Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Parathyroid hormone-related peptide (PTHrP) and the parathyroid hormone/parathyroid hormone-related peptide (PTH/PTHrP) receptor are important regulators of cell growth and differentiation during development (1). PTHrP acts mostly as a paracrine factor regulating epithelial-mesenchymal interactions or as an autocrine factor in skeleton (2), mammary gland (5, 6), pancreas (7), developing hair follicles (8), and in lung cultures (9). PTHrP and the PTH/PTHrP receptor are both expressed in adult lung and early in the developing lung. In murine lung epithelium, PTHrP can be detected on embryonic day (E)12.5 (13), and the receptor can be detected in the surrounding mesenchyme (16). In adult lung, both are expressed in the alveolar epithelium and in isolated type II cells (10, 11, 17, 18).

Because PTHrP and PTH/PTHrP receptor null mutant mice die at birth of respiratory failure (2, 3), we wondered if abnormal regulation of peripheral lung genes might contribute to the inability to sustain respiration. A number of peripheral lung genes are upregulated before birth, including surfactant proteins and the type I cell genes T1 and aquaporin-5 (Aqp-5) (19). Although it has been hypothesized that the serious bone abnormalities in receptor null animals cause respiratory failure owing to nondistensibility of the rib cage, there are a number of other possibilities that could contribute to respiratory failure, including altered expression of surfactant apoproteins, insufficient surfactant lipids, failed morphogenesis of type I cells, and others (3). Nevertheless, rescued PTHrP null mice, where the bone malformation is overcome by expression of a constitutively active receptor, survive at birth but die before 2 mo of age of unknown causes (22).

In rodents, PTHrP protein is synthesized from a single transcript but is subsequently processed (2) into at least three different peptides that appear to act through different receptors (amino acids 1-36, 38-80 or 100, and 107- 139) (1, 23). These peptides regulate calcium transport (24), and may mediate cellular proliferation, differentiation, and apoptosis (1, 23) in various tissues. In bone, where its effects have been studied in detail, PTHrP delays chondrocyte differentiation, allowing for continued cell proliferation (3, 25). PTHrP overexpression in transgenic mice alters hair follicle development (8) and impairs branching morphogenesis in mammary gland (26). PTHrP signals by binding to its cognate receptor(s) (27) or by direct translocation to the nucleus/nucleolus directed by a nuclear targeting sequence (30, 31). The PTH/PTHrP receptor, for which the animals used in this study carry a null mutation, responds to both PTH and PTHrP, is a G protein-coupled receptor, and signals through adenylate cyclase and the phospholipase C systems (27, 28).

Other investigators have shown that in vitro treatment of cultured type II cells with synthetic PTHrP peptide (1- 34) increases the number of lamellar bodies and expression of alkaline phosphatase, and stimulates secretion of phosphatidylcholine (PC) (11). These findings suggest that PTHrP peptide (1) promotes the alveolar type II cell phenotype in vitro (11). Instillation into the lung of anti-PTHrP antibodies increases expression of proliferating cell nuclear antigen in alveolar cells, suggesting that PTHrP may inhibit cell proliferation (12). Although PTHrP does not stimulate PC synthesis in isolated type II cells, other investigators have shown that PTHrP treatment of lung explants increases PC synthesis (9). This suggests that PTHrP may signal via a complex mesenchymal-epithelial pathway. Therefore the PTHrP and PTH/PTHrP receptor pathway could regulate lung genes involved in the maturation of the lung before birth.

To explore the role of the PTH/PTHrP receptor in lung morphogenesis and alveolar epithelial cell differentiation, we determined expression levels of a group of marker gene messenger RNAs (mRNAs) in late fetal mice carrying a null mutation in the PTH/PTHrP receptor gene. The genes selected include differentiation markers for the type I and type II phenotypes as well as growth factors that regulate branching morphogenesis during development. These include T1 (20) and Aqp-5 (21), both apical membrane constituents of the type I cell, and caveolin-1 (32), a structural protein expressed by both type I and endothelial cells. Surfactant protein (SP)-A, -B, -C, -D mRNAs, whose expression patterns have been extensively characterized, were used as type II cell and/or Clara cell markers. Fibroblast growth factor (FGF)-7 (or keratinocyte growth factor [KGF]) and FGF-10, important regulators of lung morphogenesis and differentiation, were used as mesenchymal markers (33, 34).

Our data show that inactivation of the PTH/PTHrP receptor decreases expression of Aqp-5 mRNA but not that of other peripheral lung genes. As this water channel is believed to mediate high rates of transepithelial water fluxes (35), its downregulation may contribute to neonatal respiratory failure in these mice by inhibiting fluid clearance during the transition to an air-filled lung. The PTH/PTHrP receptor otherwise does not play a major role in lung morphogenesis and in the differentiation of distal epithelial cells before birth.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Mice

The PTH/PTHrP type I receptor null mice, kindly provided by Dr. Henry Kronenberg (Endocrine Unit, Massachusetts General Hospital, Boston, MA), carry a deletion in the PTH/PTHrP receptor gene from exon 2 to the termination codon, which encodes most of this receptor (3). In C57BL/6, 129/SvJ, and MF-1 genetic backgrounds, these mice die between E12.5 and E14.5 of unknown causes. Null mutant homozygous Black Swiss mice survive till birth but die within a few minutes of respiratory failure and have reduced body size from at least E9.5. Histologic evaluation at E9.5 and E12.5 reveals morphologically normal organ development except the animals and all organs are smaller than normal (3).

We studied gene expression in E18.5 homozygous null, heterozygous, and wild-type Black Swiss fetuses. This developmental age was selected because both branching and late developmental changes in gene expression patterns are well established. In addition, it assured viability of the fetuses. Fetuses were removed by hysterotomy, and the lungs were isolated for RNA analysis and histologic studies.

Histology and In Situ Hybridization of Fetal Lungs

Tissues were fixed in freshly prepared 4% paraformaldehyde in phosphate-buffered saline and embedded in paraffin as previously described (36). Lung sections (5 µm) were stained with hematoxylin and eosin (H&E) to evaluate lung morphology. To determine if the general pattern of marker gene expression was normal, we localized SP-C mRNA by in situ hybridization as described previously (36). A 574-bp SP-C complementary DNA (cDNA) fragment cloned in pBluescript KS (36) was used as a template to synthesize [³⁵S]-labeled sense and antisense riboprobes (37). Hybridized slides were developed after 1 wk of exposure.

Northern Blot Analysis

Total RNA was isolated using TRIazol reagent (Life Technologies, Inc. Grand Island, NY). A total of 10 to 20 µg was electrophoresed, blotted, and hybridized as described previously (38). Probes were labeled by the hexamer-random primed method using [³²P]deoxycytidine triphosphate (39). Mouse T1, rat SP-A, SP-B, SP-C probes (36), and mouse Aqp-5 probe were obtained by reverse transcription/polymerase chain reaction (RT-PCR) and cloned into pBluescript. Human caveolin-1 probe, which recognizes murine species, was kindly provided by Drs. P. and C. Fielding, University of California, San Francisco, CA, and rat SP-D probe, by Dr. J. Fisher, University of Colorado, Denver, CO. FGF-7 and FGF-10 probes were generated by RT-PCR as described by Lebeche and coworkers (40). Blots generated with total RNA isolated from three or four wild-type, heterozygous, and homozygous embryonic lungs were hybridized with several probes and analyzed in a Molecular Dynamics densitometer (Molecular Dynamics, Sunnyvale, CA). Densitometric values, normalized to 18S RNA levels, were analyzed statistically using the Statview 5.12 software (Brain Power, Inc., Calabasas, CA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Morphology

We analyzed the gross and microscopic morphology of the lungs of E18.5 homozygous mice and compared it with the lungs of their wild-type littermates. At this gestational stage, normal lungs are highly branched, bronchioles are well defined by columnar epithelium and subjacent smooth muscle, and distal airways are lined by cuboidal epithelial cells that express surfactant genes (36). Homozygous PTH/ PTHrP receptor null mice are smaller than their wild-type littermates, and their lungs appear to be proportional to their body size. After hysterotomy, wild-type E18.5 embryos breathe so their lungs are expanded although still immature. Homozygous null mice fail to inflate their lungs. Although homozygous lungs are smaller than the wild-type lungs and are not fully inflated, we show by H&E staining that the branching pattern and alveolar structure appear normal (Figure 1). Therefore, the lack of PTH/PTHrP receptor produces no detectable differences in the morphology or branching of the distal airways.

View larger version (125K): [in this window] [in a new window]   Figure 1.   H&E staining of E18.5 wild-type (+/+) and homozygous (/) mouse lung sections. Although there is a marked difference in size between (+/+) and (/) lungs, and (/) lungs are not fully inflated, branching patterns and morphology of the distal lung are not altered. Original magnification: A, B, ×50; C, D, ×150.

The SP-C gene is expressed in mouse lung from E10 or E11 and has a very well characterized pattern of expression at the leading edge of the distal alveolar epithelium. It is a definitive marker of the type II cell phenotype in the mature lung and has been used in numerous studies to identify distal epithelial cells (19). We analyzed the SP-C mRNA expression pattern by in situ hybridization (Figure 2) and observed no difference in localization or concentrations of SP-C mRNA in distal airways between null and control animals, suggesting that the differentiation of the distal epithelium is not affected by the lack of PTH/PTHrP receptor. To evaluate this further, we analyzed expression of a large group of marker genes for the distal lung.

View larger version (124K): [in this window] [in a new window]   Figure 2.   In situ hybridization analysis of SP-C mRNA expression in E18.5 lung of wild-type (+/+) and PTH/PTHrP receptor homozygous (/) null mutant mice. Sections were hybridized with a [35S]-labeled SP-C probe. Slides were exposed for one week. Original magnification: ×75.

Expression of Lung Marker Genes

Representative Northern blots of genes specific for alveolar type I and type II cells and fibroblasts in E18.5 lungs are shown in Figure 3A. In the densitometric analysis of Northern blots, performed with total RNA from three to four different embryos run on separate Northern blots, we observed a 2.6-fold reduction in Aqp-5 mRNA. All the other markers tested were unaffected by the lack of PTH/ PTHrP receptor (Figure 3B).

View larger version (40K): [in this window] [in a new window]   View larger version (23K): [in this window] [in a new window]   Figure 3.   (A) Northern blot analysis of total RNA (10-20 µg) isolated from E18.5 lungs shows that Aqp-5, but not other peripheral lung genes, is reduced in PTH/PTHrP receptor null mice. (+/+) indicates RNA isolated from wild-type mice, (+/) from heterozygous mice, and (/) from homozygous null mice. Blots were hybridized with [32P]-labeled probes for Aqp-5, caveolin-1, SP-A, -B, -C, -D, FGF-7, and FGF-10, and normalized to 18S RNA levels. (B) Densitometric analysis of Northern blots using RNA isolated from three to five mouse embryos. Densitometric relative values are the mean of two to five blots ± standard error.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Our data suggest that the PTH/PTHrP receptor does not play a critical role in the signaling events that regulate the development of lung pattern or general characteristics of alveolar cell differentiation. We showed first that the morphology of the lung of null mice does not differ from that of wild-type control mice. The normal morphology is consistent with the finding that expression of both FGF-7 and FGF-10 are unaffected by the PTH/PTHrP receptor mutation. A number of studies have shown that these growth factors play critical roles in early lung development, namely regulation of cell proliferation and regulation of budding, respectively (40). FGF-10 in particular is important in early events in lung development because a null mutation in the FGF-10 gene results in complete agenesis of the lungs (33).

In addition, we showed that targeting of the peripheral lung gene SP-C to the correct cell population matches that observed in the wild-type animals, suggesting again little effect of loss of the PTH/PTHrP receptor on normal developmental control mechanisms. SP-C is believed to be the earliest gene expressed that is a marker for the peripheral lung epithelium. The normal pattern of expression we observed in the E18.5 embryos studied here suggests that earlier expression patterns of SP-C are also likely to be the same as those of control animals. Finally, we found that abundance of mRNAs of well-characterized alveolar cell marker genes, including the surfactant-associated proteins, was not affected by loss of the PTH/PTHrP receptor.

Our observations, however, should not be interpreted as suggesting that PTH/PTHrP has no role in lung development. Several observations argue to the contrary. Preliminary data (41) from mice with a null mutation in PTHrP (i.e., in ligand rather than the receptor) show that PTHrP deficiency results in lung hypoplasia, arrested canalicular development, impaired epithelial differentiation, fibroblastic lipid accumulation, and failed septation. Together with our findings, these observations suggest that PTH/PTHrP influences lung development via a mechanism other than the receptor we studied. These effects could be mediated by a second receptor, as hypothesized by other investigators, by translocation of PTH/PTHrP to the nucleus in an intracrine regulatory pathway, by nonlung-derived signaling molecules, or combinations of these. The presence of a second receptor that interacts with PTHrP midregion (24) has been demonstrated in some tissues of PTH/PTHrP receptor null mutant animals and in squamous carcinoma cells. This receptor seems to regulate intracellular calcium levels and fetal-placental calcium transport. Whether this receptor is expressed in the lung is unknown. Recently, a receptor specific for PTHrP was cloned from zebrafish, but its mammalian homologue has not been identified (29).

Of considerable importance is the finding that expression of Aqp-5 is markedly reduced in lungs of PTH/PTHrP receptor null fetuses. Given that expression of the other marker genes we tested was not affected by the PTH/ PTHrP receptor mutation, downregulation of Aqp-5 mRNA is probably due to inhibition of Aqp-5 transcription and not a result of a delay in overall lung maturation. Like other investigators (42), we find the Aqp-5 transcripts are easily detected in late gestational murine lung, similar to that of the rat (43). By Northern blot, we have detected low concentrations of Aqp-5 mRNA in E15.5 mouse embryonic lung (unpublished results); this is earlier than has been reported previously. Aqp-5 mRNA increases in the first postnatal week to about 3-fold its late fetal expression level (42), presumably to facilitate fluid clearance. The adult level appears to be lower than that of the immediate postnatal lung.

Aqp-5 was originally cloned from salivary gland (44). Its expression in adult rats is limited to serous acinar cells of salivary and lacrimal glands, corneal epithelial cells, and type I cells of the peripheral lung (45). In all locations, it is localized to apical plasma membranes. To date, there is little information on the molecular regulation of Aqp-5 transcription. Using primary type II cells, Borok and colleagues (46) have shown in the rat that FGF-7 (KGF), homologous serum, and cuboidal cell shape suppress Aqp-5 mRNA expression in vitro, but the signaling pathways have not been characterized. However, because it is not upregulated in the PTH/PTHrP receptor mutant lung, altered FGF-7 expression does not account for the reduced expression of Aqp-5. Like many other peripheral lung genes, Aqp-1, expressed in the vasculature of late fetal and adult lung, is markedly upregulated by prenatal glucocorticoids, but glucocorticoids apparently do not induce Aqp-5 at any time (35).

The specific function of Aqp-5 in lung is also not known but is under investigation. Permeability studies indicate that isolated rat type I cells are capable of unusually high transmembrane water fluxes, likely owing entirely or in part to the presence of Aqp-5 protein (47). Thus, these cells may play a predominant role in clearance of alveolar fluid at birth or resolution of pulmonary edema postnatally (21, 42). Low concentrations of Aqp-5 in PTHrP receptor null lungs could result in poor clearance of lung alveolar fluid at birth although recent data from an Aqp-5 null mouse argue against this. Although there is an unexplained increase in prenatal mortality, the surviving Aqp-5 null mutant animals appear able to initiate and sustain respiration after birth.

Aqp-5-deficient mice have defects in secretion of saliva in response to the cholinergic agonist pilocarpine (48). The secreted volume is reduced to about one-third normal, and the saliva is hypertonic due to increased concentrations of Na⁺, K⁺, and Cl. Protein secretion is unchanged from control animals. Thus, Aqp-5 appears to act as a highly permeable channel for near-isosmolar water transport. In contrast to salivary and lacrimal glands, which are major salt- and water-secreting organs, the postnatal peripheral lung epithelium is primarily an absorptive surface. Although Aqp-5 is likely to be involved in this function, this is unproven and the lung phenotype in Aqp-5 null mice has not yet been reported.

We conclude, therefore, that the PTH/PTHrP receptor plays little role in overall regulation of peripheral lung genes, including T1. This is surprising because T1 is highly homologous to a known target gene of PTHrP (49) expressed by a bone cell line (accession no. X95088). That T1 expression is not decreased in the PTH/PTHrP receptor null mice suggests that these similar genes are differentially regulated in bone and lung.