INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Deep inflation of the lungs provokes stretching of the alveolar epithelial barrier and is regarded as the predominant physiologic stimulus for surfactant secretion in alveolar epithelial type II cells (AET₂) (1, 2). In this distinct cell type, at the alveolar surface both lipid and protein surfactant components are stored in cytoplasmic organelles, the lamellar bodies, which are released by classic exocytosis. Translocation of the lamellar bodies to the apical epithelial membrane and fusion of the lamellar body and the plasma membrane are essential steps in the later stage of the secretory process. Cytoskeletal components and contractile proteins have been suggested as being part of mechanotransduction pathways (3, 4). Previous histologic examinations revealed the presence of actin beneath the cell surface in AET₂ in close association with lamellar bodies (5, 6), and actin-binding proteins such as spectrins and annexins have been implicated in the surfactant secretory process (7). Experiments employing cytochalasins to probe the role of actin in the process of alveolar epithelial surfactant secretion did, however, give controversial results. Both a reduction of secretory events as well as an increase in the release of surfactant was noted (1, 8). Such inconsistencies may be related to the limited specificity of cytochalasins. These agents bind to different cellular target sites, resulting in a complex interaction with the microfilamentous network (9, 10) and they may effect a redistribution of actin rather than reducing the total quantity of filamentous actin (11).

A recently described specific tool for evaluating the impact of actin assembly on cell functions is botulinum C₂ toxin, a binary toxin consisting of the components C_2I and C_2II. C_2II promotes the translocation of C_2I into eucaryotic cells, and C_2I selectively ribosylates monomeric G-actin with adenosine diphosphate (ADP), which drastically reduces the actin polymerization rate (12). In previous histologic studies employing C₂ toxin for investigating the role of filamentous actin in pulmonary cell function, we noted a critical role of this microfilament system for the maintenance of capillary endothelial barrier function as well as enhanced lamellar body membrane fusion and secretion in C₂ toxin-treated lungs (13). Here we describe that C₂ toxin treatment of purified rat AET₂ results in marked amplification of surfactant secretion in parallel with the toxin-elicited F-actin decay and G-actin accumulation. In addition, the epithelial cells were noted to respond with rapid-onset, transient actin depolymerization as part of their secretory response to different stimuli. These data support a significant role of the actin microfilament system in the regulation of surfactant secretion in alveolar epithelial cells.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Male CD-18 Sprague-Dawley rats (180-200 g) were purchased from Charles River (Sulzfeld, Main, Germany). Elastase (type EC 134; specific activity, 135 U/mg protein) was purchased from Elastin Products Company (St. Louis, MO). Dulbecco's modified Eagle's medium (DMEM) was obtained from Gibco (Karlsruhe, Germany). [³H]- methyl choline and [³²P]nicotinamide-adenine dinucleotide (NAD) was from Amersham (Dreieich, Germany). Tissue culture plastic was purchased from Becton-Dickinson (Heidelberg, Germany). Silicone elastomer Silgard 184 was obtained from Dow Corning Corporation (Wiesbaden, Germany).

Isolation of AET₂

AET₂ were isolated as described in detail previously (1). Briefly, inflated and perfused lungs from specific pathogen-free male CD-18 Sprague-Dawley rats were lavaged and filled to total lung capacity with solution containing elastase (30 U/ml) and trypsin (0.05 mg/ml). Lungs were minced and free cells were separated from lung tissue by sequential filtration through sterile gauze and 100- and 10-µm nylon mesh. "Panning" of the resultant cell suspension was performed by rat immunoglobulin G-coated plates. Nonadherent type II cells were harvested after 1 h and resuspended in DMEM containing 10% fetal calf serum. The yield of type II epithelial cells from each rat was in the range of 30-50 × 10⁶ cells. The percentage of type II cells was 95 ± 3% as assessed by modified Papanicolaou, tannic acid, and alkaline phosphatase staining. Contaminating cells included alveolar macrophages (< 4% in all experiments) and neutrophils (< 2%). AET₂ viability, as assessed by 5-carboxyfluorescein diacetate (CFDA) loading and trypan blue exclusion, ranged persistently greater than 95%. Lactate dehydrogenase (LDH) release as one indicator of cellular damage ranged less than 2% of total enzyme activity as compared with the total release in response to the pore-forming agent mellitin (100 µg/ml).

Determination of Surfactant Secretion

Freshly isolated cells were seeded at a density of 5 × 10⁵ cells per well on 12-well culture dishes. The plating density was typically 8 × 10⁴/cm² ( 64% plating efficiency). After pre-labeling with [³H]choline (1 µCi/ ml) for 18 h, the cells were washed and medium was replaced by unlabeled serum-free DMEM containing C₂ toxin or secretagogues or both. After incubation for preset time periods, medium was removed and centrifuged to sediment possibly detached cells, which were not regained and used in the assays. The monolayer was scraped off and lipids from both medium and cells were extracted according to the Folch partition (16). The amount of phosphatidylcholine (PC) secretion was calculated as the percentage of [³H]PC contained in the medium as related to the total amount of [³H]PC in cells plus medium. To determine the effect of mechanical distension on surfactant secretion of type II cells, a stretching device as described by Wirtz and Dobbs (1) was employed. Distension was achieved by applying hydrostatic pressure to the bottom of the plates (silicon membranes with an area of 9.6 cm²), with an overall increase in area of 18-25%. Separate control experiments with thin-layer chromatography ascertained that the radioactive lipids secreted from both unstimulated and stimulated cells are 95 ± 2% PC (data not given in detail).

Measurement of G- and F-Actin

Cells were cultured on 35-mm dishes at a density of 3 × 10⁶ cells per well (plating density was typically 4 × 10⁵/cm²). Measurement of G-actin was performed in the supernatant of homogenized cells by employing a deoxyribonuclease (DNAse)-I inhibition assay as described (17). Briefly, cell lysates were co-incubated with defined amounts of DNAse and DNA (reaction assessed by absorbency at 260 nm), and the amount of G-actin was calculated from the DNAse inhibition by comparison with a standard curve. F-actin was measured by binding of rhodamine-phalloidin to actin filaments in permeabilized and formaldehyde-fixed type 2 cells as described (18), and the rhodamine content in the methanolic extract was determined with spectrofluorophotometry.

Polyacrylamide Gel Electrophoresis (PAGE) of [³²P]ADP Ribolysated AET₂ Protein

ADP ribosylation was performed essentially as described by Aktories and colleagues (19). AET₂ were treated on culture dishes with different concentrations of C₂ toxin for 2 h. The cells were transferred to test tubes, pelleted, and lysed by addition of hypotonic medium (5 mM EDTA, 10 mM triethanolamine). ADP ribosylation with C₂ toxin component I was again performed on cell lysates in the presence of [³²P]NAD for 1 h. The reaction was stopped by addition of trichloracetic acid and the labeled proteins were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), according to the procedure of Lämmli (20). The gel was autoradiographed on Kodak X-Omat film with an exposure time of 2 d. Quantification was done by means of densitometric scanning at 550 nm using a thin-layer chromatography scanner II from Camag (Muttenz, Switzerland).

Statistical Analysis

For statistical comparison, one-way analysis of variance was performed. A level of p < 0.05 was considered to be significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Influence of C₂ Toxin and Phalloidin on the G- and F-Actin Ratio

Actin represents approximately 20-25% of the AET₂ total protein content with approximately equal contribution of both G- and F-actin (Table 1). Treatment of the epithelial cells with C₂ toxin caused a dose-dependent progressive depletion of the cellular F-actin content to values ranging below 20% of baseline data (Figure 1). Concomitantly, the G-actin concentration progressively increased (Figure 2). In contrast, AET₂ exposure to phalloidin, which is known to stabilize F-actin, resulted in a marked G-actin decay with nadir values of 38 ± 7% of baseline after 0.5 h (Figure 3).

View larger version (18K): [in this window] [in a new window]   Figure 1.   Time course of F-actin depletion by C2 toxin. The F-actin content of AET2 was determined at indicated times after incubation with different concentrations of C2 toxin (doses of component I/ component II given). Controls were sham-incubated with solvent only. All data are given as percent of baseline values determined before toxin application. #Significantly different from control (means ± SEM of four independent experiments each).

View larger version (22K): [in this window] [in a new window]   Figure 2.   Opposite change of G- and F-actin in response to C2 toxin. The content of G-actin and F-actin in AET2 was determined at indicated times after incubation with 250/500 ng/ml C2 toxin (component I/II). Controls were sham-incubated with solvent only. All data are given as percent of baseline values determined before toxin application. #Significantly different from control (means ± SEM of three independent experiments each).

View larger version (18K): [in this window] [in a new window]   Figure 3.   Time course of G-actin depletion by phalloidin. The G-actin content of AET2 was determined at indicated times after incubation with different concentrations of phalloidin. Controls were sham-incubated with solvent only. All data are given as percent of baseline values determined before toxin application. #Significantly different from control (means ± SEM of four independent experiments each).

Influence of C₂ Toxin and Phalloidin on Surfactant Secretion in the Absence of Secretagogues

The C₂ toxin-induced F-actin decay was paralleled by a dose-dependent augmentation of baseline epithelial surfactant secretion (Figure 4). At the maximum concentration of C₂ toxin employed (375/750 ng/ml C_2I/II), the diphosphatidyl-choline ([³H]- DPPC) release ranged approximately twofold over control values assessed in the absence of toxin. Phalloidin pretreatment did not affect the low baseline level of AET₂ surfactant secretion (Figure 5); however, it blocked the augmentation of [³H]DPPC release provoked by a subsequent C₂ toxin challenge.

View larger version (17K): [in this window] [in a new window]   Figure 4.   Decrease in F-actin and increase in [3H]PC secretion in response to C2 toxin in the absence of secretagogues. Incubation of AET2 with different concentrations of C2 toxin (component I/II) was performed for 3 h. F-actin data are given in percent of baseline; data of [3H]PC secretion are displayed as the percentage of secreted tracer in relation to the total amount of tracer contained in cells plus medium. #Significantly different from control (means ± SEM of four independent experiments each).

View larger version (10K): [in this window] [in a new window]   Figure 5.   Impact of phalloidin pretreatment on C2 toxin-induced [3H]PC secretion. Pretreatment with 10 µg/ml phalloidin or sham pretreatment was performed for 0.5 h. Next, AET2 were incubated with or without 375/750 ng/ml C2 toxin (component I/II) for 3 h. Controls received neither phalloidin nor C2 toxin. Data of [3H]PC secretion are given as the percentage of secreted tracer in relation to the total amount of tracer contained in cells plus medium. #Significantly different from control (means ± SEM of four independent experiments each).

Influence of C₂ Toxin on Secretagogue and Stretch-induced Surfactant Secretion

As anticipated, AET₂ incubation with adenosine triphosphate (ATP) provoked an augmentation of surfactant secretion to two- to threefold increased values over baseline level (Figure 6). When co-applied with concentrations of C₂ toxin, which itself did not substantially enhance baseline [³H]PC release (125/ 250 ng/ml), the ATP-elicited secretory response was significantly increased. A corresponding, although even more prominent, effect of C₂ toxin pretreatment was noted for stretch- induced surfactant secretion (Figure 7). Stretch alone caused an approximate doubling of baseline [³H]PC release. When applying such stress to epithelial cells pretreated with the per se ineffective C₂ concentration of 125/250 ng/ml, a further more than twofold increase in surfactant secretion was noted. A significant amplifying effect of C₂ toxin pretreatment on the stretch-elicited [³H]PC release was even noted for the very low toxin concentration of 50/100 ng/ml.

View larger version (13K): [in this window] [in a new window]   Figure 6.   Effect of C2 toxin pretreatment on agonist-induced [3H]PC secretion. AET2 were incubated with 125/250 ng/ml C2 toxin or sham-incubated for 3 h. During the last 2 h of this incubation period, co-admixture of ATP (2 µM) or solvent only was performed. Controls received neither C2 toxin nor ATP. The [3H]PC release was determined from the tracer amount in the medium as compared to the total tracer content. All data are given in percent of the [3H]- PC release in the controls. #Significantly different from ATP challenge in the absence of C2 toxin (means ± SEM of four independent experiments each).

View larger version (21K): [in this window] [in a new window]   Figure 7.   Effect of C2 toxin pretreatment on stretch-induced [3H]- PC secretion. AET2 on elastic membranes were pretreated with different concentrations of C2 toxin for 1 h (component I/II given). After exchange of medium, one 20-s stretch was performed, and the [3H]PC secretion was assessed at different times after this stretch. Controls received neither stretch nor C2 toxin. Data represent release of tracer in percent of the total tracer content in cells plus medium. #Significantly different from stretch-induced [3H]PC secretion in the absence of C2 toxin (means ± SEM of four independent experiments each).

Secretagogue-induced Changes of the AET₂ F- and G-Actin Ratio

All chemical secretagogues, ATP, phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA), and terbutaline induced a reversible depolymerization of the epithelial F-actin, as evident in Figure 8. Nadir values of F-actin content were noted within 3 min (ATP) to 10 min (TPA, terbutaline) after secretagogue exposure, with minimum values ranging between 75 and 85% of baseline. Time course of restoration of basal F-actin levels ranged between approximately 30 min (ATP) and greater than or equal to 180 min (TPA, terbutaline).

View larger version (19K): [in this window] [in a new window]   Figure 8.   Change of epithelial F-actin content in response to common secretagogues. AET2 were incubated with ATP (100 µM), TPA (100 nM) or terbutaline (100 µM) for up to 3 h. At different time points after onset of secretagogue exposure, the F-actin content was assessed. All data are given as percent of baseline F-actin values determined before secretagogue application. #Significantly different from control (means ± SEM of four independent experiments each).

Control Experiments

ADP ribosylation of a 43-kilodalton (kD) protein in AET₂ was demonstrated to occur in a dose-dependent manner in response to C₂ toxin. As evident from the autoradiogram in Figure 9, the higher the concentrations of components I and II to which the intact cells were exposed, the less effective was the post-labeling of lysed cells with [³²P]NAD by component I. These results indicate dose dependency preceding ribosylation of the 43-kD protein with non-labeled ADP in intact epithelial cells exposed to C₂ toxin.

View larger version (76K): [in this window] [in a new window]   View larger version (14K): [in this window] [in a new window]   Figure 9.   PAGE-gel autoradiogram of [32P]ADP-ribosylated AET2 protein. After pretreatment of AET2 with C2 toxin at indicated concentrations for 2 h (component I/II in ng/ml given) in culture dishes, the cells were transferred to test tubes, pelleted, and lysed. Cell lysates were again post-treated with C2 toxin component I (250 ng/ml) in the presence of [32P]NAD for 1 h. Labeled proteins were analyzed by SDS-PAGE and autoradiographed on Kodak X-Omat film. The top panel shows an autoradiogram of the gel. The 43-kD band represents actin. Quantification was performed by means of densitometric scanning at 550 nm (bottom panel ).

In the concentration range used, C₂ toxin incubation evoked only a very moderate, protracted LDH release as a marker of overt cell lysis. Within 60 min, 0.5-1% of total cellular LDH was liberated, both in controls and in AET₂ exposed to the highest toxin concentration currently used, 375/750 ng/ml component I/II.

AET₂ loaded with CFDA as a marker of vitality exhibited fluorochromasia within minutes and showed no decrase in fluorescence activity after application of C₂ toxin in the highest concentration range used, 375/750 ng/ml component I/II.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

C₂ toxin exposure of AET₂ was found to cause a dose-dependent decay of F-actin concomitant with G-actin accumulation in parallel with a marked amplification of the baseline surfactant secretory response. ADP ribosylation of nonmuscle G-actin, which blocks its contribution to polymerization, has been disclosed as the molecular C₂ toxin mechanism in various cell types (12), and the currently performed control experiments with [³²P]NAD labeling of a 43-kD protein demonstrate a corresponding efficacy of the clostridial toxin also in the cultured alveolar epithelial cells. As anticipated from this basic toxin effect in the C₂-treated AET₂, an impressive overall shift of the G- to F-actin ratio occurred with a reduction of the total cellular F-actin content to approximately 20% of baseline values at the highest C_2I/II concentrations used. This decay of F-actin was closely linked with enhanced baseline surfactant secretion in AET₂, and at toxin doses greater than or equal to 250/500 ng/ml, a more than twofold increase in baseline PC release was noted. The assumption that this enhancement of baseline surfactant secretion is directly related to the depolymerization of F-actin is based on: (1) the fact that C₂ is a very specific tool targeting solely the actin system, as detailed above; and (2) the finding that pretreatment with another specific tool addressing the actin system, phalloidin, virtually completely reverses the C₂ effect on baseline PC release. Phalloidin is known to bind highly specifically to actin filaments more tightly than to actin monomers, thereby decreasing the rate of actin depolymerization and shifting the equilibrium between G- and F-actin to the filament side (21, 23). The current in vitro findings thus correspond very nicely with the morphologic observations of enhanced lamellar body membrane fusion and exocytosis in intact rabbit lungs exposed to C₂ toxin in the absence of any other stimulus application (14).

In addition to provoking enhanced baseline PC release, C₂ toxin pre-exposure significantly augmented the surfactant secretory response of the alveolar epithelial cells to both soluble secretagogues and mechanical stretch. This amplification of the response pattern was particularly evident for the biophysical challenge, which provoked greater than twofold enhanced PC liberation in C₂ pretreated as compared with non-pretreated cells. Interestingly, this marked efficacy was elicited by very low botulinum toxin concentrations (50/100 and 125/250 ng/ml), which did not affect the rate of baseline surfactant secretion. Moreover, at the dosage of 50/100 ng/ml C₂, the overall cellular F-actin content was only marginally reduced by approximately 10%. These findings may suggest preferential C₂ toxin effects on a distinct, functionally specific actin pool (see below) as an underlying mechanism of such marked amplification of the mechanical stretch-evoked secretory response. This notion is further supported by the observation that a rapid-onset, transient decrease of the overall AET₂ F-actin content is a consistent part of the epithelial response pattern to all secretagogues investigated, with nadir values of approximately 15- 25% F-actin reduction below baseline. The kinetics of this transient change in actin assembly is compatible with the established time course of PC secretion in response to these secretagogues (1, 24, 25). It is tempting to speculate that a F- to G-actin shift is part of the stretch- and/or receptor-dependent surfactant secretory response of AET₂ as schematically depicted in Figure 10.

View larger version (27K): [in this window] [in a new window]   Figure 10.   Hypothetic role of the G-/F-actin ratio during stretch- and/or receptor-operated surfactant secretion and influence of G-/F-actin modulating agents: (1) control stimulation with limited F-actin G-actin transition and lamellar body exocytosis; (2) amplification of the release response by pharmacological F-actin depolymerization (C2 toxin); and (3) inhibition of the release response by F-actin stabilization (phalloidin).

The most likely explanation of the impressive impact of an altered state of actin assembly on the surfactant secretory process under both baseline conditions and in response to biochemical and biophysical stimuli is based on the notion that the cortical cytoplasm of the alveolar epithelial cells, the area ranging between the apical aspect of the plasma membrane and the lamellar bodies, is enriched in F-actin (5, 26). This subplasmalemmal actin filament system may represent a mechanical hindrance that has to be transiently depolymerized to allow lamellar body membrane fusion and exocytosis (27). In addition, the overall viscoelastic properties of the cytoplasm, also known to be largely dependent on the network of cross-linked actin filaments, may be operative in the limitation of lamellar body membrane fusion under baseline conditions. Moreover, changes in the state of actin assembly may be an integral part of signaling cascades provoked by biochemical and biophysical stimulation, as demonstrated for other cell types. The dependence of receptor kinetics (28), spatial and temporal limitation of ligand-evoked second messenger elevation (21), sensory and response elements involved in mechanotransduction (3, 29) and organization of plasma and nuclear ion channels (30) on the state of actin assembly may be relevant in this context. Further studies are clearly necessary to elucidate the role of actin in the regulation of epithelial surfactant secretion in more detail.

In conclusion, the current study demonstrates that the actin microfilament system plays a major role in the regulation of surfactant secretion in alveolar epithelial cells, occurring via exocytosis of preformed lamellar bodies. Depolymerization of actin was noted to be linked with enhanced secretion both under baseline conditions and in response to stimulation by mechanical stress and secretagogue exposure. Cytoskeletal control of compartmentalization and a role of actin in signaling cascades may be suggested as underlying events.