Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The synthesis and secretion of pulmonary surfactant, which is a complex of phospholipids and the lung-specific surfactant proteins (SPs) SP-A, SP-B, and SP-C, is a primary function of alveolar type II cells. The elucidation of the mechanisms regulating surfactant production is of great interest because insufficiency of pulmonary surfactant leads to respiratory distress, and abnormalities in its composition are associated with a number of lung diseases. Because the lung is composed of many different cell types, studies on the regulation of type II cell differentiated function have often been carried out on cultures of purified type II cells. A consistent observation in these studies has been that type II cells maintained on a tissue culture plastic substratum rapidly lose markers of differentiated function. A number of laboratories have attempted to sustain adult type II cell differentiation by culturing cells on biologic substrata, with varying degrees of success (1). Previous studies in our laboratory have shown that type II cells cultured in association with fetal rat lung fibroblasts on floating collagen gels, or on a reconstituted basement membrane gel derived from the Engelbreth-Holm-Swarm tumor, exhibit significantly improved maintenance of type II cell morphology, patterns of phospholipid biosynthesis, and expression of messenger RNAs (mRNAs) for SP-A, SP-B, and SP-C (2, 3). We have attributed the improvement in type II cell function in these culture systems to the re-establishment of requisite cell-extracellular matrix (ECM) interactions and maintenance of native type II cuboidal cell shape (4). Our previous studies also suggested that epithelial-mesenchymal interactions, which have been shown to be absolutely required for prenatal lung growth and differentiation, also had a positive influence on adult type II cell differentiation (2). In the present study we have developed a culture system that allowed us to directly examine the contribution of lung fibroblasts to adult type II cell differentiation. The results show that the presence of lung fibroblasts has a significant impact on the ability of type II cells to synthesize SPs and phospholipids, which is mediated in part by keratinocyte growth factor (KGF, also known as fibroblast growth factor [FGF]-7).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials

Matrigel was obtained from Collaborative Biomedical Products (Bedford, MA). Human recombinant KGF (or FGF-7) was obtained from Promega Biotech (Madison, WI). Neutralizing antibodies to human KGF were purchased from R&D Systems (Minneapolis, MN). Dexamethasone (DEX), forskolin (FSK), and 3-isobutyl-1-methylxanthine (IBMX) were obtained from Sigma Chemical (St. Louis, MO). Preparation of the rabbit polyclonal antibodies against rat SP-A (5) and rat SP-D (6) have been previously described and were provided by Dr. D. Voelker (National Jewish Medical and Research Center, Denver, CO). Rabbit polyclonal antibodies against mature bovine SP-B (#28031), human pro-SP-B (#96819), and human pro-SP-C (#R68514) were generously provided by Drs. J. Whitsett and T. Weaver (Children's Hospital Medical Center, Cincinnati, OH).

Isolation of Type II Cells

Type II cells were dissociated from the lungs of specific pathogen-free adult male Sprague-Dawley rats (Harlan, Indianapolis, IN) with porcine pancreatic elastase (Worthington Biochemicals, Freehold, NJ) and purified on discontinuous metrizamide gradients as previously described (7). Purified type II cells were resuspended in Dulbecco's modified Eagle's medium (DMEM) (GIBCO BRL, Gaithersburg, MD) containing 1% charcoal-stripped fetal (8) bovine serum (CSFBS) (Gemini Bio-Products, Calabasas, CA), 100 U/ml penicillin, 100 µg/ml streptomycin, 2.5 µg/ml amphotericin B (all from GIBCO BRL), and 10 µg/ml gentamicin sulfate (Sigma), and plated into the wells of six-well cluster dishes (Becton Dickinson, Lincoln Park, NJ) coated with 1 ml of Matrigel that had been diluted with DMEM (1:2, DMEM/Matrigel). Type II cells were seeded into each well at a density of 5 × 10⁵ viable cells/cm²; the day of isolation was considered Day 0 of culture.

Preparation of Fibroblast/Collagen Rafts

Lung fibroblasts isolated from Day 19 rat fetuses and from adult rats were prepared as previously described (2, 4) and used between passages 4-15. Adult human lung fibroblasts (HLF) (Repository #AG02262; Coriell Institute for Medical Research, Camden, NJ) were used between passages 3 and 6. Fibroblasts were resuspended in a small volume (approximately 250 µl) of DMEM, then mixed with neutralized rat-tail collagen solution that was prepared on ice as previously described (2), with the exception that 10× Waymouth's 752/1 medium was used for the neutralization solution. Unless otherwise noted, the final concentration of fibroblasts was 3.3 × 10⁶ cells/ml collagen. A total of 300 µl of collagen solution containing fibroblasts was carefully spread on the surface of a 25-mm-diameter, 0.8-µm-pore-size nucleopore filter (Costar, Corning, NY) and gelled at 37°C.

Preparation of Cocultures

Medium in the type II cell cultures was changed on Day 1 to remove nonadherent cells, and then a fibroblast/collagen raft was carefully transferred to each well with forceps. The hydrophobic nature of the polycarbonate filters caused the fibroblast/collagen rafts to float at or near the surface of the culture medium. Molecules secreted by the fibroblasts could therefore freely diffuse into the medium, but the physical separation of the type II cells and fibroblasts allowed us to analyze only the type II cells. Control cultures consisted of type II cells cultured with collagen rafts containing no fibroblasts, or type II cells cultured by themselves.

KGF Neutralization Experiments

To determine whether the observed effects of cocultured fibroblasts were due to the presence of KGF, antibody neutralization experiments were performed. Type II cells were isolated and plated as previously described, then cultured for 24 h. Rafts containing 1 × 10⁶ HLF were then added to the cultures, along with 10 µg/ml antihuman KGF (R&D Systems). A second group of cocultures received vehicle (phosphate-buffered saline [PBS]) only. As controls for the efficacy of the antibody, a third set of cultures was treated with 1 ng/ml human KGF, and a fourth set of cultures was treated with 1 ng/ml human KGF plus 10 µg/ml antihuman KGF. A final set of cultures was treated with 10 µg/ml antihuman KGF alone. Cultures were maintained for an additional 24 h, at which point the cells were harvested for RNA isolation and determination of SP mRNA content.

Preparation of RNA and Northern Blots

Type II cells on Matrigel were directly lysed into 4 M guanidinium isothiocyanate, 0.5% N-laurylsarcosine, and 0.1 M -mercaptoethanol in 25 mM sodium citrate buffer. Lysis was monitored visually to ensure that a minimum of Matrigel was solubilized into the lysate. Total cellular RNA was isolated by centrifugation through a 5.7 M CsCl cushion at 150,000 × g for 18 h. Northern blots were prepared and probed with complementary DNAs (cDNAs) for rat SP-A, SP-B, SP-C, and SP-D that had been radiolabeled to high specific activity with [-³²P]deoxycytidine triphosphate (dCTP) (ICN Pharmaceuticals, Costa Mesa, CA) by random-primed second-strand synthesis using a commercially available kit (GIBCO BRL). Hybridization and washing of blots were performed as previously described (3). Direct quantitation of radioactive counts on blots was performed using Molecular Dynamics ImageQuant software, version 3.22 (Molecular Dynamics, Sunnyvale, CA).

Ribonuclease Protection Assays

In some experiments, total RNA isolated from elutriated type II cells was analyzed for SP mRNA expression using a ribonuclease protection assay (RPA) that allowed simultaneous measurement of mRNAs for SP-A, SP-B, SP-C, and SP-D. Fragments of different sizes for the four SPs were isolated by polymerase chain reaction using full-length cDNAs as templates. The forward primers included a BamH1 restriction site added to the 5' end and the backward primers included an EcoR1 restriction site added to the 5' end to facilitate directional cloning into pGEM 4Z (Promega). The primers for SP-A were 5'-CGGATCCAGTCCTCAGCTTGCAAGGATC-3', coding sense and corresponding to nucleotides 424-444; and 5'-GGAATTCCGTTCTCCTCAGGAGTCCTCG-3', coding antisense and corresponding to nucleotides 549-569. The probe transcribed from this clone identified a fragment of 146 base pairs (bp). The primers for SP-B were 5'-CGGATCCGAGCAGTTTGTGGAACAGCAC-3', coding sense and corresponding to nucleotides 997-1017; and 5'-GGAATTCTGGTCCTTTGGTACAGGTTGC-3', coding antisense and corresponding to nucleotides 1152-1172. The probe transcribed from this clone identified a fragment of 176 bp. The primers for SP-C were 5'-CGGATCCCATACTGAGATGGTCCTTGAG-3', coding sense and corresponding to nucleotides 202-222; and 5'-GGAATTCTCTGGAGCCATCTTCATGATG-3', coding antisense and corresponding to nucleotides 381-401. The probe transcribed from this clone identified a fragment of 200 bp. The primers for SP-D were 5'-CGGATCCCGGAAGAGCCTTTTGAGGATG-3', coding sense and corresponding to nucleotides 831-851; and 5'-GGAATTCACAGTTCTCTGCCCCTCCATTG-3', coding antisense and corresponding to nucleotides 1054-1075. The probe transcribed from this clone identified a fragment of 245 bp.

The vectors were linearized with BamH1 and radiolabeled antisense probes were transcribed in vitro using a commercially available kit (Promega) and [-³²P]CTP (800 Ci/mmol; ICN). The RNA probe was purified on an 8% polyacrylamide/7 M urea gel and eluted from the most intense band detected by autoradiography. An 18S ribosomal RNA (rRNA) probe, which was synthesized using pT7 RNA 18S template and the T7 MEGAshortscript Kit (Ambion, Austin, TX) and [-³²P]CTP, was used as an internal standard for RNA quantitation. The amount of 2 to 5 µg of total cellular RNA was hybridized at 45°C for 24 h with 1 × 10⁵ counts/min (cpm) of antisense probe. Probe not protected by hybridization with target RNA was digested with a mixture of RNase A and T1 (RPAII kit; Ambion). The protected RNA duplex fragments were precipitated, resuspended, and separated on an 8% polyacrylamide/7 M urea gel. The gel was dried and exposed to Hyperfilm (Amersham Life Science Products, Arlington Heights, IL) at 70°C. Radioactive counts were obtained from protected fragments for SP-A, SP-B, SP-C, and SP-D by ImageQuant analysis and normalized to 18S rRNA. This assay gives linear results from inputs of 1 to at least 15 µg of total RNA.

Metabolic Labeling and Immunoprecipitation of SP

On Day 4 of culture, the medium and the fibroblast/collagen rafts were removed from the cultures and replaced with 2 ml of cysteine-methionine-free DMEM (ICN) plus 1% CSFBS and incubated for 30 min. This medium was then replaced with 0.5 ml of fresh medium containing 1 mCi/ml of TRAN³⁵S-Label (ICN; specific activity > 1,000 Ci/mmol). After 4 h of metabolic labeling, the cells were removed from the Matrigel with dispase; washed twice with PBS (pH 7.4); then lysed into 190 mM NaCl, 60 mM Tris (pH 7.4), 6 mM ethylenediaminetetraacetic acid, 4% sodium dodecyl sulfate containing 1 mM phenylmethylsulfonyl fluoride, 5 mM N-ethylmaleimide, and 10 µg/ml pepstatin A (all from Sigma). Immunoprecipitation of radiolabeled proteins was then performed as described elsewhere (9) and autoradiograms were prepared. Rabbit polyclonal antibodies against rat SP-A, rat SP-D, human pro-SP-C, and bovine SP-B (antibody #28031) were used for immunoprecipitation.

Immunohistochemistry

For immunohistochemistry, type II cells were first released from the Matrigel by incubation in dispase for 1 h at 37°C. Cell aggregates were pelleted by centrifugation at 150 × g, resuspended in PBS, and recentrifuged. The pelleted aggregates were then resuspended in 250 µl of cold neutralized rat-tail collagen solution. This suspension was transferred to a small piece of Parafilm and allowed to gel at 37°C. The gel was then fixed in freshly prepared 4% paraformaldehyde in PBS and embedded in paraffin. This collagen-enrobing procedure allowed us to evaluate cross sections of many aggregates in one section. Sections 4 µm thick were immunostained with polyclonal antibodies against rat SP-A, human pro-SP-B (antibody #96819), and human pro-SP-C using standard avidin-biotin complex techniques (6). Sections that were immunostained for SP-A were subjected to antigen retrieval using 6 N guanidinium hydrochloride and 0.2 mg/ml trypsin as previously described (6).

Labeling and Analysis of Lipids

The pattern and quantity of lipid synthesis by type II cells cultured under the various conditions was performed as described previously (2). Briefly, on Day 3 of culture the cells were incubated with 10 µCi/ml [1-¹⁴C]acetate (ICN; sp. act. 25-60 µCi/ mmol) for the final 24 h before harvesting. Lipids were extracted with methanol and chloroform, then separated by two-dimensional thin layer chromatography on DC-Fertigplatten Kieselgel 60 plates (E. Merck, Darmstadt, Germany). Individual phospholipid spots were scraped directly into scintillation vials, their incorporated radioactivity was determined, then the incorporation of acetate into each phospholipid species as a percentage of total phospholipids was determined. The percentage of saturated phosphatidylcholine (PC) was determined by the method of Mason and Nellenbogen (10). We also determined the amount of radiolabeled acetate incorporated into neutral lipids. Because some of the treatments might result in cell proliferation, the relative amount of lipid synthesis was normalized to DNA content. Therefore, parallel cultures of each condition were prepared in triplicate. On the day of harvest, the type II cells were removed from the Matrigel substratum by incubation in dispase for 1 h at 37°C and pelleted by centrifugation, then DNA content was determined fluorimetrically.

Statistics

Data were analyzed using the JMP 4.0 and SAS 6.12 computer software packages (SAS Institute, Cary, NC).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Lung Fibroblasts Increase Expression of SP mRNAs

Our initial experiments compared lung fibroblasts from three different sourcesembryonic rat lung, adult rat lung, and adult human lungfor their ability to maintain SP mRNA expression. As shown in Figure 1, coculture of type II cells with any of the three of the fibroblast lines resulted in a marked increase in the relative abundance of mRNAs for SP-A, SP-B, SP-C, and SP-D. Because the efficacy of the different fibroblasts appeared essentially equivalent, the remainder of the experiments were done using the HLF.

View larger version (100K): [in this window] [in a new window]   Figure 1.   Coculture of adult rat type II cells with lung fibroblasts increases SP gene expression. Adult rat type II cells were seeded onto Matrigel and allowed to adhere for 24 h, whereupon the medium was changed to remove nonadherent cells. The cells were then cultured for an additional 72 h with no additions (lane 1), or cocultured with collagen rafts containing 1 × 106 fibroblasts from either Day 19 fetal rat lung (lanes 2 and 3), adult rat lung (lanes 4 and 5), or adult human lung (lanes 6 and 7). The filter supporting the collagen gel was placed in the medium with the gel facing either down (lanes 2, 4, and 6) or up (lanes 3, 5, and 7). RNA was isolated by the guanidinium-CsCl method; electrophoresed under denaturing conditions; blotted; and sequentially probed with radiolabeled cDNAs for rat SP-A, SP-B, SP-C, and SP-D. Each lane was loaded with 3.5 µg total RNA. The bottom panel shows a photomicrograph of the ethidium bromide-stained gel to demonstrate equivalence of loading. The data show that all three types of lung fibroblasts increased expression of all four SP mRNAs, and that they were equivalent in their efficacy. Whether the collagen raft faced down or up had no effect on SP gene expression.

We next determined the number of HLF required to elicit the observed response. Rafts containing 1 × 10⁴ HLF gave no response above that seen in cells cultured alone or with a bare nucleopore filter (Figure 2). Rafts containing 1 × 10⁵ HLF elicited an apparent small increase in the amount of all four SP mRNAs, but these increases reached statistical significance only for SP-A and SP-B (Table 1). Significant increases in all of the SP mRNAs (SP-A mRNA: 15-fold; SP-B mRNA: 8-fold; SP-C mRNA: 3- to 4-fold; SP-D mRNA: 4- to 5-fold) were seen when type II cells were cocultured with 1 × 10⁶ HLF. Increasing the number of HLF to 2 and 3 × 10⁶ cells per raft, however, had no further beneficial effect on SP gene expression (data not shown).

View larger version (70K): [in this window] [in a new window]   Figure 2.   The response of adult rat type II cells to stimulation by HLF is dose-dependent. Adult type II cells were seeded onto Matrigel and cultured for 4 d. The cells were cultured for the final 72 h in the presence of no additions (lane 1), or cocultured with a collagen raft containing no cells (lane 2), or 1 × 104 (lane 3), 1 × 105 (lane 4), 1 × 106 (lane 5) adult HLF. Northern blots were prepared with 2.5 µg total RNA loaded per lane, then sequentially probed for SP-A, SP-B, SP-C, and SP-D. The bottom panel shows a photomicrograph of the ethidium bromide-stained gel to demonstrate equivalence of loading. This figure shows that 1 × 104 HLF had no effect on SP gene expression. Increasing the number of HLF to 1 × 105 resulted in a slight increase in mRNA levels for all four SPs. Increasing the number of HLF to 1 × 106, however, had a marked stimulatory effect on SP mRNAs.

We then examined the ability of type II cells cultured in the presence of HLF to translate and process SP mRNAs. Type II cells cultured for 72 h in the presence of HLF were labeled for 4 h with [³⁵S]-methionine and [³⁵S]-cysteine and harvested, and cell lysates were sequentially immunoprecipitated with specific antibodies against SP-A, SP-B, SP-C, and SP-D. The results are shown in Figure 3. Immunoprecipitated SP-A gave a single dominant band of molecular mass 32-34 kD, consistent with what has been described for forms of SP-A containing immature oligosaccharide (5). Immunoprecipitation with an antibody against mature bovine SP-B (antibody #28031) gave a pattern similar to that seen in cultured human fetal lung explants (11): precursor peptides with molecular masses of 42 and 25 kD, along with processed proteolipids at 8 and 12 kD. An antibody against amino acids 1-20 of human pro-SP-C (antibody #68514) precipitated a 22-kD band of rat pro-SP-C, as well as its processing intermediates (12). Immunoprecipitation with antirat SP-D gave a single band at approximately 43 kD. Thus, it appears that type II cells in this culture system not only accumulated SP mRNAs but also translated them and processed the resultant peptides through normal pathways. Because type II cells cultured on Matrigel form multicellular aggregates with their apical surfaces directed inward, it should be noted that the SPs immunoprecipitated in these experiments represent intracellular peptides as well as any radiolabeled proteins that may have been secreted into the lumina during the labeling period.

View larger version (37K): [in this window] [in a new window]   Figure 3.   Translation and processing of SPs by adult type II cells cocultured with HLF. Adult type II cells were seeded onto Matrigel and cultured for 4 d, with the final 72 h in the presence of 1 × 106 HLF. During the last 4 h of culture the cells were labeled with 35S-cysteine and -methionine; lysed; immunoprecipitated with specific antibodies against SP-A, SP-B, SP-C, and SP-D; and electrophoresed on polyacrylamide gels under reducing conditions. Antibodies against SP-A and SP-D precipitated single bands at 32-34 and 43 kD, respectively, which correspond to the reported size of monomers of these proteins. Antibodies against SP-B precipitated 42- and 25-kD precursor proteins, as well as processed proteolipids at 8 and 12 kD. The antibodies against SP-C precipitated rat pro-SP-C at 22 kD, along with processing intermediates. These data demonstrate that adult type II cells cocultured with HLF translate SP mRNAs and process the precursor proteins.

We estimated the percentage of type II cells in the aggregates that were positive for the different SPs by immunohistochemistry. As can be seen in Figure 4, most of the epithelial cells were positive for SP-A, SP-B, and SP-C. The presence of immunoreactive material in the aggregate lumina suggested that these proteins were being secreted.

View larger version (140K): [in this window] [in a new window]   Figure 4.   Immunohistochemical staining of SPs in adult type II cells cocultured with HLF. Type II cells were cultured for 4 d on Matrigel, with the last 72 h in coculture with 1 × 106 HLF. Aggregates were released from the substratum with dispase, resuspended in liquefied rat-tail collagen that was subsequently gelated, and fixed in paraformaldehyde. Paraffin sections were stained with antibodies against rat SP-A (A), human pro-SP-B (B), and human pro-SP-C (C). The epithelial cell aggregates contained many cells that stained positive for SP-A, SP-B, or SP-C. Eliminating primary antibodies resulted in no immunostaining (D). All panels are shown at the same magnification; original magnification, ×140.

DEX Affects SP Expression in Type II Cell-Lung Fibroblast Cocultures

We next examined the effects of soluble factors in the coculture system. We chose to test DEX (10⁷ M) and cyclic adenosine monophosphate (cAMP). DEX was chosen because glucocorticoids have been shown to affect the surfactant system in the developing and adult lung, both in vivo and in vitro (13). We tested the effect of increasing intracellular cAMP (via treatment with 10 µM FSK + 10 µM IBMX) because it has been shown to affect both SP gene expression and surfactant phospholipid biosynthesis (14). In addition, we have previously observed that cAMP improved functional differentiation of adult rat type II cells maintained in serum-free conditions (15). The results of these experiments are shown in Figure 5 and Table 2.

View larger version (86K): [in this window] [in a new window]   Figure 5.   Effect of soluble factors on SP gene expression in cultured adult rat type II cells. Adult type II cells were seeded onto Matrigel and cultured for 4 d. Total RNA was isolated, and 3 µg from each condition was used in a RPA for SP mRNAs, along with 18S rRNA as a normalization control. The final 72 h of culture were under the following conditions: lane 1, no additions; lane 2, + 107 M DEX; lane 3, + 10 µM FSK; 10 µM FSK/IBMX; lane 4, cocultured with 106 HLF; lane 5, + HLF, DEX; lane 6, + HLF, FSK/IBMX; lane 7, + 10 ng/ ml KGF; lane 8, + KGF, DEX; lane 9, + KGF, FSK/IBMX. Lane 0 is RNA extracted from freshly isolated type II cells. This representative RPA shows that coculture of type II cells with HLF had a decided positive effect on the accumulation of SP mRNAs, which was partially mimicked by KGF. Although the addition of DEX by itself had no effect on SP gene expression, when combined with HLF or KGF it had either positive (SP-B), negative (SP-A, SP-C), or no effects (SP-D). FSK/IBMX by itself stimulated SP-C mRNA accumulation, but did not alter expression of any of the SPs in response to HLF or KGF.

DEX given by itself had no significant effect on the level of SP-A mRNA. However, DEX antagonized the effects of HLF on SP-A mRNA expression, reducing the stimulation from 13-fold to 6-fold. The effect of DEX on SP-B expression was quite dramatic. Addition of DEX alone to type II cell cultures caused no significant stimulation of SP-B mRNA levels. However, when DEX was added to type II cells cocultured with HLF, the level of SP-B mRNA was increased 15-fold over that seen in type II cells cultured in the absence of any additions. The level of SP-B mRNA expression in these cells was equivalent to that seen in freshly isolated rat type II cells. The effect of DEX on SP-C mRNA expression was similar to that seen for SP-A: DEX by itself had no significant effect on SP-C mRNA expression, and significantly antagonized the positive (3- to 4-fold increase) effects of HLF on SP-C expression. DEX treatment by itself had no significant effect on SP-D mRNA levels, and also had no effect on the stimulation of SP-D mRNA caused by coculture with HLF.

Type II cells treated with FSK/IBMX as a single addition to the medium showed no significant differences in their content of mRNAs for SP-A, SP-B, and SP-D (Figure 5 and Table 2). SP-C mRNA levels, however, were significantly increased by elevating intracellular cAMP. When added to type II cells cocultured with HLF, FSK/IBMX had no effect on the accumulation of mRNAs for SP-B, SP-C, and SP-D, but significantly antagonized the positive effects of HLF on the accumulation of SP-A mRNA, reducing HLF stimulation from 13- to 8-fold.

KGF Is a Mediator of Lung Fibroblast Effects on Type II Cells

A number of recent studies have suggested that KGF (FGF-7) may serve as a signaling molecule mediating interactions between lung mesenchyme and epithelium (6, 16). In a previous study (17) we demonstrated that KGF stimulated accumulation of mRNAs for SP-A and SP-B, but not SP-C, when given for the last 2 d of a 6-d culture period. In the present study we examined type II cells treated with KGF for 3 d, beginning 24 h after plating. The results of exposing type II cells to increasing amounts of KGF are shown in Figure 6 and Table 3. KGF stimulated increased expression of all four SP mRNAs at a concentration of 1 ng/ml. The stimulation of SP-A, SP-B, and SP-D mRNAs peaked at 10 ng/ml and remained level thereafter. SP-C mRNA levels also peaked at 10 ng/ml, but then significantly declined with further increases in KGF.

View larger version (73K): [in this window] [in a new window]   Figure 6.   The effects of KGF on SP gene expression by cultured adult type II cells is dose-dependent. Type II cells were seeded onto Matrigel and cultured for 4 d, with the final 72 h in the presence of different concentrations of KGF. Northern blots were prepared using 2.5 µg of total RNA per lane and sequentially probed with rat cDNAs for SP-A, SP-B, SP-C, and SP-D. The bottom panel shows a photomicrograph of the ethidium bromide-stained gel to demonstrate equivalence of loading. Conditions: lane 1, no additions; lane 2, + 0.1 ng/ml KGF; lane 3, + 1 ng/ml KGF; lane 4, + 5 ng/ml KGF; lane 5, + 10 ng/ml KGF; lane 6, + 50 ng/ml KGF; lane 7, + 100 ng/ml KGF; lane 8, + 250 ng/ml KGF. This representative Northern blot shows that as little as 1 ng/ml KGF elicits an obvious increase in SP mRNA expression in cultured type II cells. These responses peaked at approximately 10 ng/ml. High concentrations of KGF (250 ng/ml) resulted in decreased signal intensity.

We then compared the effects of KGF (at 10 ng/ml) on SP gene expression with those obtained when type II cells were cocultured with HLF. The results are shown in Figure 5 and Table 2. In these experiments there was no significant difference between KGF and HLF in their ability to stimulate accumulation of mRNAs for SP-B, SP-C, and SP-D. KGF treatment also induced a significant (7-fold) stimulation in SP-A mRNA, but this was less than that seen with HLF (13-fold). The further addition of DEX to KGF-containing medium gave results identical to those seen when DEX was added to cocultures of type II cells and HLF; DEX augmented stimulation of SP-B mRNA, but antagonized the positive effects of KGF on SP-A and SP-C mRNAs. Addition of FSK/IBMX to KGF-containing medium had no effect on mRNA levels for SP-B, SP-C, and SP-D. SP-A mRNA levels appeared to be reduced, but these changes did not reach statistical significance.

The similarity in the responses of type II cells to HLF to that seen with KGF suggested that KGF, which is produced by lung fibroblasts (16), might be mediating the positive effects of HLF. To address this possibility, we performed experiments in which type II cells and HLF were cocultured in the presence of a neutralizing antibody against human KGF. Because of uncertainty about how long the antibody would retain activity in culture, the duration of the experimental treatment was reduced to 24 h. The results are shown in Figure 7 and Table 4. Treatment of type II cells with 1 ng/ml KGF resulted in a significant stimulation in the levels of all four SP mRNAs. Simultaneous incubation with anti-KGF antibody ablated the effects of exogenously added KGF in every case. Incubation with the antibody by itself had no effect on SP mRNA levels. Type II cells cocultured with HLF for 24 h showed significant increases in levels of SP-A and SP-B mRNAs, but no significant effect on SP-C and SP-D mRNAs. Addition of anti-KGF antibody to the cocultures neutralized the positive effects of HLF, but only partially: whereas SP-A and SP-B mRNA levels were significantly reduced compared with those seen in cocultures, they remained significantly elevated over those seen in type II cells cultured with no additions.

View larger version (91K): [in this window] [in a new window]   Figure 7.   Anti-KGF antibodies partially neutralize the effects of HLF on cultured adult type II cells. Adult alveolar type II cells were seeded onto Matrigel and cultured with no additions for 24 h, then for an additional 24 h in the conditions listed. Total RNA was isolated, and 2.5 µg used in a RPA for SP mRNAs, with 18S rRNA used for a normalization control. Conditions: lane 1, no additions; lane 2, + anti-KGF, 10 µg/ml; lane 3, + anti-KGF, 10 µg/ml; lane 4, + 106 HLF; lane 5, + HLF, anti-KGF; lane 6, + HLF, anti-KGF; lane 7, + KGF, 1 ng/ml; lane 8, + KGF, anti-KGF; lane 9, + KGF, anti-KGF. This representative RPA shows that HLF stimulated accumulation of SP-A and SP-B, but not SP-C and SP-D, mRNAs. KGF increased expression of all four SP mRNAs. The addition of anti-KGF antibody by itself to the medium had no effect on the cells. Anti-KGF antibody completely ablated the effects of KGF on SP gene expression, but only partially neutralized the effects of HLF.

Lung Fibroblasts and KGF Increase Synthesis of Phospholipids in Cultured Type II Cells

Because surfactant is composed primarily of phospholipids, we next determined the effects of HLF on the ability of type II cells to incorporate radiolabeled acetate into lipids during the last 24 h of culture. Because the type II cells and HLF are physically separated in the coculture system, we were able to examine acetate incorporation in type II cells only. Because of the significant effects that KGF had on SP mRNA levels, we also examined acetate incorporation in type II cells treated with KGF. Finally, we determined the effects of adding DEX to HLF/type II cell cocultures, and to type II cells cultured with KGF. The results are shown in Table 5.

Normalized to micrograms of DNA, HLF elicited a 9- to 10-fold increase in the amount of acetate incorporated into phosopholipids, and a 2- to 3-fold increase in acetate incorporated into neutral lipids. This effect was mimicked by the substitution of KGF for HLF. The effects of HLF and KGF differed substantially, however, when DEX was added to the cultures. DEX by itself stimulated acetate incorporation into phospholipids 3- to 4-fold, but did not significantly affect incorporation into neutral lipids. When added to type II cell/HLF cocultures, DEX had a significant inhibitory effect on acetate incorporation into phospholipids, but no effect on neutral lipid synthesis. In contrast, DEX added in combination with KGF had a significant positive effect on acetate incorporation into phospholipids, but no apparent effect on neutral lipid synthesis.

We next examined the pattern of acetate incorporation into individual phospholipid species (Table 6). Normalized to cellular DNA, type II cells cocultured with HLF incorporated 10 times more acetate into PC, saturated PC (sat PC), and phosphatidylglycerol (PG) than type II cells cultured with no additions. These effects were essentially duplicated by adding KGF to type II cell cultures in place of HLF. Significant differences were observed, however, when DEX was added to the cultures. DEX added by itself to cultured type II cells caused a significant increase in the amount of acetate incorporated per microgram of DNA into PC (4-fold), sat PC (4-fold), and PG (6-fold). When added to type II cell/HLF cocultures, DEX significantly antagonized the positive effects of HLF on the incorporation of acetate into PC and sat PC, but appeared to increase acetate incorporation into PG, although this did not reach statistical significance. When compared with cultures treated with KGF alone, concomitant treatment with DEX and KGF did not significantly affect acetate incorporation into PC or sat PC. Synthesis of PG, however, was substantially increased, rising 32-fold above that seen in type II cells cultured in control medium.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

It has long been appreciated that pulmonary mesenchyme is required for normal lung development. Diffusible signals produced by the embryonic mesenchyme have profound effects on the growth, patterning, and differentiation of the developing lung epithelium (for reviews, see 18). The normal alveolar wall in the adult, however, contains relatively few fibroblasts compared with the number seen associated with the developing epithelium. Our present observations suggest an important role for fibroblasts in the adult alveolusthat epithelial-fibroblast interactions may also be required for the maintenance of type II cell differentiation in the adult. Our data show that coculture of adult type II cells with lung fibroblasts has a marked positive effect on their differentiated function: mRNA levels for SP-A, SP-B, SP-C, and SP-D were all substantially increased; these mRNAs were translated and processed; and surfactant phospholipid biosynthesis was elevated 10-fold. Because the epithelial cells and fibroblasts were not allowed to come into contact, this improved differentiation was mediated by diffusible factors produced by the fibroblasts. Further, the improvement we observed in type II cell differentiation is not transitory. Whereas the duration of culture in the present experiments was 4 d, in other studies (Pan and colleagues, submitted) we have shown that type II cells cocultured with HLF will maintain highly differentiated characteristics for at least 2 wk.

We have previously observed (2) that adult rat type II cells cultured in association with lethally irradiated fibroblasts on floating collagen gels incorporated a higher percentage of acetate into phospholipids, notably PC and sat PC, than cells cultured on tissue culture plastic or in association with fibroblasts on attached collagen gels. This culture system, although useful, had several disadvantages. The most obvious of these is that the fibroblasts and type II cells were intimately associated with each other, thus all analyses were necessarily done on the combined cell types. This has been eliminated in the present culture system, where the epithelial cells and fibroblasts are physically separated from one another. A second disadvantage of the floating feeder layer system was that the fibroblasts were growth-arrested with irradiation. Although this prevented them from overgrowing the type II cells, the fibroblasts eventually died, making long-term culture difficult. This problem has been circumvented in the present culture system, where the fibroblasts are left untreated.

Coculture of fibroblasts with adult type II cells has been done by other investigators. Mangum and associates (19) cocultured primary rat type II cells and early passage rat lung fibroblasts on opposite sides of a collagen-coated polycarbonate filter. The type II cells formed a tight monolayer, but the cells appeared ultrastructurally dedifferentiated. Adamson and Young (20) showed that type II cells cultured on an endothelial cell ECM coating the surface of a culture insert incorporated significantly more [³H]thymidine when fibroblasts were present in the lower culture chamber. This activity was neutralized by the addition of anti-KGF antibodies. Pasternack and coworkers (21) demonstrated that medium conditioned by low-passage adult lung fibroblasts stimulated a significant increase in thymidine incorporation by adult type II cells. In contrast to the studies of Adamson and Young, Pasternack and colleagues could not demonstrate that the increased proliferation was due to KGF. A common characteristic of these culture systems is that the type II cells assumed a more flattened, squamous cell shape, which we have previously shown is not conducive to maintenance of type II cell differentiation (4, 22). The Matrigel substratum used in the present study allowed the type II cells both to interact with a basement membrane and to assume their native cuboidal cell shape. We believe that these conditions are necessary for the cells to be competent to respond to soluble factors provided by HLF, or to those added to the medium.

Our data indicate that KGF, which is produced by lung fibroblasts, is a major mediator of the positive effects of HLF on type II cells. These observations are in agreement with those of Chelly and associates (23), who demonstrated that KGF stimulated expression of SP-A, SP-B, and SP-C in cultured fetal rat type II cells, as well as incorporation of choline and acetate into sat PC. These increases in choline and acetate incorporation were correlated with increases in choline phosphate cytidyltransferase (CPCT) and fatty-acid synthase (FAS) activities, respectively. Together these data suggested an important role for KGF in the maturation of the surfactant system during late development. Our results indicate that the requirement for KGF may extend beyond fetal lung development, and that adult type II cells require KGF to sustain maximum levels of differentiation.

As suggested by our antibody neutralization experiments, however, KGF may not be the only active molecule provided by HLF. In these studies the anti-KGF antibody was able to completely neutralize the effects of exogenous KGF at a concentration (1 ng/ml) that elicited a response equal to or greater than the effects of coculturing with HLF. When added to type II cell/HLF cocultures, however, the antibody was only partially effective: it reduced SP-A and SP-B mRNAs to levels that were both significantly less than in cocultures treated with vehicle, but the levels remained significantly greater than type II cells cultured by themselves. This suggests that HLF produce a factor(s) in addition to KGF that stimulates accumulation of SP-A and SP-B mRNAs. SP-C mRNA levels did not increase in type II cells cocultured with HLF during the 24-h time course of these experiments, thus the antibody had no measurable neutralizing effect. Addition of KGF, however, modestly increased SP-C mRNA expression, and this was neutralized by the anti-KGF antibody.

How KGF affects expression of SP-C has yet to be completely defined. In a previous study we observed that KGF had little effect on SP-C mRNA levels when given to type II cells that had been precultured for 4 d in medium with no additions except serum (17). In the present study, however, we observed a 4-fold increase in SP-C mRNA in response to 10 ng/ml KGF, with levels decreasing significantly at higher concentrations of KGF; this biphasic dose effect was also observed by Chelly and coworkers (23) using fetal rat type II cells. Our present observations are in agreement with those in a previous report (6), in which type II cells were cultured at an air-liquid interface. In these earlier experiments KGF significantly increased SP-C mRNA levels, although not to the same extent we report here. Some insight into the results of Sugahara and colleagues (17) may be gained from results obtained in cultures of embryonic epithelium. KGF can sustain SP-C expression in embryonic distal lung epithelium when these rudiments are maintained in mesenchyme-free culture (24). However, we also observed (Shannon, unpublished observations) that embryonic lung epithelial rudiments became refractory to KGF stimulation when they were cultured in its absence for 48 h. Thus, it is possible that the lack of a KGF effect on SP-C mRNA levels observed in our earlier study was the result of withdrawing KGF stimulation for 96 h. The observation that SP-A and SP-B mRNAs were stimulated in that study (17) whereas SP-C mRNA was unaffected or inhibited illustrates the independent regulation of expression of all the SP genes.

KGF has myriad effects on pulmonary epithelial cells. Prenatally, KGF stimulates proliferation and differentiation of the presumptive distal lung epithelium in culture (24). KGF is necessary, but insufficient by itself, to effect transdifferentiation of embryonic tracheal epithelium to express a distal lung phenotype (25). The concentration of KGF to which the epithelial cells are exposed is clearly important, however, because targeted overexpression of KGF leads to developmental malformations consistent with pulmonary cystadenoma (26). Treatment of fetal lung explants in vitro with KGF results in cystic dilation of the epithelium (27), which results from epithelial fluid secretion driven by cystic fibrosis transmembrane conductance regulator-independent Cl transport (28).

The lung epithelial hyperplasia in the studies cited earlier results from pharmacologic levels of KGF. The question arises as to the role of KGF in the normal lung. The adult lung contains levels of KGF mRNA equivalent to those seen in skin (29), yet lung epithelial proliferation under normal conditions is comparatively low. KGF in the normal lung may be sequestered by binding to heparan sulfate proteoglycan in the ECM and thus might not become bioactive until released. Our data, however, suggest that KGF may play a nonmitogenic role in the adult lung that of maintaining type II cell differentiation. The role of KGF as a differentiation factor is supported by its ability to specify the distal lung epithelial phenotype in embryonic respiratory epithelium (25). If KGF is involved in maintenance of adult type II cell differentiation, however, its absence can be compensated for by other FGFs, because mice null for KGF have no lung phenotype (30).

The ability of glucocorticoids to accelerate maturation of the surfactant system in the developing lung has been documented in numerous studies (13), but less is known about their effects on the surfactant system in the adult lung. Young and Silbajoris (31) demonstrated that treatment of adult rats with DEX for 1 wk resulted in 2- to 4-fold increases in total phospholipid and sat PC. Betamethasone administered to adult rats has been shown to upregulate CPCT, the rate-regulatory enzyme for PC synthesis (32). In our cultures, DEX treatment by itself significantly stimulated overall acetate incorporation into phospholipids, which is probably due to enhanced expression of FAS. Xu and Rooney (33) have shown that DEX increases FAS transcription and mRNA stability in explants of fetal rat lung, with a resultant increase in FAS mass and activity. When DEX was given in conjunction with KGF to our cultures, the effect was additive, which implies that DEX and KGF increase FAS activity by different mechanisms. In contrast, when given in conjunction with HLF, DEX significantly reduced acetate incorporation into phospholipids.

When individual phospholipid species were examined, DEX significantly stimulated acetate incorporation into PC, sat PC, and PG. DEX antagonized the positive effects of HLF on PC synthesis, but had no effect on sat PC or PG. DEX did not significantly alter the effects of KGF on PC and sat PC synthesis, but had a superadditive effect on PG synthesis. The different responses of type II cells to DEX when given in combination with KGF or HLF cells support the idea that HLF produce a bioactive factor(s) in addition to KGF. These data also underscore the complexity and independent regulation of individual surfactant phospholipids.

In summary, we have demonstrated that adult lung fibroblasts exert a decided positive influence on primary cultures of adult type II cells. These effects are mediated in part by KGF, but other as-yet-unidentified factors also play a positive role. The type II cell aggregates formed in this culture system do not easily allow the measurement of secretion. However, combining the strategies described in this study with other variations in culture conditions, such as maintaining the cells at an air-liquid interface, where cell apices are exposed to the medium (6, 34), may lead to a culture system in which regulation of the synthesis and secretion of surfactant components can be studied in long-term primary culture. These possibilities are currently under investigation.