|
|
||||||||
Department of Physiology, University of Innsbruck, A-6020 Innsbruck, Austria
| Abstract |
|---|
| Introduction |
|---|
|
|
|---|
One of the most fascinating aspects of the surfactant system is the stabilization of a collection of ~300 million bubbles (alveoli), an inherently unstable structure, by a balanced reduction of the surface tension. In each respiratory unit (the acinus), surface tension must be adequately reduced to prevent its own collapse. This implies that there should be local regulatory mechanisms preventing depletion of the alveolar surface lipid film. A unique feature of natural (compared with industrial) surfactant is its compression dependence on the reduction in surface tension, such that during lung deflation, surface tension approaches 0 dyn/cm2 (note that, according to LaPlace's law, this is an important mechanism to keep small alveoli open). Along with this process, some squeezing out of material occurs. This material can be recycled, i.e., reincorporated into the lipid layer at the air-liquid interface, in two ways. First, it can be recycled within the respiratory acinus by recruitment from the subphase during inspiration. This recycling mechanism is a passive one and does not involve a cellular action. Second, part of "used" surfactant is recycled (via endocytosis) by the type II cell, resulting in a continuous surfactant turnover through this cell type. (Another route of surfactant recycling is through macrophages, which will not be discussed here.) The entire recycling process is slow; the turnover time for surfactant has been estimated at several hours (reviewed in Refs. 6 and 14). From estimates of flux, it was calculated that type II cells must be secreting between 11 and 47% of the lamellar body pool per hour (14). The turnover rate of surfactant is not constant. Consistently, secretion of surfactant by the type II cell is regulated, which has been demonstrated by numerous studies (see below). It is generally assumed that surfactant turnover through the type II cell is primarily governed by factors that control the rate of the exocytotic process of lamellar bodies (this assumption will be discussed below).
Physiologically, the most relevant condition associated with elevated surfactant turnover is probably exercise. It is well established that an increase in tidal volume, achieved by muscle force or mechanical ventilation, increases surfactant secretion. This effect may be considerable; 14-fold rates of secretion were reported in swimming rats (10).
| What makes surfactant secretion so unique? |
|---|
|
|
|---|
1 µm), among the largest secretory vesicles in mammalian cells. Second, vesicle contents are secreted into a small fluid volume termed hypophase, subphase, or alveolar lining fluid (ALF), which covers the epithelial cell layer and separates it from the air space. Its average thickness was estimated to be 200 nm, varying from a few nanometers to almost micrometer sizes (2). Third, the secreted product (surfactant) is lipophilic and therefore poorly soluble in an aqueous environment. This probably results in the accumulation of various macromolecular aggregates of surfactant in the ALF. Notably, circular structures of surfactant in the ALF suggest that, in some cases, vesicle contents from lamellar bodies are secreted into the ALF without significant transformation. The mechanisms of surface lipid film formation are not entirely understood. | A challenge to match demand for surfactant with supply |
|---|
|
|
|---|
Stretch-mediated enhancement of surfactant secretion during exercise prevents a loss of alveolar surfactant. There is a rapid increase of the total amount of phospholipid in the alveolar compartment after onset of swimming in rats, consistent with an increased alveolar surfactant pool during exercise, which recycles from and to the air-liquid interface at an elevated rate. This intra-alveolar surfactant pool is long-lived, since it was sustained for at least 2 h following termination of swimming (10).
Large-breath-induced enhancement of surfactant secretion is accompanied by a dramatic loss (~30%) of lamellar bodies from type II cells within 1 h (9). What this clearly indicates is that lamellar body synthesis does not keep pace with lamellar body loss, at least during initial periods of stimulation. Does this imply that at some point during exercise, the supply of lamellar bodies to the apical plasma membrane becomes rate-limiting for secretion (an idea that would challenge the concept of regulated secretion)? How many lamellar bodies would a Tour de France bicyclist loose on his 6-h ride through the French Alps? Each of his type II cells contains ~150 lamellar bodies, a smaller number than in most other secretory cell types. Considering that a single deep breath is a stimulus for secretion, a regulated exocytotic process comparable to that in neuroendocrine cells would quickly deplete his cellular vesicle pool, assuming that each stimulus would cause the fusion of at least one vesicle (this is a minimum assumption for regulated exocytosis, in which readily releasable vesicles instantaneously fuse in response to a Ca2+ signal). The low number of vesicles but the high frequency of stimulation during strenuous exercise is an apparent problem. Here we summarize currently available data on the mechanisms and regulation of surfactant secretion and discuss how a type II cell could keep the balance between stimulated vesicle fusion and smooth, long-term release of surfactant.
| Surfactant secretion can be stimulated by various intracellular signaling pathways |
|---|
|
|
|---|
It is still largely unknown which pathways are activated in a defined in vivo condition associated with increased surfactant secretion and how these pathways are activated in a physiological setting. As noted above, since a single stretch activates secretion and is associated with a Ca2+ signal, it is reasonable to assume that Ca2+ is the physiological messenger coupling stretch to secretion. Since mechanical stimulation may also elicit inositol 1,4,5-trisphosphate (IP3) generation in type I cells, and since IP3 may permeate to the type II cell via gap junctions, it is conceivable that the type I cell controls stretch-mediated responses of the type II cell in vivo. Importantly, the response to stretch in vivo (i.e., whether the type I cell or the type II cell is the primary target of stretch) should depend on the mechanical properties (distensibility) of each cell type, and these properties have not yet been sufficiently determined. It could also be possible that, in vivo, stretch induces ATP release from lung cells, causing secretion via activation of purinergic receptors and phosphoinositide hydrolysis, resulting in both Ca2+ mobilization and the activation of PKC. Extracellular ATP is one of the most potent stimuli of surfactant secretion in vitro (8), but essentially nothing is known about its origin and role in the alveolus.
| Regulated secretion of surfactant in vitro is a slow and multiphasic process |
|---|
|
|
|---|
A standard technique to study the regulation of surfactant secretion is to measure the accumulation of radiolabeled phospholipids in cell supernatants of isolated alveolar type II cells in primary culture. It is evident that, in these types of experiments, the bath solution is only a vague substitute for the poorly defined ALF, and mechanical forces may be absent. Only very recently have techniques been developed that enable the study of single type II cell function within the isolated lung (1). Naturally, the time course of lipid accumulation in the bath solution is a function of all exocytotic steps, yielding neither information about the site of regulation nor about the rate-limiting steps within the overall process. These in vitro studies revealed that, irrespective of the mode of activation, the appearance of phospholipids in the cell supernatant is a very slow process, which terminates up to several hours after the instance of stimulation (Fig. 1
). Hence, a single stimulus can cause a sustained secretory process in vitro.
|
The implication of these findings is that surfactant secretion consists of at least two phases, a prefusion and a postfusion phase, which are clearly separated in time. Which cellular mechanisms underlie these phases and which are regulated?
| The postfusion phase: fusion pores or related structures limit release in vitro |
|---|
|
|
|---|
|
| The prefusion phase: the interface of constitutive and stimulated exocytosis |
|---|
|
|
|---|
, assuming circular fusion pores) and the number of fusion pores (n)
![]() | (1) |
where F is a yet-undefined factor related to viscoelastic properties of surfactant.
A type II cell monolayer in primary culture always exhibits a certain number of fusions even without stimulation. These are defined as constitutive fusion events. If there is also some amount of constitutive fusion in vivo (which is likely), surfactant delivery to the air-liquid interface could theoretically be efficiently enhanced by increasing the radius of these fusion pores, even without a change in the fusion rate. This effect, if present, does not appear to be sufficient, however, to satisfy the demand of surfactant during stimulation.
Type II cells have developed a fusion machinery allowing them to temporarily increase the fusion rate by manyfold without depleting their stored vesicles within short periods of time. First, type II cells do not exhibit a pool of vesicles in a docked, readily releasable state, at least in vitro (3a, 4, 5). This implies that, unlike in the classically defined condition of regulated exocytosis, type II cells do not respond to a Ca2+ signal (elicited, for instance, by the flash photolysis of caged Ca2+) with immediate and synchronous fusions of many vesicles, but rather with cumulative fusion events starting with highly variable delays of many seconds to minutes (3a). In primary cell cultures, the number of stimulated fusion events per cell is in the range between 0 and ~10. The likelihood of fusion in response to a short Ca2+ signal is highest ~1 min thereafter. The percentage of nonresponders is >50% (3a), however, and if this number is the same in vivo, a single deep breath would activate a fraction of all type II cells only. This could be an economical way to save intracellular lamellar bodies for future stimuli (nonresponsiveness is probably related to the shape and duration of the Ca2+ signal; see below). Importantly, the secretory response is not a simple function of the amplitude of the Ca2+ signal, because fusion occurs in an apparent "all-or-none" fashion at a very low threshold [Ca2+]c of ~320 nM (4). Since such low Ca2+ concentrations should easily be present at cytoplasmic sites remote from Ca2+ channels, it appears unlikely that a local Ca2+ accumulation, as present in the active zone of synaptic terminals, is necessary for fusion in this system. This is consistent with the lack of published data on voltage-gated Ca2+ currents in type II cells.
A Ca2+ signal induced by a single type II cell stretch is similar in shape to ultraviolet-flash-induced Ca2+ release (looking at global cytoplasmic [Ca2+]c, not taking into account possible local Ca2+ concentration gradients). Therefore, one could assume that a single deep breath causes fusion of between zero and a few vesicles, with a delay between seconds and minutes. Since repetitive short Ca2+ signals are difficult to generate experimentally, the long-term fusion behavior in response to repetitive Ca2+ signals (deep breaths) is currently unknown. Wirtz and Dobbs found that repetitive stretch causes only a 15% increase in stimulated secretion versus a single stretch (13). From surfactant turnover rates during exercise (see above), it is reasonable to assume that a continuous fusion activity severalfold higher than basal fusion activity is maintained in vivo. Owing to the greatly varying delays between Ca2+ signals and corresponding fusion events, it is unlikely that there is a close temporal correlation between the breathing phase and the instance of fusion.
Regardless of whether the response to stretch is mediated by Ca2+ mobilization alone or by purinergic stimulation in an autocrine fashion, Ca2+ is definitely the most important second messenger involved in either condition. This is demonstrated by a clear Ca2+ dependence of agonist-induced fusion and, moreover, by a close temporal correlation between fusion activity and the course of the Ca2+ signal. ATP-induced Ca2+ elevations are longer than ultraviolet-flash-induced Ca2+ signals and are the result of both intracellular Ca2+ release and Ca2+ entry (3a). Both phases influence the magnitude and temporal behavior of the fusion response. Accordingly, the amount of ATP-induced fusion is dependent on the magnitude of the integrated Ca2+ signal (3a). Naturally, as noted, there may be a common downstream messenger in all Ca2+-mediated signals, such as calmodulin. The efficacy of this Ca2+ sensor could be modified by various factors, explaining why repetitive signals may be subject to either tachyphylaxis or potentiation. A candidate for potentiation is PKC, which has been shown to sensitize Ca2+-induced fusion by modulating the exocytotic machinery in other cell types and which is also a strong activator of surfactant secretion in type II cells. GTP-binding proteins may also play a role, because GTP
S induces surfactant secretion under "Ca2+-clamped" conditions.
On the basis of in vitro data (3a), we propose a model for vesicle processing and fusion that is consistent with slow (compared with neuroendocrine cell types), continuous secretion and that appears to be perfectly adapted to avoid quick depletion of the cellular vesicle pool (Fig. 3
). We postulate a more-or-less random distribution of vesicles at various stages, such that some but not all cells happen to have vesicles close to fusion (Fig. 3
). Since there is a clear relationship between the duration and amplitude (integral) of a Ca2+ signal and the amount of fusion, it is thought that, depending on the shape of the Ca2+ signal, vesicles are processed to some degree toward fusion competence in each cell. In case of long (or repetitively spiking) Ca2+ signals, many vesicles from various initial stages may undergo fusion (Fig. 3
). On a mechanistic basis, the simplest explanation would be that lamellar bodies are constantly "pushed" toward the apical membrane by mechanical forces. Cytoskeletal structures, however, may act as both barrier and motor by preventing the contact of the vesicle membrane with the plasma membrane. Each Ca2+ signal could activate gelsolin and thereby solate the cortical actomyosin. This idea is strongly supported by the finding that actin depolymerization enhances fusion and that major stimuli of surfactant secretion cause a reduction of F-actin (11). Once a close apposition between membranes is reached, elevated local Ca2+ concentrations would complete the process of fusion. It should be noted here that in vitro fusion assays for isolated lamellar body membranes are still lacking. Therefore, we do not yet know the Ca2+ dependence of the membrane fusion event itself. Cytoplasmic Ca2+ measurements cannot exclude local Ca2+ elevations at the site of fusion, even if they are not detected with common Ca2+ indicators.
|
| Future perspectives |
|---|
|
|
|---|
| Acknowledgments |
|---|
This work was supported by the Austrian Science Foundation (FWF), Grants P12974-MED and P13263-MED, and the Austrian National Bank, Grant 7413.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. Tarran, B. Button, M. Picher, A. M. Paradiso, C. M. Ribeiro, E. R. Lazarowski, L. Zhang, P. L. Collins, R. J. Pickles, J. J. Fredberg, et al. Normal and Cystic Fibrosis Airway Surface Liquid Homeostasis: THE EFFECTS OF PHASIC SHEAR STRESS AND VIRAL INFECTIONS J. Biol. Chem., October 21, 2005; 280(42): 35751 - 35759. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. S. Patel, D. Reigada, C. H. Mitchell, S. R. Bates, S. S. Margulies, and M. Koval Paracrine stimulation of surfactant secretion by extracellular ATP in response to mechanical deformation Am J Physiol Lung Cell Mol Physiol, September 1, 2005; 289(3): L489 - L496. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Murao, A. Shimizu, K. Hosoi, A. Iwagaki, K.-Y. Min, G.-i. Kishima, T. Hanafusa, T. Kubota, M. Kato, H. Yoshida, et al. Cell shrinkage evoked by Ca2+-free solution in rat alveolar type II cells: Ca2+ regulation of Na+-H+ exchange Exp Physiol, March 1, 2005; 90(2): 203 - 213. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Qiao, B. Zhou, J. M. Liebler, X. Li, E. D. Crandall, and Z. Borok Identification of Three Genes of Known Function Expressed by Alveolar Epithelial Type I Cells Am. J. Respir. Cell Mol. Biol., July 1, 2003; 29(1): 98 - 105. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Singer, M. Frick, T. Haller, S. Bernet, M. Ritsch-Marte, and P. Dietl Mechanical Forces Impeding Exocytotic Surfactant Release Revealed by Optical Tweezers Biophys. J., February 1, 2003; 84(2): 1344 - 1351. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |