Published ahead of print on September 4, 2003, doi:10.1165/rcmb.2003-0136OC
© 2004 American Thoracic Society DOI: 10.1165/rcmb.2003-0136OC Apoptosis and Necrosis Induced by Cyclic Mechanical Stretching in Alveolar Type II CellsDepartment of Respiratory Medicine and Critical Care, University of Leipzig, Leipzig, Germany Address correspondence to: Dr. Stefan Hammerschmidt, Medizinische Universitätsklinik I, Pneumologie, Universität Leipzig, Johannisallee 32, 04103 Leipzig, Germany. E-mail: stefan.hammerschmidt{at}t-online.de
Alveolar type II (ATII) cells are exposed to mechanical stretch during breathing and mechanical ventilation. Increased stretch may contribute to lung injury. The influence of three stretching patterns (characterized by frequency [min-1] - increase in surface area [%]: S40-13, S60-13, S40-30) on parameters of apoptosis, necrosis, and membrane integrity of rat ATII cells was compared with that in static cultures. The S40-13 stretching pattern simulated normal breathing. The other patterns were chosen to study increased amplitude and frequency. There were no significant differences between the S40-13 group and static cultures. Lactic acid dehydrogenase (LDH) release and early apoptotic cells were significantly increased in S60-13 and S40-30 in comparison with static cultures (LDH: 0.089 ± 0.014 µg/ml and 0.177 ± 0.050 µg/ml versus 0.050 ± 0.011 µg/ml; early apoptosis: 17 ± 3.5% and 23 ± 3.1% versus 9.7 ± 1.4%) at 24 h. Necrosis was significantly increased only in the S40-30 group (13 ± 2.4% versus 6.1 ± 0.9% in static culture at 24 h). Captopril as well as L-Arginine prevented apoptosis and reduced apoptotic cells to static culture levels in the S40-30 group, but did not influence necrosis and LDH release. Increased mechanical stretch may contribute to lung injury by induction of apoptosis and necrosis in ATII cells. Apoptosis induced by high-amplitude mechanical stretch is prevented by captopril and L-Arginine.
Abbreviations: angiotensin converting enzyme, ACE • acute respiratory distress syndrome, ARDS • alveolar type II cell, ATII cell • Dulbecco's modified Eagle's medium, DMEM • epithelial basement membrane surface area, EBMSA • fetal calf serum, FCS • functional residual capacity, FRC • lactic acid dehydrogenase, LDH • nitric oxide, NO • nitric oxide synthase, NOS • propidium iodide, PI • total lung capacity, TLC
Alveolar type II (ATII) cells produce and secrete pulmonary surfactant and contribute to the regeneration of the alveolar epithelium. These processes contribute to alveolar integrity. Mechanical stimuli, i.e., cyclic stretch, play an increasingly recognized role in the regulation of ATII cell functions. Mechanical stimuli associated with physiologic ventilation have been shown to stimulate ATII cell calcium signaling, phospholipid secretion (1), and surfactant protein expression (2). Beyond these physiologic effects, mechanical stretch exhibits adverse effects to ATII cells. It induces apoptosis (3) and cell membrane stress failure (4) resulting in cell death (5, 6). Increased mechanical forces may occur as a consequence of mechanical ventilation. It has been recognized that mechanical ventilation may induce or aggravate acute lung injury. High tidal volume has been identified to worsen the prognosis of ventilated patients with acute respiratory distress syndrome (ARDS) (7, 8). These clinical data have attracted broad interest in the mechanisms linking increased mechanical forces with acute lung injury. Increased mechanical forces have been shown to induce several injurious processes, including release of prostanoids and cytokines, leading to an activation of the immune system (4, 9–11). Although these sequels of mechanical stress are cellular reactions following mechanically stimulated signaling pathways, mechanical forces may also induce necrotic cell death (5, 6) or activate the apoptotic pathway of cell death (3). This in turn may also stimulate immune responses through the proinflammatory effects of the cytosol released from damaged cells (10, 12). Apoptosis has been shown to represent a major pathway responsible for the disappearance of ATII cells in acute lung injury (13). Acute lung injury during ARDS is characterized by surfactant dysfunction or deficiency (14). Mechanical stretch–induced apoptosis or necrosis of ATII cells may contribute to acute lung injury via disturbances in surfactant secretion and recycling as well as via impairment of the regeneration of the alveolar epithelium. Previous studies using ATII cells in models of mechanical strain described the induction of apoptosis using a nonphysiologic stretching pattern at 3 cycles/min (3) or characterized a single stretching pattern only (15, 16). The effect of different stretching patterns on cell viability was also tested in ATII cells cultured for 5 d assuming that these cells were similar to alveolar type I cells (5, 6). The aim of our study was to characterize the influences of amplitude, frequency, and duration of mechanical stimuli on both apoptosis and necrotic cell death. Pharmacologic compounds with known effects on apoptosis were then evaluated for an effect on mechanical stretch–induced apoptosis.
ATII Cell Preparation ATII cells were isolated according to the method developed by Dobbs and coworkers (17) as described previously (1, 18, 19). In brief: ATII cells were isolated from male Sprague Dawley rats (150–200 g) by elastase digestion and differential adherence on IgG-coated dishes. ATII cells were 89 ± 3.4% pure at the time of plating, as proven by modified Papanicolaou staining.
ATII cells were placed on the central area (
Experimental Protocol
Stretching Pattern
Pure cell culture medium was added to wells without cells cycled for 24 h (S40-30 protocol) and then applied to static ATII cell cultures to exclude the possibility that toxic compounds might leak from the distended membranes. Medium from static membranes served as a control. There was no effect on apoptosis/necrosis of medium from cyclically stretched membranes.
Lactic acid Dehydrogenase Measurement
Determination of Apoptosis and Necrosis
Annexin V Binding and Propidium Iodide Staining
Oligonucleosome Enrichment
Caspase-8 Activity
Influence of Captopril and L-Arginine
Angiotensin II ELISA
Statistics
LDH LDH activity was determined in supernatants of all experimental groups (Figure 1). LDH activity in supernatants of static cell cultures increased slightly with time to a maximum of 0.050 ± 0.011 µg/ml at 24 h. LDH activity in group S40-13 (0.059 ± 0.011 µg/ml at 24 h) was not different from those of static cultures. LDH activity in supernatants of groups S60-13 and S40-30, in contrast, were significantly increased compared with static cultures at 12, 18, and 24 h (S60-13: 0.089 ± 0.014 µg/ml, P < 0.0001 and S40-30: 0.177 ± 0.050 µg/ml, P < 0.0001 at 24 h). Cyclic stretching with the larger amplitude (S40-30) resulted in a significantly greater LDH release compared with all other groups (P = 0.0002 versus the S40-13 and P = 0.031 versus the S60-13 group at 24 h).
Annexin V Binding and PI Staining A representative plot of a flow cytometry analysis of stretched cells for annexin V binding and PI staining is shown in Figure 2. Cells negative for PI and annexin V were regarded as viable cells (Figure 2, lower left quadrant). PI-negative cells, which bind annexin V, were defined as cells early in the apoptotic process (Figure 2, lower right quadrant). PI-positive cells with annexin V binding represent late apoptotic/necrotic cells (Figure 2, upper right quadrant). PI-positive cells without annexin V binding (Figure 2, left upper quadrant) were labeled as necrotic cells.
The plots of Figure 3A demonstrate the changes in the percentage of viable (annexin V–negative and PI-negative) cells over the experimental time. The percentage of viable cells in the S40-13 and S60-13 groups was not significantly different from static cultures. However, a significant decrease in the percentage of viable cells occurs in the S40–30 group (55 ± 12% versus 81 ± 2.4%, P = 0.0002 at 24 h). This decrease may be explained by an early increase in necrotic cells (Figure 3D) as well as a subsequent increase in early apoptotic (Figure 3B) and late apoptotic/necrotic cells (Figure 3C). A significant increase in early apoptotic cells in comparison with static cultures (Figure 3B) was found in the S60–13 group at 24 h (17 ± 3.5% versus 9.7 ± 1.4%, P = 0.037) and in the S40–30 group at t = 18 and 24 h (21 ± 3.2% versus 13 ± 1.7%, P = 0.035 and 23 ± 3.1% versus 9.7 ± 1.4%, P = 0.0044). The percentage of necrotic cells was not different in static cultures compared with cells subjected to cyclic stretch in the S40-13 and S60-13 group (Figure 3D). High amplitude cyclic stretch (S40-30) resulted in an early increase of necrotic cells at 12 h and a second peak at 24 h (18 ± 2.2% versus 6.6 ± 1.1%, P = 0.0002 and 13 ± 2.4% versus 6.1 ± 0.9%, P = 0.0034). In contrast, the number of cells in early apoptosis increased continuously starting at 18 h. Cells in late apoptosis/necrosis increased at 12 h and rose to a maximum at 18 h. These findings indicate that there may be an early phase of necrosis followed by a second phase of cell death, most likely initialized by apoptosis.
Oligonucleosome Enrichment
Caspase-8 Activity Figure 5 summarizes caspase-8 activity measured in cell lysates of cyclically stretched ATII cells and in static controls. A significant difference between static controls and stretched cells was observed in response to the S40-30 stretching pattern (1.17 ± 0.19 versus 2.31 ± 0.43 U/mg protein).
Angiotensin II We were not able to detect any Angiotensin II with the detection system used (Angiotensin ELISA; SpiBio).
Influence of Captopril and L-Arginine
Both compounds also resulted in a considerable increase in viable cells at 24 h of cyclic stretching (77 ± 2.6% and 76 ± 3.8%) almost to the level of static cultures (82 ± 3%).
In this study of the effect of mechanical stimulation on LDH release, apoptotic death, and necrotic death we first chose the stretching pattern to simulate the mechanical stimulation of ATII cells associated with normal breathing in rats. The frequency of 40/min is quiet breathing for the rat, and the increase in (two-dimensional) membrane surface area of 13% roughly corresponds to a lung inflation from FRC to 75% TLC according to literature models (20). Cell culture is always accompanied by some LDH release, e.g., during exocytosis. However, in our experiments we have demonstrated that LDH release in cells cyclically stretched with a pattern that was assumed to be physiologic did not exceed LDH release in cells cultured on identical static membranes. In addition, this stretching pattern did not alter the rate of apoptosis or necrosis in ATII cells in comparison with cells in static culture. A stretching pattern chosen to represent approximately total lung inflation (30% increase in membrane surface area), however, led to an increase in LDH release starting at 12 h. This stretching pattern also resulted in a significant increase in the percentage of apoptotic and necrotic cells. The percentage of viable cells consequently fell with time in an almost linear fashion. It appears that there are two mechanisms of cellular damage recognizable from the time course of LDH release and the appearance of markers of apoptosis or necrosis: Necrosis peaks early at 12 h, with only a moderate increase in the number of apoptotic cells. However, at 18 and 24 h apoptosis increases greatly while the number of necrotic cells is decreasing (the first round of necrotic cells probably now gone). With further time, apoptosis leads to necrosis again increasing the percentage of dead cells (cells in late apoptosis or necrosis). A possible interpretation that would fit our data as well as the model of membrane stress failure (4) is that some cells that cannot achieve sufficient membrane reseal will die by necrosis early on. However, in other cells short-lived stress failure may trigger the apoptotic pathway. An increase in frequency (from 40 to 60/min) alone was clearly not a comparable mechanical insult. Here we observed only insignificant changes in viable cells although at 24 h an increase in early apoptotic cells (measured as annexin V binding) was apparent and LDH release was increased throughout the experimental time in comparison with the S40–13 pattern. A second assay measuring caspase-8 activity, which increases in early apoptosis, failed to detect this small increase in apoptotic cells in the S60–13 group. The difference in the outcome of the two tests may be due to a lower sensitivity of the caspase-8 test or due to apoptosis-independent regulation of this enzyme. Remarkably unaltered was the fraction of necrotic cells at all times in the increased frequency group, again arguing for stretch extent induced membrane stress failure as the major mechanism leading to cellular damage. A cell surface area increase of 30% roughly corresponds with that cellular surface area increase that has been observed in experimental models to occur during lung inflation from residual volume to TLC (20). This extent of cell distension may be reached in high tidal volume ventilation and may also be reached or even exceeded in those parts of the lung remaining open following derecruitment in severe lung injury. The release of intracellular components associated with necrosis of ATII cells may result in an activation of the immune system (10, 12) and further contribute to acute lung injury. However not only necrosis of ATII cells but also apoptosis may induce, maintain or aggravate acute lung injury: ATII cell apoptosis induced by a Fas–activating monoclonal antibody applied intratracheally in mice has been shown to result in acute lung injury (21). ATII cell necrosis and apoptosis appear to be a crucial link between mechanical stress and induction or aggravation of acute lung injury. We have used various methods to quantify ATII cell viability: flow cytometry analysis for annexin V binding and PI staining is frequently used to quantify viable, early apoptotic, late apoptotic/necrotic, and necrotic cells. To strengthen our flow cytometry data we chose two additional methods with predominance for early apoptosis and late apoptosis/necrosis, respectively. Caspase-8 was chosen because this enzyme is proximal in the caspase cascade. We do recognize, though, that caspase-8 may also be influenced by a number of endogenous regulatory proteins (22). Oligonucleosome enrichment in the supernatant may be the result of pre-lysis DNA fragmentation and subsequent release during cell lysis late in the process of apoptosis. However, it may also occur due to post-lytic DNA fragmentation. Therefore the information from the oligonucleosome enrichment assay should parallel the information from the annexin V–positive/PI-positive cell fraction detected by flow cytometry analysis. In this study we demonstrated that the increase in late apoptotic/necrotic cells (Figure 6D), and also the increase in early apoptotic cells (Figure 6C) following high amplitude cyclic stretch, is completely inhibited by two pharmacologic compounds: the ACE inhibitor captopril and the NOS substrate L-Arginine. In contrast, the increased percentage of necrotic cells (Figure 6E) remains unaffected by these compounds. It appears likely that captopril or L-Arginine inhibit the process of stretch-induced apoptosis at some point. Medium constituents among other experimental factors may explain why the absolute percentage of apoptotic cells in our static cultures was higher than that reported in a study using comparable experimental conditions (16). The authors of this study reported 5% apoptotic cells in static cultures. However, in their study apoptosis was detected by measurement of DNA condensation with the nuclear stain Hoechst 33258, which detects late stage apoptosis. In contrast, we have used annexin V binding, which detects apoptosis at an earlier stage. In addition, we have used a considerably lower concentration of FCS (2 versus 10%). Previously we have observed a dependency of the percentage of apoptotic cells in static as well as in stretched cultures on the concentration of FCS in the culture medium (Hammerschmidt and coworkers, unpublished data). As reported in our study, Edwards and colleagues (16) reported a maximum 3-fold increase in the percentage of apoptotic cells in response to cyclic stretch. Thus the relative effect of mechanical stretch on ATII cell apoptosis was comparable in both experimental systems. Another paper reports 40–50% nonviable cells following 1 h cyclic stretching (15/min, 25% increase in surface area) in comparison with 5–10% under static conditions (15). This noticeable difference to our findings and the findings of Edwards and associates (16) may result from a difference in the method used to differentiate between viable and nonviable cells. Ethidium homodimer-1, which is excluded by the intact plasma membrane of viable cells, was used to mark dead cells. This dye may enter the cells during transient, nonlethal membrane damage. Therefore, the number of stained cells may greatly exceed the number of necrotic cells as measured by PI staining in our study. An important constituent of the extracellular environment influencing the rate of ATII cell apoptosis is NO. NO has been shown to exhibit antiapoptotic as well as proapoptotic effects (23) on pulmonary structures. Edwards and coworkers (16) demonstrated first the inhibitory effect of the NO donor S-nitroso-N-acetyl-D,L-penicillamine (SNAP) on mechanical stretch induced apoptosis in ATII cells and proposed a physiologic role for macrophages that might protect ATII cells from apoptosis via NO release. We used L-Arginine, the physiologic substrate of NOS, and found a comparable inhibitory effect on mechanical stretch–induced ATII cell apoptosis. This suggests that NO was released by an NOS present in the cultured ATII cells (24). Endogenous NO has been shown to inhibit myocardial caspase 3 activity (25). Caspase inhibition may therefore represent a possible mechanism of the antiapoptotic activity of NO in our model system. Inducible NOS, which is usually induced by inflammatory cytokines with or without LPS exposure (26) might also be induced by mechanical stimulation either directly or indirectly via resulting inflammatory mechanisms. Alveolar distension due to mechanical ventilation has been related to exhaled NO in several experimental systems (27, 28). Data from patients with acute lung injury from our group suggest nitrite (which results from NO in an aqueous environment) to increase in breath condensate with increasing alveolar distension (29). Together this supports the hypothesis that alveolar distension induces ATII cell apoptosis and simultaneously induces a protective mechanism by the release of NO, a potent inhibitor of ATII cell apoptosis. Among the pharmacologic substances influencing ATII cell apoptosis ACE inhibitors and angiotensin II receptor antagonists have gained some interest. Both groups of compounds inhibit apoptosis for example in the myocardium, in renal tubular cells and in vascular smooth muscle cells (30–32). Both inhibit Fas-ligand–induced apoptosis in ATII cells (33). Similar to Fas-ligand–induced apoptosis, we found that mechanical stress induced apoptosis was inhibited by the ACE inhibitor captopril. Although angiotensin II was not detected in the supernatant of stretched cells, we cannot exclude the role of angiotensin II in stretch-induced apoptosis. Rapid degradation of angiotensin II or elevated intracellular angiotensin II concentrations may also explain our negative findings. The beneficial effects of both NO and ACE inhibition may also be interrelated: The action of ACE inhibitors on the cardiovascular system is partially mediated via inhibition of the enzymatic bradykinin breakdown. Bradykinin has been shown to activate endothelial NOS via phosphorylation in the heart (34) and to evoke NO release from tracheal epithelial cells via activation of constitutive NOS (35). Both interventions of our study, the addition of L-Arginine and of captopril, almost entirely inhibited stretch-induced ATII cell apoptosis but did not influence the percentage of necrotic cells. This further suggests that both compounds most likely act somewhere along the pathway leading to apoptosis, but may be without effect when membrane stress failure exceeds a critical level.
This work was supported by a grant from Deutsche Forschungsgemeinschaft (Ha 3263/1-1). The expert technical assistance of Ms. Konstanze Büttner is greatly acknowledged. Received in original form April 14, 2003 Received in final form August 29, 2003
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
1.5 cm diameter) of fibronectin-coated silicon membranes (Bioflex; Flexcell International, Hillsborough, NC; coated additionally with 150 µM bovine fibronectin for at least 3 h at 4°C) of six-well plates (106 per well with Dulbecco's modified Eagle's medium [DMEM] with 10% fetal calf serum [FCS], 1%wt/vol gentamicin, and 1% glutamine). After 22 h of adherence, medium was replaced by fresh medium containing 2% FCS. These plates were used for experiments. 
P < 0.05 versus static controls and versus the S40-13 and the S60-13 groups.
