Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Pulmonary alveolar type II cells carry out highly specialized functions that include the synthesis, secretion, and reutilization of surfactant, a surface-active lipoprotein which acts to reduce surface tension at the alveolar air-liquid interface (1) and is essential for normal breathing. Type II cells are unique in their ability to produce surfactant which contains relatively large amounts of dipalmitoylphosphatidylcholine (DPPC), a saturated glycerophospholipid with singular surface-active properties, and store surfactant lipids and proteins in organelles termed lamellar bodies. Type II cells synthesize four lung-specific surfactant-associated proteins, SP-A, SP-B, SP-C, and SP-D, which serve a number of important functions (2). Surfactant synthesis is initiated in type II cells only after 75% of gestation is complete (3). Surfactant synthesis by the fetal lung is regulated by a number of hormones and factors, including glucocorticoids and agents that increase cyclic AMP (4, 5).

To define the molecular mechanisms involved in developmental, hormonal, and type II cell-specific regulation of surfactant synthesis in fetal lung, we have focused on the gene encoding SP-A, a major surfactant protein that is synthesized primarily by type II pneumonocytes and to a lesser extent in nonciliated bronchioalveolar epithelial (Clara) cells (6, 7). SP-A gene expression is developmentally regulated in fetal lung in concert with the induction of surfactant glycerophospholipid synthesis (8, 9) and with the appearance of differentiated type II cells. SP-A appears to serve a number of critical roles, including facilitation of tubular myelin formation (10), enhancing the rapid adsorption of DPPC to an air-liquid interface (11), mediating reutilization of surfactant phospholipids and proteins (12), and activating immune defense within the alveolus (13, 14).

SP-A gene expression is undetectable in fetal lung at mid-gestation. When human fetal lung explants are placed in organ culture in serum-free medium they differentiate spontaneously and develop the capacity for synthesis of surfactant glycerophospholipids and SP-A (15). Treatment of the lung explants with the cAMP analogue dibutyryl cAMP (Bt₂cAMP) increases the rate of appearance of type II pneumonocytes, as well as the rate of SP-A gene transcription (16, 17). The factors that cause spontaneous differentiation of the cultured lung explants have not been defined. However, we have found that the mid-gestation human fetal lung explants produce relatively high levels of the prostanoids, PGE₂, TXA₂, and PGI₂, and that inhibition of cyclooxygenase activity by indomethacin treatment causes a pronounced reduction in cyclic AMP formation and prevents the spontaneous induction of SP-A gene expression (18). In other studies, we observed that spontaneous differentiation and cyclic AMP induction of SP-A gene expression in the cultured lung tissue is dependent upon the oxygen tension of the environment; at environmental oxygen tensions of  5% the effect of Bt₂cAMP to induce SP-A gene expression was abolished (19).

The organ culture system provides an effective means for promoting and maintaining type II cell differentiation, which likely depends upon precise cellular interactions and tissue architecture. In this regard, organ culture affords a powerful system for study of the multifactorial regulation of synthesis of surfactant lipids and proteins. However, the absence of cell lines or of a primary culture system in which phenotypic properties of type II pneumonocytes are maintained, has hampered the progress of studies of the regulation of surfactant secretion and reutilization, and the implementation of DNA transfection studies to functionally map regulatory elements of type II cell-specific genes. In the present study, we have developed a primary type II cell-enriched culture system in which major morphologic and biochemical properties of type II cells are sustained for up to 2 wk. By use of recombinant replication of defective human adenoviruses, we have found that reporter gene constructs containing sequences flanking the 5'-end of the SP-A gene are expressed specifically in islands of type II cells, but not in the surrounding fibroblasts. Thus, this system provides an excellent model for studies of cis-acting elements involved in type II cell-specific expression of lung-specific genes.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Organ Culture and Type II Pneumonocyte Isolation and Primary Culture

The procedure used to isolate and maintain type II pneumonocytes in primary culture is outlined in Figure 1. Lung tissues from mid-trimester human abortuses were obtained in accordance with the Donors Anatomical Gift Act of the State of Texas; consent forms and protocols were approved by the Human Research Review Committee of the University of Texas Southwestern Medical Center at Dallas. The tissues were minced (1-2 mm³) and cultured on lens paper (#11-996; Fisher Scientific, Pittsburgh, PA) supported by stainless steel grids in serum-free Waymouth's MB752/1 medium (#51400; Gibco/BRL Inc., Gaithersburg, MD) in the presence of dibutyryl cAMP (1 mM, Bt₂cAMP, #104396; Boehringer Mannheim Corp., Indianapolis, IN) as described previously (16). After 5 days of organ culture with daily medium changes, lung explants were dissociated by digestion with collagenase type I (0.5 mg/ml, C-0130; Sigma Chemical Co., St. Louis, MO) and collagenase type IA (0.5 mg/ml, C-9891; Sigma Chemical Co.) for 15 min at 37°C with vigorous pipeting. Following collagenase digestion, the cells were treated with DEAE-dextran (250 µg/ml; Sigma Chemical Co.) and incubated for 45 min with shaking at 37°C. We have found that DEAE-dextran treatment selectively eliminates fibroblasts. The cells were pelleted at 400 × g and plated either onto Transwell^TM dishes (#3425; Costar Corp., Cambridge, MA) (1-2 × 10⁶ cells per dish), on 60-mm tissue culture dishes or on Thermanox coverslips (#5413; Nunc, Inc., Naperville, IL) that were coated with extracellular matrix (ECM) prepared by derived from MDCK cells (ATCC CCL34) (2-5 × 10⁶ cells per 60-mm dish). The extracellular matrix-coated dishes were prepared from confluent monolayers of MDCK cells that were treated with deoxycholate (1%) for 5 min. The dishes were washed 3 times with Hank's balanced salt solution and stored at 37°C until used. Plated epithelial cell-enriched cultures were incubated overnight in Waymouth's medium with 10% fetal bovine serum (#16000; Gibco/BRL Inc.). Dishes were washed twice with medium to eliminate dead and non-adherent cells, and then incubated in Waymouth's MB752/1 medium without fetal bovine serum. In studies where type II pneumonocytes were plated onto Transwell^TM dishes, culture medium was placed either below or above and below the cell monolayer. The plating density of the cells after overnight incubation was approximately 50-60%.

View larger version (26K): [in this window] [in a new window]   Figure 1.   Schematic of the technique used for isolation and culture type II pneumonocytes. Details of the procedure are described in MATERIALS AND METHODS. Fetal lung was minced and incubated in organ culture in serum-free medium with Bt2cAMP for 5 days. The tissue was then treated with collagenase (0.5 mg/ml) followed by incubation with DEAE-dextran (250 µg/ml) to selectively eliminate fibroblasts. The cells were then plated on ECM-coated 60-mm tissue culture dishes and incubated overnight in medium containing 10% serum. The dead and/or unattached cells were then washed away, resulting in a relatively pure population of type II pneumonocytes in primary culture (~ 2-5 million cells per dish). Based on previous morphometric studies (16, 38), it is estimated that approximately 50% of the cells in the cultured human fetal lung explants are type II cells. After DEAE-dextran treatment, > 90% of the plated cells contain osmophilic inclusions (Figure 2) which manifested the morphologic properties of lamellar bodies upon electron microscopic examination (Figure 3). In transfection studies, the cells were incubated with recombinant adenovirus for 1 h. The cells were then maintained in 1.0 ml serum-free medium for the remainder of culture period. Media were changed daily.

View larger version (124K): [in this window] [in a new window]   Figure 2.   Morphology of type II pneumonocytes in primary culture. Light micrograph of type II pneumonocytes after 5 days in primary culture that were stained with osmium tetroxide for presence of lamellar bodies. Shown is a typical aggregate of isolated type II cells containing numerous osmiophilic granules, which are seen as dark circular organelles clustered around the nuclei of the cells. Scale bar = 5 µm.

View larger version (131K): [in this window] [in a new window]   View larger version (146K): [in this window] [in a new window]   Figure 3.   Ultrastructure of type II pneumonocytes in primary culture. (A) Electron micrograph of isolated type II pneumonocytes after 4 days in primary culture in serum-free Waymouth's medium containing Bt2cAMP. Scale bar = 1 µm. (B) A higher magnification micrograph of the lamellar bodies in the type II pneumonocytes. Scale bar = 0.2 µm.

Osmium Tetroxide Staining of Lamellar Bodies in Type II Pneumonocytes in Primary Culture

To assess the presence of lamellar bodies in the cultured type II pneumonocytes, a modification of the method of Mason and associates (20) was used. The cells were plated onto ECM-coated Thermanox coverslips and cultured in serum-free medium containing Bt₂cAMP. Prior to analysis, the cells were rinsed in phosphate-buffered saline (PBS, pH 7.3) and fixed with glutaraldehyde (1.5% in PBS) for 15 min. The cells were then washed twice and treated with osmium tetroxide (1.0%; Sigma Chemical Co.) in PBS for 90 min at room temperature. The cells were subsequently washed twice and incubated in tannic acid (1.0% in PBS, pH 6.8) overnight. After washing with PBS, the cells were examined and photographed under light microscopy.

Electron Microscopy of Type II Pneumonocytes in Primary Culture

For analysis of the ultrastructure of the cultured type II pneumonocytes, transmission electron microscopy was performed. The cultured cells were scraped from the dishes, fixed in glutaraldehyde (2%), and routinely processed for electron microscopy (21). In brief, the cells were dehydrated in graded alcohols, embedded in resin, sectioned, and stained with lead citrate and uranyl acetate. Sections were viewed and photographed with a JEOL 100SX (Japan Electronic Optic Laboratories USA, Inc., Boston, MA) scanning electron microscope.

Immunoblot Analysis of SP-A Protein

Cellular proteins were isolated as described previously (22). Briefly, cells were scraped off the plates with a rubber policeman and homogenized in ice-cold homogenization buffer (0.25 M sucrose, 10 mM Tris-HCl [pH 7.4], 1 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride [PMSF]) using 10 strokes in a Teflon-glass homogenizer. The homogenized samples were pelleted at 600 × g for 5 min to remove debris, and the resulting homogenate was assayed for protein content. Cellular proteins (50 µg) were subjected to one-dimensional sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE) and the separated proteins transferred to nitrocellulose by electrophoretic transfer. The samples on the resulting blot were analyzed for SP-A content as described previously (23); the blot was incubated with rabbit anti-human SP-A immunoglobulin G (IgG) and ¹²⁵I-labeled goat anti-rabbit IgG. The blots were subjected to autoradiography using Kodak X-OMAT radiographic film (#165 1579; Eastman Kodak Co., Rochester, NY) for visualization of the resulting protein-antibody complexes. Relative amounts of protein were assessed by scanning densitometry of the autoradiograms.

Northern Analysis of Type II Pneumonocyte mRNA

Total RNA was extracted from the cells by homogenization in guanidinium isothiocyanate (4.0 M) using a Teflon-glass homogenizer. The cell extracts were centrifuged through a cesium chloride gradient (5.7 M), and the pelleted RNA was resuspended in water (24). Total RNA (15 µg) was electrophoresed, transferred to nitrocellulose, and probed using either a ³²P-labeled rabbit SP-A cDNA, rabbit SP-B cDNA, or rabbit SP-C cDNA, as described in detail previously (25). The relative levels of mRNA were assessed by autoradiography using Kodak X-OMAT radiographic film.

Transfection of Human Type II Pneumonocytes with Recombinant Adenovirus

A recombinant replication-defective human adenovirus containing 991 base pairs of sequence flanking the 5'-end of the rabbit SP-A linked to the human growth hormone (hGH) structural gene, as reporter (Ad:SP-A₉₉₁:hGH), was constructed as described previously (28). A recombinant replication-defective adenovirus containing the gene encoding bacterial -galactosidase (lacZ) driven by the cytomegalovirus early (E1) promoter (AdCMV:lac) has been described elsewhere (29). Type II pneumonocytes were infected with adenovirus for transfection of reporter genes as described previously (28). Briefly, type II pneumonocytes were plated onto 60-mm ECM-coated dishes and maintained overnight in Waymouth's MB752/1 medium containing fetal calf serum (10%, v/v). The cells were then washed several times with medium to remove non-adherent cells and debris, and incubated for 1 h with recombinant adenovirus. The medium was then replaced with fresh medium.

Immunocytochemical Analysis of hGH and -galactosidase in Transfected Human Type II Pneumonocytes

Cells to be infected with recombinant adenoviruses were plated onto ECM-coated Thermanox glass coverslips, as described above. After DNA transfection, cells were incubated as described above until day 5 of culture, when the cells were fixed with formaldehyde (3.7%) in PBS for 10 min. To visualize nuclear -galactosidase, the cells were incubated in PBS containing 5 mM potassium ferricyanate, 5 mM potassium ferrocyanate, 1 mM MgSO₄, and 0.5% 5-bromo-4-chloro-3-indolyl--D-galactopyranoside (X-gal) at 37°C. When a blue color indicating the presence of -galactosidase was observed, the cells were washed twice with PBS. To visualize the presence of hGH, the cells were permeabilized with NET/gel (50 mM Tris-HCl, 1 mM EDTA, 150 mM NaCl, 0.25% gelatin, 0.05% NP-40, 0.01% sodium azide, pH 8.0) and Triton X-100 (0.1%) for 30 min prior to antibody addition. The coverslips then were treated with rabbit IgG specific for hGH (50 µg/ml; Dako Corp., Carpinteria, CA) in NET/gel (without Triton X-100) for 60 min. The blot was then treated with the components of the Vectastain Elite ABC kit (#PK-6101; Vector Labs, Burlingame, CA) to promote the binding of horseradish peroxidase to the hGH/antibody complexes. Visualization of the complexes as red granules was accomplished by use of the aminoethyl carbazole substrate kit from Zymed Laboratories (#00-2007; Zymed Labs. Inc., San Francisco, CA).

Quantitative Analysis of hGH Production in Transfected Human Type II Pneumonocytes

Media from cells infected with the AdSP-A₉₉₁:hGH recombinant adenovirus were collected at 24-h intervals. The concentration of hGH in the medium was quantitated by radioimmunoassay using an Allegro hGH kit (Nichols Institute Diagnostics, San Juan Capistrano, CA).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Type II Pneumonocytes Isolated from Human Fetal Lung Explants Maintain the Characteristics of Type II Cells in Primary Monolayer Culture

In developing a method for isolation and culture of differentiated type II cells, a number of factors were considered. First, we previously observed that fetal lung explants differentiate when maintained in organ culture in serum-free medium; there is spontaneous appearance of type II cells and induced synthesis of surfactant phospholipids (15) and SP-A (16). Furthermore, we found that addition of Bt₂cAMP to the culture medium increased the rate of type II cell differentiation and SP-A gene transcription (16, 30, 31). In fact, after several days of organ culture, ~ 50% of the cell population in the lung explants is comprised of type II pneumonocytes (16). Thus, there is a marked enrichment of type II cells in the cultured fetal lung explants as compared to late fetal, neonatal, or adult lung in situ, in which it is estimated that type II cells comprise only 10-15% of the cell population. Second, in our prior attempts to transfect DNA into lung cell suspensions, we serendipitously observed that DEAE-dextran (250 µg/ml) eliminated the majority of fibroblasts, resulting in a relatively pure population of alveolar epithelial cells (32). Finally, in light of the potentially important role of extracellular matrix (ECM) in the maintenance of alveolar epithelial cell differentiation (33, 34), the isolated type II pneumonocytes were cultured on ECM derived from MDCK cells. The method for type II cell isolation and culture is diagrammed in Figure 1.

We observed that the cultured cells contained numerous lamellar bodies, the major morphologic characteristic of type II pneumonocytes (35), for at least 5 days after the initiation of primary culture. Shown in Figure 2 is a light micrograph of a typical cluster of type II pneumonocytes after 5 days of culture in serum-free medium containing Bt₂cAMP. The cells, which were stained for the presence of lamellar bodies using osmium tetroxide (20), contained numerous large osmiophilic granules clustered around the nuclei. By electron microscopy, it is evident that these granules have the morphologic characteristics of lamellar bodies (Figure 3B) (36) and many of the cells contain these granules (Figure 3A). In addition, in all of the micrographs taken of the cells, we always observed a cuboidal morphology regardless of the plane of the section, suggesting that the cells isolated have the phenotypic properties of type II cells.

Type II Pneumonocytes in Primary Culture Maintain Expression of SP-A for Up to 14 Days

Type II pneumonocytes, when isolated from adult lung and plated on plastic culture dishes, have been reported to rapidly lose morphologic properties of type II cells and the ability to synthesize surfactant phospholipids and SP-A (33, 37). To ascertain the differentiated state of the fetal type II cells in the culture system developed in the present study, the levels of SP-A were analyzed by immunoblot analysis of cell lysates as a function of time in culture. The cells were cultured in 60-mm dishes in 1 ml of serum-free Waymouth's medium containing Bt₂cAMP with daily medium change for up to 14 days. As can be seen in the autoradiogram in Figure 4, the amount of immunoreactive SP-A appeared relatively low in the fetal lung explants that had been incubated for 5 days in the presence of Bt₂cAMP prior to cell isolation (ST). The amount of immunoreactive SP-A increased in the isolated cells through day 8 of culture and remains relatively constant throughout day 14. It should be noted that the levels of immunoreactive SP-A appeared low in the cultured lung explants (ST) relative to the cultured cells, because the extremely high levels of SP-A in the cultured type II cells necessitated a short exposure time of the autoradiogram. In fact, the levels of immunoreactive SP-A in the cultured explants were comparable to those observed in previous studies (16, 17). The high levels of SP-A expression in the type II cell cultures as compared with the fetal lung explants is due, in part, to the marked enrichment of type II cells in the former, as well as to undefined features of the cell culture system that enhance type II cell phenotypic properties which have not as yet been defined.

View larger version (57K): [in this window] [in a new window]   Figure 4.   Immunoblot analysis of SP-A in human type II pneumonocytes as a function of time in primary culture. Human type II pneumonocytes were maintained for up to 14 days in primary culture in serum-free Waymouth's medium containing Bt2cAMP (1 mM). Cells were harvested after various times of incubation; cell homogenates were analyzed for SP-A by immunoblotting using a rabbit IgG specific for human SP-A, followed by incubation with 125I-labeled goat anti-rabbit IgG. Shown is an autoradiogram of the resulting immunoblot.

Expression of Genes Encoding SP-A, SP-B, and SP-C Are Regulated by Cyclic AMP and Glucocorticoids in Type II Pneumonocytes in Primary Culture

To analyze mRNA levels for SP-A, as well as the other surfactant proteins SP-B and SP-C in the cultured type II cells, as well as the effects of cyclic AMP and glucocorticoids, we analyzed the steady-state levels of mRNAs encoding the surfactant proteins in cells after 5 days of primary culture in the absence or presence of Bt₂cAMP (1 mM) and dexamethasone (10¹⁰ to 10⁷ M), added alone or in combination. After culture, total RNA isolated from the cells was analyzed for SP-A, SP-B, and SP-C mRNA transcripts by Northern blotting using radiolabeled cDNAs for rabbit SP-A (25), SP-B, and SP-C (the SP-B and SP-C cDNAs were a gift from Dr. V. Boggaram, University of Texas Health Science Center at Tyler, TX). As can be seen in Figure 5A, SP-A mRNA levels in type II cells incubated with Bt₂cAMP were markedly increased over those of cells incubated in control medium. As can be seen in the autoradiogram in Figure 5B, dexamethasone had a dose-dependent biphasic effect on SP-A mRNA levels in type II cells co-incubated with Bt₂cAMP; dexamethasone at 10¹⁰ M increased SP-A mRNA levels as compared with those observed in cells incubated with Bt₂cAMP alone, whereas, at concentrations  10⁹ M, dexamethasone antagonized the stimulatory effect of Bt₂cAMP. These effects of Bt₂cAMP and dexamethasone on SP-A mRNA are similar to those observed using human fetal lung tissue in organ culture (38).

View larger version (43K): [in this window] [in a new window]   Figure 5.   Effects of Bt2cAMP and dexamethasone on the levels of SP-A, SP-B and SP-C mRNA in type II pneumonocytes in primary culture. Type II pneumonocytes in primary culture were incubated in the absence or presence of Bt2cAMP and/or dexamethasone for 5 days. The RNA was subjected to Northern analysis using radiolabeled cDNA probes specific for rabbit SP-A, SP-B, or SP-C. (A) Northern blot of SP-A mRNA levels in cells incubated in the absence (Con) or presence of Bt2cAMP (Bt; 1 mM). (B) Northern blot of SP-A mRNA levels in cells incubated in the presence of Bt2cAMP (1 mM) (Bt), in the absence or presence of various concentrations of dexamethasone (Dex, 1010 to 107 M). The position of SP-A mRNA in A and B is indicated. (C) Northern blot of SP-B and SP-C mRNA isolated from type II cells incubated in the absence (Con) or presence of Bt2cAMP (1 mM), dexamethasone (107 M), or the two agents in combination (Bt+Dex). The positions of SP-B and SP-C mRNA in C are indicated.

As can be seen in Figure 5C, SP-B mRNA levels in the cultured type II cells were increased by Bt₂cAMP and by dexamethasone as compared with those of untreated cells. In type II cells incubated with Bt₂cAMP and dexamethasone in combination, there was an apparent synergistic induction of SP-B mRNA levels. On the other hand, SP-C mRNA levels were not detectably altered by Bt₂cAMP treatment. Dexamethasone had a modest effect to increase the levels of SP-C mRNA; in type II cells incubated with Bt₂cAMP and dexamethasone in combination, SP-C mRNA was increased to levels that were greater than those observed with dexamethasone alone.

Primary Culture of Type II Pneumonocytes in Transwell^TM Dishes: Effect of Medium Placement on SP-A Expression

Transwell^TM dishes were originally designed for analysis of regulated secretion by polarized epithelial cells. The dishes contain a collagen-coated membrane platform with upper and lower chambers for placement of medium. In this study, we analyzed the effects of medium placement in the lower chamber only with medium placement in both upper and lower chambers on the levels of immunoreactive SP-A in human fetal type II cells in primary culture. Type II pneumonocytes were isolated from cultured human fetal lung explants, as described above, and plated in serum-free medium containing Bt₂cAMP (1 mM) either on ECM-coated 60-mm culture dishes (volume of medium = 1 ml), or on Transwell^TM dishes with serum-free medium in both the upper and lower chambers, or in the lower chamber alone (air/liquid interface). After 4 or 8 days of primary culture with daily medium changes, the cells were harvested and the proteins were subjected to immunoblotting for analysis of SP-A. As described above, when the enriched type II cell preparation was cultured in dishes coated with ECM, the levels of SP-A protein were elevated and increased between days 4 and 8 of culture (ECM lanes, Figure 6). When the cells were cultured in the Transwell^TM dishes at an air/liquid interface (membrane/air lanes, Figure 6), the levels of immunoreactive SP-A were increased to levels greater than those of the cells cultured on the ECM-coated culture dishes. By contrast, when the type II cells were cultured in the Transwell^TM dishes with medium in both upper and lower chambers (membrane/media lanes, Figure 6), the levels of immunoreactive SP-A were reduced markedly as compared with cells cultured in Transwell^TM dishes at an air/ liquid interface or in the ECM-coated culture dishes. In fact, the levels of SP-A protein in the cells cultured in the Transwell^TM dishes with medium in upper and lower chambers declined between days 4 and 8 of culture, suggesting a loss of the type II cell phenotypic properties. These findings indicate that culture of type II pneumonocytes at an air/liquid interface contributes to the maintenance of a differentiated phenotype.

View larger version (42K): [in this window] [in a new window]   Figure 6.   Immunoblot of SP-A in human type II pneumonocytes in primary culture plated either on ECM-coated 60-mm tissue culture dishes or in TranswellTM dishes. Human type II pneumonocytes in primary culture were incubated for 4 or 8 days in serum-free Waymouth's medium containing Bt2cAMP (1 mM) after plating on ECM-coated 60-mm tissue culture dishes (ECM) or in TranswellTM dishes containing medium either in the lower chamber only (Membrane/Air) or in both the lower and upper chambers (Membrane/Media). Shown is an immunoblot of cell lysates probed for SP-A protein. The position of the SP-A band is indicated.

Expression of SP-A in Isolated Type II Pneumonocytes in Primary Culture Depends on the Volume of Medium

In consideration of the findings using the Transwell^TM dishes, it was of interest to determine the effects of medium volume on SP-A expression in type II cells cultured in ECM-coated dishes. Type II pneumonocytes were incubated in 60-mm ECM-coated culture dishes containing 0.25- 3.0 ml of serum-free medium containing Bt₂cAMP for 5 days. The cells were harvested, and the levels of immunoreactive SP-A protein were analyzed as a marker of type II pneumonocyte differentiation. As shown in the autoradiogram in Figure 7, the levels of SP-A were equivalent in cells cultured in medium volumes  1.0 ml. However, at volumes of medium > 1.0 ml the levels of immunoreactive SP-A were decreased. These findings further suggest that culture of type II cells at an air/liquid interface promotes maintenance of a differentiated phenotype.

View larger version (33K): [in this window] [in a new window]   Figure 7.   Immunoblot analysis of SP-A in human type II pneumonocytes in primary culture as a function of medium volume. Human fetal type II pneumonocytes plated on ECM coated 60-mm dishes were incubated in 0.25, 0.5, 1.0, 2.0, or 3.0 ml of serum-free Waymouth's medium containing Bt2cAMP (1 mM) for 5 days. Cell homogenates were analyzed for SP-A by immunoblotting. Shown is the resulting immunoblot. The position of the SP-A band is indicated.

Type II Pneumonocytes in Primary Culture Are Readily Infected by Recombinant Human Adenoviruses and Provide a System for Analysis of Gene Regulatory Elements Required for Cell-Specific Expression

We have used enriched populations of type II cells cultured on ECM-coated dishes to define genomic elements upstream of the rabbit (28, 39, 40) and human (41) SP-A genes that mediate cyclic AMP-regulated expression. In our initial studies, we found that traditional methods of DNA transfection, including use of calcium phosphate (42), DEAE-dextran (43), liposomes (44), and electroporation (45), were ineffective as means for transfer of DNA into the primary cultures of type II cells (unpublished observations). In fact, as mentioned above, we serendipitously observed that although DEAE-dextran was ineffective as a means for transfecting DNA into the isolated lung cells, it selectively eliminated fibroblasts, providing a means for enriching the cultures for type II cells (unpublished observations). We, therefore, turned to the use of recombinant replication-defective human adenoviruses (46), which are trophic for lung airway epithelial cells (47), for introduction of fusion genes comprised of various amounts of SP-A 5'-flanking DNA linked to the human growth hormone (hGH) structural gene into the type II cells in primary culture.

It should be noted that the lung cell cultures are comprised of islands of type II cells surrounded by fibroblasts. To validate the use of this system for studies to define regions of the SP-A gene that mediate type II cell-specific expression, it is important to determine whether both type II cells and fibroblasts are infected by the recombinant adenoviruses and whether one or both cell types have the capacity to express the SP-A:hGH fusion genes. To address these issues, in the present study, we infected the enriched type II cell cultures with two recombinant adenoviruses, AdSP-A₉₉₁:hGH (28) and AdCMV:lac (29). Human fetal lung cells were isolated by collagenase treatment of lung explants that had been incubated for 5 days in medium containing Bt₂cAMP. Half of the cell suspension was treated with DEAE-dextran to reduce the number of fibroblasts; the remainder was not treated with DEAE-dextran. Both cell suspensions were plated onto ECM-coated Thermonox glass coverslips and cultured for 2 days in Waymouth's medium containing Bt₂cAMP. On the third day of culture, the cells were simultaneously infected for 1 h with 10⁷ plaque-forming units (PFU) of AdSP-A₉₉₁:hGH and AdCMV:lac. A multiplicity of infection (m.o.i.) of 3 for each recombinant adenovirus was used so that essentially all of the cells would be transfected with each gene construct. Three days later, the cells were fixed on the coverslips and stained for the presence of -galactosidase and hGH.

Light micrographs of the adenovirus infected cells stained for -galactosidase and hGH are shown in Figures 8A and 8B. The type II cell cultures shown in Figure 8A were prepared using DEAE-dextran treatment to selectively remove fibroblasts, as described in MATERIALS AND METHODS. The cultures shown in Figure 8B were prepared without DEAE-dextran treatment to more critically evaluate cell-specific expression of AdSP-A₉₉₁:hGH. Since the CMV:lacZ expression vector contains a nuclear localization signal, expression of AdCMV:lac is reflected by blue staining for -galactosidase in cell nuclei. Expression of AdSP-A₉₉₁:hGH was assessed by immunocytochemistry for hGH and is indicated by the presence of red granules in the cytoplasm. As can be seen by the blue staining of the nuclei, it is evident that nearly all of the epithelial cells and fibroblasts were infected and express -galactosidase from the AdCMV:lac adenovirus (Figures 8A and 8B), demonstrating the efficiency of transfection cells by adenovirus. By contrast, only cells with an epithelial cell morphology were immunoreactive for hGH. This finding suggests that, whereas the lung epithelial cells contain transcription factors that interact with elements within the 991 bp SP-A gene 5'-flanking region, the fibroblasts do not. The absence of staining for -galactosidase or hGH in some of the cells may be due to the fact that adenoviruses are replication defective in cells that do not express E1a (48). Therefore, when a cell divides, it is possible that one of the daughter cells may not contain adenoviral DNA. These findings indicate that the primary type II cell cultures can effectively be used to functionally map cis-acting elements that mediate type II cell-specific expression of SP-A promoter activity (28, 39).

View larger version (109K): [in this window] [in a new window]   View larger version (113K): [in this window] [in a new window]   Figure 8.   Analysis of hGH and -galactosidase expression in human type II pneumonocytes transfected with AdSP-A991:hGH and AdCMV:lac adenoviruses. Midgestational human fetal lung explants that had been cultured in the presence of Bt2cAMP were incubated with collagenase as described in MATERIALS AND METHODS. Half of the cell suspension was treated with DEAE-dextran to selectively remove fibroblasts and then plated on ECM-coated coverslips (A). The other half of the suspension was plated onto the ECM-coated coverslips without prior DEAE-dextran treatment (B). On the third day of culture, each dish of coverslips was incubated with 107 PFU of AdSP-A991:hGH viruses and 107 PFU of AdCMV:lac viruses for 1 h, washed, and subsequently incubated in serum-free medium containing Bt2cAMP. Three days later, the cells were fixed and stained for the presence of -galactosidase protein using 5-bromo-4-chloro-3-indolyl--D-galactopyranoside (X-gal). Since the lacZ gene contains a nuclear localization signal, the resulting blue reaction product should be localized to the nucleus. The cells were then incubated with anti-hGH IgG, and the complexes were visualized using AEC chromagen (Vector Labs), resulting in a red color.

To characterize the expression of recombinant adenoviruses for use in promoter studies, we analyzed hGH production in type II cells infected with AdSP-A₉₉₁:hGH adenovirus as a function of increasing m.o.i. from 0.1 to 10, and as a function of increasing numbers of cells at a constant number of PFU, where the number of infectious viral particles becomes limiting (Figure 9). Human type II pneumonocytes were cultured and infected with recombinant adenovirus as described in MATERIALS AND METHODS with the following modifications. In the study shown in Figure 9A, the number of type II cells was maintained at a constant level (4 × 10⁶ cells per dish), while the number of infectious viral particles added to the cells was varied from 4 × 10⁵ to 4 × 10⁷ PFU per dish, resulting in increasing gene dosage per cell. In the parallel study shown in Figure 9B, the number of type II cells was varied from 10⁵ to 10⁷ cells per dish, while the number of PFU was constant at 10⁶, resulting in a decrease in the m.o.i. After 5 days of culture in serum-free medium without Bt₂cAMP, the media were collected and the amount of hGH produced was analyzed. As shown in Figure 9A, as the number of recombinant adenoviral particles was increased from an m.o.i. of 0.1 to 10, hGH production was increased in a linear fashion. These findings indicate hGH expression increases in a linear manner with increased AdSP-A₉₉₁:hGH infection up to an m.o.i. as high as 10, and suggest that the concentration of transcription factors within the type II cells that interact with the SP-A promoter do not become limiting under these conditions. As shown in Figure 9B, when the number of infectious viral particles became limiting relative to the number of type II cells, so that the m.o.i. was  0.3, hGH expression reached a constant level. In previous studies (28, 39), we found that hGH expression by transfected type II cells was highly reproducible from one experiment to another for each fusion gene construct under the same treatment and assay conditions when the m.o.i. was within the range of 0.15-0.20. Under these conditions, expression of the AdSP-A₉₉₁:hGH was readily detectable under basal conditions, and alterations in expression upon addition of various factors (i.e., Bt₂cAMP, dexamethasone) could readily be assessed (28, 39).

View larger version (23K): [in this window] [in a new window]   Figure 9.   Expression of AdSP-A991:hGH in type II pneumonocytes in primary culture under conditions of constant cell number with increasing virus (A), or constant virus number with increasing number of cells (B). (A) Human type II pneumonocytes were isolated and plated at a density of 4 × 106 cells per dish in medium containing Bt2cAMP (1 mM). The cells were transfected with recombinant adenovirus virus (AdSP-A991:hGH) in amounts ranging from 4 × 105 to 4 × 107 PFU per dish, resulting in a range of multiplicity of infection (m.o.i.) from 0.1-10. The transfected cells were incubated for 5 days. As can be seen, with increasing m.o.i. up to a value of 10, expression was linear. (B) Cells plated at densities ranging from 105 to 107 cells per dish were incubated with a constant amount of virus (106 PFU) and incubated for 5 days as above. Expression of the SP-A991:hGH fusion gene increased in an almost linear manner in a range of m.o.i. from 1-10; however, at an m.o.i. < 1, the number of recombinant adenoviral particles became limiting and hGH expression remained constant.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

The lack of alveolar epithelial cell lines that manifest both morphologic properties and gene expression patterns of type II pneumonocytes has hampered studies to define the molecular mechanisms involved in type II cell-specific gene regulation, and in surfactant synthesis, processing, secretion, and reutilization. Although murine lung epithelial (MLE) cell lines have been established from distal bronchiolar/alveolar lung tumors of transgenic mice carrying the simian virus 40 (SV40) large tumor antigen under control of the human SP-C promoter (49), these cell lines express some, but not all of the cellular markers characteristic of type II cells. For example, the MLE-15 cell line was reported to be comprised of subsets of cells that contained multilamellar inclusion bodies and relatively high levels of SP-B mRNA, whereas SP-A mRNA was found to be present at low levels and SP-C mRNA was undetectable. On the other hand, the MLE-12 cell line expressed SP-C mRNA; however, SP-A mRNA was undetectable and lamellar bodies were absent.

Thus, there has been considerable effort by a number of laboratories to develop methods for primary culture of type II pneumonocytes which maintain a differentiated phenotype for an extended period of time in monolayer culture. In the present study, we have developed a relatively simple method for isolation and culture of type II pneumonocytes isolated from midgestation human fetal lung explants. Since SP-A gene expression is initiated in fetal lung tissue in closer association with the induction of surfactant glycerophospholipid synthesis and the appearance of differentiated type II cells than the surfactant proteins SP-B, -C, and -D, we have focused on SP-A as a marker of the differentiated state.

In previous studies, we observed that midgestation human fetal lung explants differentiate spontaneously when maintained in organ culture in serum-free defined medium (15). Essentially all of the epithelial cells lining the prealveolar ducts differentiate into type II cells containing numerous lamellar bodies. It was found that ~ 50% of the cells in the cultured fetal lung explants had the morphologic characteristics of type II cells (16). Thus, the proportion of type II cells in the cultured fetal lung tissue is markedly increased as compared with adult lung tissue, in which it is estimated that ~ 10-20% of the cells are type II cells (50). Furthermore, we previously observed that treatment of human fetal lung explants with Bt₂cAMP increases the size of the prealveolar ducts, enhances the rate of type II cell differentiation, and increases transcriptional activity of the SP-A gene (16). In consideration of the increased proportion of type II cells and elevated levels of SP-A gene transcription in cyclic AMP treated human fetal lung explants we chose to use these tissues as the source of type II cells for the primary culture system. The same technique also has been successfully applied to the isolation and culture of type II cells from fetal rat (28) and rabbit lung tissues (Alcorn, Smith, and Mendelson; unpublished observations).

The mechanisms whereby cyclic AMP enhances the rate of type II cell differentiation have not been defined. In previous studies to determine the mechanisms whereby cyclic AMP increases transcriptional activity of the SP-A gene, we identified conserved cyclic AMP-response element (CRE)-like sequences upstream of the rabbit (28, 40) and human (41) SP-A2 genes. In studies to characterize the transcription factors that bind to this element, it was found that neither cyclic AMP-response element-binding protein (CREB), cyclic AMP-response element modulator (CREM), or activating transcription factor-1 (ATF-1) bind to this site (40). Rather, it appears that this sequence may serve as a binding site for a member of the nuclear receptor superfamily (40). This is of interest, since several members of this gene family are known to serve important roles in cellular differentiation (51, 52).

In previous attempts to transfect DNA into freshly isolated human fetal type II cells, we made the serendipitous observation that treatment of collagenase-dispersed fetal lung explants with DEAE-dextran (which was ineffective in promoting DNA transfection) caused a selective depletion of fibroblasts from the cell suspension, without substantially altering the number of type II cells. After plating of this enriched type II cell suspension on extracellular matrix, the type II cells aggregate to form islands that are surrounded by small numbers of fibroblasts. Since the cells are placed in serum-free medium after the first day of culture, fibroblasts are maintained at relatively low numbers. In consideration of the possible importance of type II cell-fibroblast interactions in the maintenance of type II cell differentiation (53), we elected to carry out our studies using this type II cell-enriched culture system rather than attempting to create pure type II cells cultures.

The findings obtained in this study suggest that culture of these cells at an air-liquid interface is of great importance in the maintenance of type II cell differentiation, an effect previously noted by Yamaya and associates (54) and Whitcutt and colleagues (55) in their studies of trachea epithelial cell differentiation. This finding was particularly evident in the studies using Transwell^TM dishes, in which medium was placed either in the lower chamber alone, or in upper and lower chambers of the dish. When type II cells were cultured in Transwell^TM dishes that contained medium only in the lower chamber, so that the cells were maintained at an air-liquid interface, the levels of SP-A expression were markedly increased as compared with cells cultured with medium in the upper as well as lower chambers. Furthermore, when type II cells were incubated on ECM in 60-mm tissue culture dishes, the levels of SP-A expression were increased when the volume of culture medium was  1.0 ml as compared with cells cultured in larger amounts of medium. Therefore, under conditions in which the cells are immersed in medium, the levels of SP-A expression are decreased. These findings are of interest, since the type II cell exists at an air-liquid interface within the alveolus. As a result, the type II cell, together with the type I cell, are exposed to unique surface forces, and a higher oxygen tension than is any other cell type in the body.

The findings of our previous studies suggest that oxygen plays a permissive role in the spontaneous differentiation of midgestation human fetal lung explants in culture at both morphologic and biochemical levels (19). When fetal lung explants were maintained in organ culture at an oxygen tension of 1%, SP-A mRNA and protein failed to increase spontaneously with time in culture and there was no inductive effect of Bt₂cAMP. Furthermore, when cultured in a 1% oxygen-containing environment, the volume densities of the lumina of the prealveolar ducts and of the epithelium lining those ducts were markedly reduced as compared with explants maintained in a 20% oxygen-containing environment; again, cyclic AMP had no inductive effect on morphologic development. When fetal lung tissues that had been maintained in 1% oxygen were placed in a 20% oxygen environment, there was rapid enlargement of the prealveolar ducts and an induction of SP-A gene expression (19). In oxygen "dose-response" studies, we also observed that the inductive effects of cyclic AMP on SP-A gene expression and on morphologic development were prevented when the oxygen tension was reduced to levels  5%. Whereas, the spontaneous increase in volume density of the alveolar lumina and of epithelial cells were maximal at an oxygen tension of  10%, the stimulatory effect of cyclic AMP on SP-A mRNA levels was far greater at 20% oxygen than at lower oxygen tensions. In those studies, we also observed that SP-C expression was markedly reduced in the human fetal lung explants at low oxygen tension (19).

In light of these findings, we suggest that increased oxygen availability to type II cells cultured at an air-liquid interface may contribute to promotion and maintenance of type II cell differentiation and the induction of SP-A gene expression. Alternatively, cell polarization is likely increased by culture of the type II cells at an air-liquid interface, and this may contribute to maintenance of the differentiated state. It is evident from the DNA transfection studies that the cultured type II cells have retained the transcription factors required to mediate expression of SP-A promoter activity. In those experiments, we found that although both lung fibroblasts and type II cells were equivalently infected by recombinant adenoviruses and have the ability to express -galactosidase under control of the CMV promoter, only the type II cells had the capacity to express SP-A:hGH fusion genes containing 991 bp of 5'-flanking sequence from the rabbit SP-A gene. Based on these findings, we conclude that the primary type II cell culture system reported herein will provide a suitable means for mapping of genomic elements required for expression of type II cell-specific genes. It will be of interest to determine whether these cells retain the ability to process, secrete, and metabolize surfactant components in a manner similar to that of type II cells in situ.