INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE PROCESS BY WHICH CELL fates are determined in an embryo is known as pattern formation (18). Intercellular communication, which mediates the biological interactions between mesenchyme and epithelium, is central to this process. These biological interactions control cellular proliferation and differentiation during organ and tissue development. Mesenchymal and epithelial interactions are necessary for the development of many organ systems, including those of the gastrointestinal, integument, urogenital, and respiratory systems (11). Lung development serves as a classical model for such biological interactions (28). Formed initially as an outpouching of the primitive foregut, the lung subsequently undergoes growth and branching of the primitive respiratory epithelium into the surrounding mesenchyme to form the bronchial tree. A complex process of interactions among cells, cytokines, extracellular matrix, and cell membrane receptors is necessary for lung morphogenesis and regional specification of the respiratory system. Presently, the mechanisms linking extracellular signals to intracellular reactions, such as gene expression and proliferation in the developing lung, are not well understood (19).

An essential mechanism in the regulation of signal transduction involves the activities of phosphoproteins capable of reversible phosphorylation and dephosphorylation (14, 30). The ratio of these phosphoproteins in their phosphorylated and dephosphorylated states together with the relative activity of protein kinases and phosphatases determines the activity of a given target phosphoprotein. Normal signal transduction is therefore dependent on a complex interplay of protein kinases and phosphatases. In eukaryotic cells, there are two large families of phosphatases that are divided into either serine/threonine phosphatases or tyrosine phosphatases, depending on which respective amino acid residue is dephosphorylated (5). Protein phosphatase 2A (PP2A), a serine/threonine phosphatase, accounts for a large portion of total cellular phosphatase activity. Importantly, PP2A in cell culture studies has been shown to have a role in control of the cell cycle (4), growth and proliferation (25), and cell fate determination (16). This suggests that PP2A is likely to be a key signaling intermediate in the regulation of a number of cellular processes necessary for organ growth and development.

We have demonstrated previously that PP2A is broadly expressed during early lung development, becoming localized later predominantly in the earliest developing epithelium and endothelium as the lung matures (37). Because the role of PP2A in mammalian organogenesis remains largely unexplored, the present study demonstrates that PP2A has a significant role in the regulation of lung growth, regulating the cell cycle of mesenchymal cells and differentiation of the respiratory epithelium.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Fetal lung explant culture. Primordia of 14-day embryonic rat lung were microdissected from the embryo with the trachea intact and placed on prewetted porous, culture plate inserts (3-µm pore, 12 mm in diameter, Falcon) in six-well culture plates (Costar) with 1,500 µl of complete medium (BGJ_b with L-glutamine, 1% FCS, 100 U/ml penicillin, 100 µg/ml streptomycin, and 0.5 mg/ml ascorbic acid; GIBCO BRL). The PP2A inhibitors okadaic acid (OA) (3) and cantharidin (Can) (15) were added to the medium at 3, 6, and 9 nM and 1,200, 2,400, and 3,600 nM, respectively. The highest concentration of each inhibitor was equal to ~100× the IC₅₀ of that compound for PP2A. The PP2B inhibitor deltamethrin (Del) (7) was added to the medium at 5 and 10 µM, again with the highest concentration equal to ~100× the IC₅₀ for PP2B as an additional control for PP2A-specific effects. Control explants were incubated in medium containing the diluent DMSO. Each insert supported up to 10 lung blocks. The fetal lung explants were incubated in a 5% CO₂ incubator at 37°C for 48 h, and the medium with and without inhibitors was changed after 24 h in culture.

Determination of airway counts. Photographs of each individual explant were taken at the start of the experiment (0 h), at 24 h, and after 48 h in culture. The number of airway buds present in each photograph was counted by a single reviewer (B. K. Taylor), with both the treatment group and the time of incubation blinded to the reviewer. Any section of an airway bud with three sides was counted as a branch. The number of new airway buds is expressed as airway buds at 24 or 48 h minus those at 0 h.

Detection of 5-bromo-2'-deoxyuridine by immunocytochemistry. For the final 12 h of culture 5-bromo-2'-deoxyuridine (BrdU, 10 mM) was added to explant cultures with and without the phosphatase inhibitors. Lungs were fixed subsequently in 70% ethanol for 20 min at 4°C and cryoprotected overnight in 20% sucrose-optimum cutting temperature (OCT) embedding medium (Miles) at 4°C. Lungs were then transferred to cryomolds containing OCT and stored at 80°C. Tissue sections (4-6 µm) were cut using a cryostat (Reichert-Jung), thaw mounted on Superfrost-Plus slides (Fisher Scientific), and stored at 20°C. Tissue sections were allowed to warm to room temperature, briefly incubated in PBS. Slides were then incubated at 4°C in 1× saline-sodium citrate containing 1% Triton X-100 for 2 min and subjected to DNase digestion (50 U/ml) for 15 min to improve nuclear penetration. After a brief PBS rinse, slides were preincubated with 1% BSA for 20 min, washed (2 × 10 min in PBS), and incubated at room temperature for 1 h with an anti-BrdU-alkaline phosphatase-conjugated antibody (1:500 dilution, Boehringer Mannheim). Control slides were incubated with PBS alone. After removal of unbound primary antibodies by washing with PBS, the sections were incubated for 20 min with Alkaline Phosphatase Substrate Kit 1 (SK-5100, Vector Laboratories) and mounted with Vectashield Mounting Medium for Fluorescence (Vector Laboratories). Alkaline phosphatase activity was visualized by a color reaction using the Vector red alkaline phosphatase substrate that allows both bright-field and fluorescent detection. Finally, the slides were examined and photographed under an Olympus Vanox AHBS3 bright-field fluorescent microscope.

Hoechst DNA staining. Serial frozen sections were prepared as previously described and hydrated with 1× PBS with 0.001% Tween before staining. A 10 mg/10 ml stock solution of Hoechst DNA dye 33258 was diluted to 10 µM, and 1 µl was added to 200 µl of 1× PBS-0.001% Tween (1:200) for each section. The sections were then incubated for 1 h at room temperature, washed two times with distilled H₂O, and mounted with Vectashield Mounting Medium. Finally, the slides were examined and photographed as described previously.

Protein and DNA content. Fetal lung explants (10-13 fetal lungs pooled for each treatment group, repeated 3 times) were treated with OA (0-9 nM) for 48 h and assayed for DNA and protein content. Pooled explants were homogenized by polytron disruption in 100 µl of homogenate buffer containing 25 nM HEPES, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 2 µM leupeptin, and 100 µM EDTA, pH 8.0. Protein content was determined by fluorescamine assay using 5 µl of sample with BSA as a standard. Protein concentration is expressed as micrograms per microliter. The remaining homogenate was assayed for DNA content using the DNA dye (Hoescht 33258) and a DNA fluorometer (TKO, Pharmacia). DNA concentration is expressed as micrograms per milliliter.

Terminal deoxynucleotidyltransferase dUTP nick end labeling reaction. Lungs were fixed in 70% ethanol for 20 min at 4°C and cryoprotected overnight in 20% sucrose-OCT embedding medium (Miles) at 4°C. Lungs were then transferred to cryomolds containing OCT and stored at 80°C. Tissue sections (4-6 µm) were cut using a cryostat (Reichert-Jung), thaw mounted on Superfrost-Plus slides (Fisher Scientific), and stored at 20°C. The slides were rinsed with PBS, and the tissue sections were incubated in a permeabilization solution (0.1% Triton X-100 and 0.1% sodium citrate) for 2 min at 4°C. The slide sections were rinsed twice with PBS, and 50 µl of terminal deoxynucleotidyltransferase dUTP nick end labeling (TUNEL) reaction mixture (TUNEL Kit, Boehringer Mannheim) were added to each section. Sections were covered with parafilm and incubated for 60 min in a humidified chamber at 37°C. Subsequently, the sections were rinsed three times in PBS; 50 µl of Converter-AP (TUNEL Kit, Boehringer Mannheim) were added to each section, and the sections were covered with parafilm and incubated for 60 min in a humidified chamber at 37°C. After three washes with PBS at room temperature, labeled nuclei were detected with an alkaline phosphatase red substrate (Vector Laboratories) and mounted with Vectashield Mounting Medium for Fluorescence (Vector Laboratories). Finally, the slides were examined and photographed under an Olympus Vanox AHBS3 bright-field fluorescent microscope.

Cell cycle analysis. Cultured fetal lungs (15 per group, repeated 3 times) from each treatment group were dissociated in a 0.1% collagenase solution in a spinner flask at 37°C for 30 min. Cells were collected by centrifugation at 1,000 rpm for 7 min. Cells (5 × 10⁵) were resuspended in 1 ml of propidium iodide solution (7.5 × 10⁵ M propidium iodide, 10 mM NaCl, 0.01 M Tris base, 0.001% Nonidet P-40, and 700 U of RNase) and analyzed by fluorescent flow cytometry using a FACScan (Becton Dickinson).

Northern analysis. Total RNA was extracted from cultured fetal lungs (n = 15-17 lungs/group, repeated 2 times) using TRI Reagent (Molecular Research Center) and quantitated at 260 nm with an ultraviolet (UV) spectrophotometer (Shimadzu). Northern analysis of 8 µg of total RNA was performed by electrophoresis under denaturing conditions in a MOPS-formaldehyde 1.2% agarose gel with transfer to a charged nylon membrane (Zeta-Probe, Bio-Rad) by capillary action in high salt (20× saline-sodium phosphate-EDTA). Posttransfer membranes were UV cross-linked (Stratalinker, Stratagene) and stained with methylene blue to document equal transfer. Blots were hybridized with a surfactant protein C (SP-C) cDNA (gracious gift from Mary Williams, Boston University) and a cDNA for glyceraldehyde-3-phosphate dehydrogenase as a loading control and labeled by random priming (Ready to Go, Amersham) with hybridization and washes at 65°C. Relative amounts of mRNA after hybridization were determined by PhosphorImager autoradiography (Molecular Dynamics) and quantitated using the ImageQuant software package (Molecular Dynamics).

Statistics. Differences in airway number between treatment groups were determined by one-way ANOVA. New airway numbers are expressed as means ± SE. Pairwise multiple comparisons were made by the Student-Newman-Keuls method. P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Effect of protein phosphatase inhibition on branching morphogenesis. To determine a role for PP2A in lung development, 14-day fetal rat lungs were grown in culture with increasing concentrations of the cell-permeable PP2A inhibitors (OA and Can) or a PP2B inhibitor (Del) for 48 h. OA-treated explants (0, 3, 6, and 9 nM) and Can-treated explants (0, 1,200, 2,400, and 3,600 nM) demonstrated obvious decreases in new airway formation and overall size of the explant as the concentrations of the PP2A inhibitors were increased, with a maximal effect at the highest concentrations (Fig. 1). The Del-treated explants grew similar to control explants in both the number of new airway branches and size (Fig. 1). The effect of PP2A inhibition with OA and Can on airway branching was statistically significant at the highest concentrations (~100× IC₅₀), 9 nM and 3,600 nM, respectively, at both 24 and 48 h of culture (Fig. 2, A and B). New airway formation was significantly decreased by 30.2 and 36.1% at 24 and 48 h, respectively, in the 9 nM OA-treated explants and by 45.0 and 50.2% at 24 and 48 h, respectively, in the 3,600 nM Can-treated explants. PP2B inhibition with Del did not affect new airway development (Fig. 2C). Therefore, PP2A inhibition with two concentration-specific inhibitors resulted in a significant decrease, but not an arrest, in lung growth and airway formation.

View larger version (127K): [in this window] [in a new window]   Fig. 1.   Effect of protein phosphatase 2A (PP2A) and PP2B inhibition on fetal lung growth. Serial photographs of representative explants at 0, 24, and 48 h grown in culture with the pharmacological PP2A inhibitors okadaic acid (OA) and cantharidin (Can) or the PP2B inhibitor deltamethrin (Del) are shown. Magnification, ×20.

View larger version (201918K): [in this window] [in a new window]   Fig. 2.   Effect of PP2A and PP2B inhibition on new airway formation in 14-day gestation rat lung explants grown in culture with PP2A inhibitors OA (A) and Can (B) or PP2B inhibitor Del (C). Number of new airways formed above baseline number recorded at 0 h is shown for each treatment group of the 3 compounds (n = 14-22 explants/time point) after 24 and 48 h in culture. * P < 0.05 treatment group vs. control (1-way ANOVA).

Histological evaluation of fetal lungs. Inasmuch as a previous report describes severe kidney developmental defects with PP2A inhibition (27), fetal lungs were examined histologically. Representative control and OA-treated explant cryosections are demonstrated in Fig. 3. As shown, both the control and OA-treated explants are grossly similar. Branching airways are readily apparent in all treatment groups, and of note, there is no significant alteration of the epithelial and mesenchymal cell compartments. As expected from the previous data, there were fewer airway branches and the overall size of the explant was smaller in the 9 nM OA-treated sections. Del treatment produced no alteration in lung morphology as well (data not shown). Therefore, unlike the kidney, PP2A inhibition with OA did not alter lung morphology.

View larger version (142K): [in this window] [in a new window]   Fig. 3.   Effect of 48 h of PP2A inhibition with OA on cellular morphology of 14-day gestation fetal lungs. Representative hematoxylin and eosin stained sections from control (A) and OA-treated (9 nM, B) fetal lungs at ×100 are shown.

PP2A did not induce apoptosis in the developing fetal lung. Because PP2A inhibition has been shown to induce apoptosis in vitro in a number of cell lines (17, 21, 29), we sought to determine whether the decreased lung explant growth and airway branching were the result of increased apoptosis. To assess apoptosis, explant cryosections were subjected to the TUNEL reaction as a sensitive marker of DNA fragmentation, indicating programmed cell death. TUNEL-positive nuclear labeling in control and 9 nM OA-treated explants is shown in Fig. 4. As expected in normally growing embryonic lung tissue (23), there were TUNEL-positive nuclei in the control lung (Fig. 4, A and B) in both mesenchyme and epithelium. OA-treated explants demonstrated a dose-dependent decrease in TUNEL-positive nuclei, with a maximal effect at 9 nM. In the 9 nM OA treatment group, there were no TUNEL-positive nuclei in the explant sections (Fig. 4, D and E). The negative controls in both treatment groups (0 and 9 nM, Fig. 4, C and F, respectively), which were not treated with the enzyme terminal deoxynucleotidyltransferase, indicate the efficacy of the labeling. Therefore, PP2A inhibition decreases fetal lung growth by a mechanism that does not include increased apoptosis. These studies suggest that in the growing fetal lung, PP2A plays a positive role in the induction of apoptosis.

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Terminal deoxynucleotidyltransferase dUTP nick end labeling (TUNEL) reaction as marker of apoptosis in control and OA-treated explants. A and B: ×100 and ×400 views, respectively, of control (0 nM) explant subjected to TUNEL reaction demonstrating TUNEL-positive nuclei (red) in A and in B (arrows) showing normal background apoptosis found in developing lung. D and E: ×100 and ×400 views, respectively, of representative 9 nM OA-treated fetal lung that show no TUNEL-positive nuclei (i.e., no apoptosis). C and F: negative-control sections for TUNEL reaction for both 0 nM (C) and 9 nM OA-treated (F) fetal lungs lacking labeling reagent.

PP2A inhibition increases DNA synthesis and decreases protein synthesis. To determine whether the effects of PP2A inhibition on lung growth were due to inhibition of cell proliferation and not toxicity, DNA and protein content measurements on fetal lung explants treated in culture with 0, 3, 6, and 9 nM OA (n = 10-13 lungs pooled for each treatment group) were made after culture for 48 h. As shown in Fig. 5, PP2A inhibition with OA treatment produced a dose-dependent increase in DNA content (>4-fold increase at 9 nM) and a dose-dependent decrease in protein content (2.7-fold at 9 nM). These findings indicate that OA is not toxic to the fetal lung explants and suggest that PP2A regulates both DNA synthesis and protein synthesis in the fetal lung.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   DNA and protein content of fetal lung explants after 48 h of OA treatment. Fetal lung explants (10-13 fetal lungs for each treatment group, n = 3) were pooled after treatment in culture with 0, 3, 6, and 9 nM OA for 48 h. DNA content () was measured by DNA fluorometry using the DNA dye Hoechst 33258 and protein content () was measured by fluorescamine assay. Open bars, protein-to-DNA ratio.

PP2A regulates cell cycle in growing fetal lung. Because PP2A inhibition results in decreased lung growth that is not related to altered morphology or apoptosis, the effect of PP2A inhibition on cell division was examined. To determine whether the increase in DNA content was related to an increased rate of DNA synthesis, BrdU incorporation, as a marker of DNA synthesis and therefore cell division, was examined in tissue sections of 0 (control), 3, 6, and 9 nM OA-treated explants after 48 h in culture (Fig. 6). BrdU labels cells in the S phase of the cell cycle as they replicate their DNA and has been used to study dividing lung cells (6, 9). PP2A-inhibited explants demonstrated a dose-dependent increase in the number of BrdU-labeled cells, with maximal effect at 6-9 nM (Fig. 6). Immunostaining for BrdU coupled with Hoescht 33258 staining of DNA demonstrates that the cells with the highest BrdU uptake also had the highest DNA content (Fig. 7). The labeled cells were predominantly mesenchymal, with no appreciable increase in BrdU uptake in the respiratory epithelium. Treatment with Del (0-10 µM) did not affect DNA synthesis as measured by BrdU incorporation (data not shown). PP2A inhibition with OA did not result in an increase in cell number because the actual number of mesenchymal cells present did not change (Fig. 8). The increase in BrdU labeling of the mesenchymal cells indicates a cell-specific increase in DNA synthesis without increased cell division. Consistent with these findings are the DNA content results of fluorescent flow cytometric analysis of enzymatically dissociated fetal lung explants. PP2A inhibition was found to cause a release of the G₁/S checkpoint and a cell cycle arrest at the G₂/M phase of the cell cycle (Fig. 9). The number of cells in the quiescent phase (G₀/G₁) of the cell cycle decreased from 65% in control explants to 42% in the OA-treated explants, indicating accelerated entry into the S phase. Furthermore, there was an accumulation of cells in the G₂/M phase of the cell cycle in the OA-treated explants that exceeded control explants by greater than 300% (8 vs. 29%), indicating a specific cell cycle arrest at the G₂/M phase.

View larger version (79K): [in this window] [in a new window]   Fig. 6.   Effect of OA treatment on 5-bromo-2'-deoxyuridine (BrdU) incorporation in 14-day gestation fetal lungs treated with increasing concentrations of OA (A, control; B, 3 nM; C, 6 nM; D, 9 nM) for 48 h and pulsed with 10 µM BrdU for 12 h. Fetal lungs were sectioned and immunostained for BrdU with anti-BrdU antibody. Nuclei that incorporated BrdU are red.

View larger version (124K): [in this window] [in a new window]   Fig. 7.   BrdU immunostaining and DNA staining of serial sections of fetal lungs treated with OA for 48 h. A and B: serial sections of control (0 nM) OA-treated explants after immunostaining for BrdU (A, red nuclei) and DNA staining (B, blue nuclei) with Hoescht dye 33258. C and D: serial sections from OA-treated lungs stained for BrdU (C, red nuclei) and DNA (D, blue nuclei). Arrows, mesenchymal cells surrounding airway (a).

View larger version (15K): [in this window] [in a new window]   Fig. 8.   Mesenchyme cell number in 14-day gestation fetal lungs treated for 48 h with OA.

View larger version (19K): [in this window] [in a new window]   Fig. 9.   PP2A inhibition arrested fetal lung cells in G2/M phase of cell cycle. The 14-day fetal lung explants were cultured with 0 nM (A) and 9 nM OA (B) for 48 h (n = 15 for each treatment group) and then dissociated with collagenase and stained with propidium iodide. Cells (200,000) were analyzed for DNA content using fluorescent flow cytometry. Proportions of cells at various stages of cell cycle are plotted from representative experiment, but mean values from 2 separate experiments are provided.

PP2A regulates expression of SP-C mRNA. Given the cell-specific effects of PP2A inhibition on proliferation of the mesenchyme, altered differentiation was also explored as another possible mechanism of decreased fetal lung growth. We examined expression of the specific respiratory epithelial cell gene markers (SP-A, SP-B, SP-C, LAR, and CCSP) in explants treated with 0 or 9 nM OA for 48 h. Of the developmentally expressed surfactant proteins, only SP-C, the earliest marker of terminally differentiated respiratory epithelium (12, 35), was detectable at this gestational age. As shown in Fig. 10, the level of SP-C mRNA was decreased in the PP2A-inhibited explants.

View larger version (72K): [in this window] [in a new window]   Fig. 10.   Surfactant protein C (SP-C) expression in control and 9 nM OA-treated fetal lungs. Representative Northern analysis of 8 µg of total RNA extracted from control and 9 nM OA-treated fetal lungs (n = 15-17) probed with radiolabeled SP-C cDNA as marker of epithelial differentiation and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA as control is shown. The 28S and 18S ribosomal bands are shown as size references.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In this study, fetal lung explants in the early pseudoglandular stage of development were treated with concentration-specific PP2A inhibitors and monitored for airway branching and maturation. PP2A inhibition resulted in 1) decreased explant size and fewer airway branches without gross morphological changes, 2) decreased apoptosis, 3) increased mesenchymal cell DNA synthesis, 4) decreased differentiation of the respiratory epithelium, and 5) a cell cycle arrest at the G₂/M phase of the cell cycle. We believe that this is the first report to describe PP2A as a mediator of lung growth and development.

The signaling proteins that transduce membrane receptor developmental and growth events to the nucleus in the lung are as yet largely unidentified. The balance of phosphorylation and dephosphorylation of key regulatory proteins appears to be the predominant mechanism for transferring signals to the nucleus in many if not all cell types. Phosphatases in particular are necessary to balance the phosphorylation activated by receptor stimulation. The serine/threonine phosphatases PP2A and PP1 are necessary for normal development and cell fate determination. In the lung, PP2A and PP1 activities are developmentally regulated in whole lung homogenates peaking after birth (31). PP2A expression is also developmentally regulated, being broadly expressed by both epithelial and mesenchymal cells in the early fetal 14- to 17-day rat lung (37). By 19 days, expression predominates in the earliest forming airways with diminished expression at this stage in the mesenchyme or in the more differentiated bronchial epithelium (37). Because PP2A is likely to have many substrates in the developing lung and could therefore regulate many developmental processes, the role of PP2A, if any, in regulating lung development was unclear. In the present study, a combination of PP2A (OA and Can) and PP2B (Del) inhibitors was used to assay for the effects of these phosphatases on lung development. The concentrations used in these studies were based on the reported IC₅₀ of the respective enzyme inhibitors to be selective for PP2A rather than to the related phosphatase PP1 based on assays of cell protein lysates. Of note in intact cells (7a), rather than in previously studied whole cell extracts (5, 31), concentrations of OA of greater than 1 µM are necessary to inhibit PP1 by 50%. Therefore, it is likely, considering that our concentration range for OA was 3-9 nM, that relatively selective inhibition of PP2A rather than of PP1 accounts for the present findings.

Our results demonstrate that PP2A inhibition leads to an arrest predominantly in the replication of mesenchymal cells, with relatively little effect on the differentiated respiratory epithelium. The decrease in lung explant growth and airway branching together with the increased DNA synthesis demonstrated in the mesenchymal cells indicates that PP2A inhibition in the developing lung results in a release of the G₁/S checkpoint and a cell cycle arrest before the M phase, or mitosis, in the mesenchyme. Although the role of PP2A in the regulation of the cell cycle is implied from the combination of genetic and pharmacological manipulations, the exact mechanisms are poorly understood (reviewed in Ref. 20). Yeast genetic models of PP2A deficiency and PP2A-inhibited bone marrow-derived macrophages have shown PP2A to be required for progression of the cell cycle (13, 36). As in the present study, other investigators also have shown PP2A inhibition with OA to result in a cell cycle arrest in the G₂/M phase of the cell cycle (10). Similarly, in vitro mammalian cell culture studies using the pharmacological PP2A inhibitor fostriecin also have reported a G₂/M phase arrest (22). The exact mechanisms of PP2A regulatory control of the cell cycle are speculative but may involve phosphorylation of cell cycle proteins. Treatment of cells in vitro with OA results in increased phosphorylation of the cell cycle suppressor proteins retinoblastoma protein (Rb) and p53 (38). Phosphorylation of Rb leads to cell proliferation by controlling progression through the restriction point within G₁ phase of the cell cycle (24); p53, on the other hand, is a transcription factor activated by phosphorylation leading to a G₂/S cell cycle arrest (1, 26). Concurrent with the present study, PP2A appears to have a role in the regulation of the G₂/M phase of the cell cycle as cdc2 and the dual-specificity protein phosphatase, cdc25, are both substrates for PP2A (13). Taken together, PP2A has an important role in the regulation of the cell cycle by dephosphorylation of key regulatory proteins. Unregulated DNA synthesis without cell division (i.e., no increase in cell number), as the result of PP2A inhibition, could explain the increased BrdU labeling, small explant size, and decreased airway branching we observed in the OA-treated explants. Therefore, in the lung, the predominant effect of PP2A on the cell cycle appears to be maintaining the G₁/S checkpoint and the transition to mitosis.

Mesenchymal cells are necessary for the normal growth of respiratory epithelium (11). Recombination experiments have demonstrated that peripheral embryonic lung mesenchyme produces soluble factors that are inductive of early embryonic lung branching morphogenesis and epithelial cell fates (reviewed in Ref. 33). It is unclear from previous expression studies (37) why the mesenchyme would be more sensitive to the effects of PP2A inhibition than the epithelium. This may have to do with relatively decreased pools of PP2A such that the stoichiometry of inhibitor to enzyme was tilted in favor of the mesenchyme vs. the epithelial cells. It is unclear at this point whether there are possible differences in PP2A expression in the lung in vivo vs. in vitro as an explanation for the predominant mesenchymal effect. The in vitro model of lung development closely mimics in vivo lung growth and expression of lung epithelial markers (35). Therefore, it is unlikely that significant changes in expression of PP2A are likely but as of yet unexplored. The present study also suggests that another mechanism regulating lung development is the proliferation of the mesenchyme. Proliferation of the mesenchyme may be necessary to support tubular growth and branching by providing appropriate local levels of growth factors necessary for the process. Further studies will be necessary to determine if PP2A regulates expression levels of growth factors within mesenchymal cells.

Our findings of decreased or absent apoptosis in PP2A-inhibited explants together with the expected and readily detectable baseline apoptotic activity in the control explants suggest that PP2A is a positive regulator of apoptosis in the embryonic lung. This finding is supported by studies in transgenic knockout mice lacking the catalytic subunit of PP2A where null mice die early in gestation but maintain persistent embryonic masses much later into gestation than would be anticipated (8). The present study suggests that PP2A in the lung may be involved in the signal transduction pathways controlling apoptosis.

PP2A has an important role in the regulation of SP-C expression as demonstrated by the decrease in SP-C mRNA levels with OA treatment. PP2A mRNA and protein are developmentally regulated in the lung, being maximally expressed in the least mature epithelium (37). SP-C is expressed in early developing respiratory epithelium in a pattern similar to PP2A (34). Other investigators (32) have suggested that PP2A may be involved in the regulation of surfactant production in the developing lung, supporting the notion that PP2A could have a role in surfactant synthesis. PP2A is known to affect transcription by direct phospho-regulation of a number of transcription proteins including c-Jun of the AP-1 complex (2). It is not possible at this time to determine whether decreased SP-C mRNA levels with PP2A inhibition result from decreased SP-C gene transcription or an actual decrease in the number of forming type II pneumocytes. The decreased expression of SP-C along with the decreased airway branching in PP2A-inhibited explants observed in our study, suggests that PP2A is necessary for normal mesenchymal-epithelial differentiation or possibly SP-C transcription. Similarly, Drosophila and mouse genetic models of PP2A deficiency have demonstrated that PP2A is required for cell fate determination and differentiation of mesenchymal derivatives (8, 16).

In summary, the data presented in this study demonstrate for the first time that PP2A is a mediator of rat lung growth and development in embryonic whole organ culture. We believe that PP2A has a positive role in the regulation of fetal lung growth by specific regulation of mesenchymal cell proliferation, supporting branching morphogenesis. Furthermore, we speculate that PP2A may be a key transduction intermediate in regulating epithelial cell differentiation.