Comber, R. Sladek, T. J. Hudson, L. J. Muglia, T. Macrae, S. Gagnon, M. Asada, J. A. Brewer, and N. B. Sweezey. 2003. The growth factor midkine is modulated by both glucocorticoid and retinoid in fetal lung development. Am. J. Respir. Cell Mol. Biol. 28:33–41. 17. Cole, T. J., H. J. Harris, I. Hoong, N. Solomon, R. Smith, Z. Krozowski, and M. J. Fullerton. 1999. The glucocorticoid receptor is essential for maintaining basal and dexamethasone-induced repression of the murine corticosteroidbinding globulin gene. Mol. Cell Endocrinol. 154:29–36. 619 18. Shepherdley, C. A., C. B. Daniels, S. Orgeig, S. J. Richardson, B. K. Evans, and V. M. Darras. 2002. Glucocorticoids, thyroid hormones, and iodothyronine deiodinases in embryonic saltwater crocodiles. Am. J. Physiol. Regul. Integr. Comp. Physiol. 283:R1155–R1163. 19. Flecknoe, S., R. Harding, G. Maritz, and S. B. Hooper. 2000. Increased lung expansion alters the proportions of type I and type II alveolar epithelial cells in fetal sheep. Am. J. Physiol. Lung Cell. Mol. Physiol. 278:L1180–L1185. 20. Flecknoe, S. J., M. J. Wallace, R. Harding, and S. B. Hooper. 2002. Determination of alveolar epithelial cell phenotypes in fetal sheep: evidence for the involvement of basal lung expansion. J. Physiol. 542:245–253. 21. Crapo, J. D., S. L. Young, E. K. Fram, K. E. Pinkerton, B. E. Barry, and R. O. Crapo. 1983. Morphometric characteristics of cells in the alveolar region of mammalian lungs. Am. Rev. Respir. Dis. 128:S42–S46. 22. Crapo, J. D., B. E. Barry, P. Gehr, M. Bachofen, and E. R. Weibel. 1982. Cell number and cell characteristics of the normal human lung. Am. Rev. Respir. Dis. 126:332–337. 23. Alcorn, D. G., T. M. Adamson, J. E. Maloney, and P. M. Robinson. 1981. A morphologic and morphometric analysis of fetal lung development in the sheep. Anat. Rec. 201:655–667. 24. Williams, M. C. 1977. Conversion of lamellar body membranes into tubular myelin in alveoli of fetal rat lungs. J. Cell Biol. 72:260–277. 25. Nardo, L., G. Maritz, R. Harding, and S. B. Hooper. 2000. Changes in lung structure and cellular division induced by tracheal obstruction in fetal sheep. Exp. Lung Res. 26:105–119. 26. Rishi, A. K., M. Joyce-Brady, J. Fisher, L. G. Dobbs, J. Floros, J. VanderSpek, J. S. Brody, and M. C. Williams. 1995. Cloning, characterization, and development expression of a rat lung alveolar type I cell gene in embryonic endodermal and neural derivatives. Dev. Biol. 167:294–306. 27. Verkman, A. S., M. A. Matthay, and Y. Song. 2000. Aquaporin water channels and lung physiology. Am. J. Physiol. Lung Cell. Mol. Physiol. 278:L867– L879. 28. Williams, M. C. 2003. Alveolar type I cells: molecular phenotype and development. Annu. Rev. Physiol. 65:669–695. 29. King, L. S., S. Nielsen, and P. Agre. 1996. Aquaporin-1 water channel protein in lung: ontogeny, steroid-induced expression, and distribution in rat. J. Clin. Invest. 97:2183–2191. 30. Christoffels, V. M., T. Grange, K. H. Kaestner, T. J. Cole, G. J. Darlington, C. M. Croniger, and W. H. Lamers. 1998. Glucocorticoid receptor, C/ EBP, HNF3, and protein kinase A coordinately activate the glucocorticoid response unit of the carbamoylphosphate synthetase I gene. Mol. Cell. Biol. 18:6305–6315. 31. Finotto, S., K. Krieglstein, A. Schober, F. Deimling, K. Lindner, B. Bruhl, K. Beier, J. Metz, J. E. Garcia-Arraras, J. L. Roig-Lopez, P. Monaghan, W. Schmid, T. J. Cole, C. Kellendonk, F. Tronche, G. Schutz, and K. Unsicker. 1999. Analysis of mice carrying targeted mutations of the glucocorticoid receptor gene argues against an essential role of glucocorticoid signalling for generating adrenal chromaffin cells. Development 126:2935–2944. 32. Purton, J. F., R. L. Boyd, T. J. Cole, and D. I. Godfrey. 2000. Intrathymic T cell development and selection proceeds normally in the absence of glucocorticoid receptor signaling. Immunity 13:179–186. 33. Godfrey, D. I., J. F. Purton, R. L. Boyd, and T. J. Cole. 2000. Stress-free Tcell development: glucocorticoids are not obligatory. Immunol. Today 21: 606–611. 34. Purton, J. F., Y. Zhan, D. R. Liddicoat, C. L. Hardy, A. M. Lew, T. J. Cole, and D. I. Godfrey. 2002. Glucocorticoid receptor deficient thymic and peripheral T cells develop normally in adult mice. Eur. J. Immunol. 32:3546– 3555. 35. Liggins, G. C. 1976. Adrenocortical-related maturational events in the fetus. Am. J. Obstet. Gynecol. 126:931–941. 36. Crone, R. K., P. Davies, G. C. Liggins, and L. Reid. 1983. The effects of hypophysectomy, thyroidectomy, and postoperative infusion of cortisol or adrenocorticotrophin on the structure of the ovine fetal lung. J. Dev. Physiol. 5:281–288. 37. Mantamadiotis, T., T. Lemberger, S. C. Bleckmann, H. Kern, O. Kretz, A. Martin Villalba, F. Tronche, C. Kellendonk, D. Gau, J. Kapfhammer, C. Otto, W. Schmid, and G. Schutz. 2002. Disruption of CREB function in brain leads to neurodegeneration. Nat. Genet. 31:47–54. 38. Suzuki, T., H. Sasano, S. Suzuki, G. Hirasawa, J. Takeyama, Y. Muramatsu, F. Date, H. Nagura, and Z. S. Krozowski. 1998. 11Beta-hydroxysteroid dehydrogenase type 2 in human lung: possible regulator of mineralocorticoid action. J. Clin. Endocrinol. Metab. 83:4022–4025. 39. Nanjundan, M., and F. Possmayer. 2001. Pulmonary lipid phosphate phosphohydrolase in plasma membrane signalling platforms. Biochem. J. 358:637– 646. 40. Harding, R., and S. B. Hooper. 1996. Regulation of lung expansion and lung growth before birth. J. Appl. Physiol. 81:209–224. 41. Danto, S. I., J. M. Shannon, Z. Borok, S. M. Zabski, and E. D. Crandall. 1995. Reversible transdifferentiation of alveolar epithelial cells. Am. J. Respir. Cell Mol. Biol. 12:497–502. 42. Gutierrez, J. A., R. F. Gonzalez, and L. G. Dobbs. 1998. Mechanical distension modulates pulmonary alveolar epithelial phenotypic expression in vitro. Am. J. Physiol. 274:L196–L202. 43. Alcorn, D., T. M. Adamson, T. F. Lambert, J. E. Maloney, B. C. Ritchie, and P. M. Robinson. 1977. Morphological effects of chronic tracheal ligation and drainage in the fetal lamb lung. J. Anat. 123:649–660.