Main

Inhibitors of pulmonary surfactant contribute to the pathophysiology of lung injuries manifested clinically as ARDS(1,2). In ARDS, inflammation gives rise to phospholipases, proteases, and other mediators within lung tissue. Damage to the alveolocapillary membrane allows these compounds, along with cellular degradation products and blood-derived lipids and protein, access to the alveoli where they can impair the surface-active function of lung surfactant. Plasma protein, Hb, cell membrane lipids, FFA, and many other blood- and tissue-derived compounds have been found in bronchoalveolar lavage from animals or humans with ARDS-related acute lung injury (3–8 for review). The ability of many of these substances to inhibit the adsorption and/or dynamic surface activity of pulmonary surfactant in vitro is well-documented(9–19), and their presence in alveoli correlates with impaired surfactant function in lung-injured animals(20–24) and in human patients with ARDS(25–27).

In contrast to lytic enzymes and act chemically to degrade pulmonary surfactant components, cellular lipids and blood-derived proteins typically impair lung surfactant activity through biophysical interactions. Several biophysical mechanisms of action appear to be involved(13,18). Blood proteins, cell membrane lipids, and FFA all have intrinsic surface activity. These substances adsorb to the air-water interface and form surface films just as do the components of pulmonary surfactant. Competitive adsorption from large proteins like albumin and Hb has been shown to interfere significantly with the entry of lung surfactant constituents into the interface(13). FFA also adsorb readily, but unlike large proteins, these small molecules tend to interpenetrate and form mixed films with lung surfactant phospholipids. When the content of fluid unsaturated fatty acids like oleic acid rises to a sufficient level in the surface film, its ability to reach low surface tensions during subsequent dynamic compression is compromised18. Cell membrane lipids, including lysophospholipids, have also been shown to impair the surface activity of pulmonary surfactant(11,15), but detailed mechanistic studies have not been done. This is the focus of the current paper, which compares the mechanisms and characteristics of LPC-induced and albumin-induced inhibition of CLSE.

Inhibition mechanisms for LPC and albumin are examined by several interfacial methods. Dynamic surface studies used a pulsating bubble surfactometer with a specially designed sample chamber and associated fluid delivery system for exchanging the hypophase(28,29). This hypophase exchange system allowed the effects of LPC and albumin on dynamic surface tension lowering in CLSE films to be assessed during rapid cycling at physiologically relevant rates without the influence of continued adsorption. Experiments in a separate absorption apparatus with diffusion minimized by stirring then examined specific inhibitor effects involving this surface behavior. In addition, Whilhelmy balance studies were also done on mixed films of DPPC and LPC to help define their miscibility and interactions at the interface. Results from all these methodologies indicated that a major part of LPC-induced inhibition of lung surfactant involved interactions within the surface film during dynamic compression, whereas albumin acted primarily by competing with lung surfactant during adsorption.

MATERIALS AND METHODS

CLSE. CLSE was prepared by organic solvent extraction from endogenous surfactant lavaged with saline from the lungs of freshly killed calves (Conti Packing Co., Henrietta, NY) (e.g.30,31). Lavage fluid was immediately centrifuged at 250 × g for 10 min to remove a small cellular pellet. The supernatant was centrifuged at 12,000 × g for 30 min to sediment whole surfactant, and CLSE was extracted into chloroform:methanol by the method of Bligh and Dyer(32). CLSE contained approximately 93% phospholipid and 1.3% protein by spectrophotometric assay(33,34). Phosphatidylcholines made up >80% of phospholipids by thin layer chromatography(35), and a spectrum of minor phospholipid classes was present as reported previously(30,31,36). CLSE was stored in chloroform at -20°C.

Other surfactants and inhibitors. DPPC was obtained from Avanti Polar Lipids (Alabaster, AL), and LPC and albumin were purchased from the Sigma Chemical Co. (St. Louis, MO). LPC was reagent grade, >99% pure as supplied. Albumin was bovine albumin, essentially fatty acid free, prepared from fraction V albumin.

Combination of CLSE with LPC or albumin. Stock suspensions of CLSE (30 mg phospholipid/mL) in saline buffer (150 mM NaCl, 10 mM HEPES, 5 mM CaCl2 at pH 7.0) were prepared and stored under sterile conditions. Final surfactant dispersions for oscillating bubble and adsorption studies were obtained by diluting the stock dispersion with the same buffer to the desired surfactant concentration. LPC and albumin dispersed in buffer were added in known amounts into these working solutions. Composition in CLSE:inhibitor mixtures was expressed as content in percent by weight, with surfactant and inhibitor concentrations given as mg/mL.

Pulsating bubble apparatus. a pulsating bubble surfactometer (Electronetics, Amherst, NY)(37) was used to measure the overall surface tension-lowering ability of CLSE in the presence and absence of inhibitors. Measurements with this instrument reflect the combined effects of adsorption and dynamic film compression and were performed as a function of surfactant and inhibitor concentrations at 37°C. A small air bubble, communicating with ambient air was formed in an aqueous dispersion of CLSE (and inhibitors) held in a small plastic sample chamber. The bubble was monitored through a microscope and was pulsated in size between minimum and maximum radii of 0.4 and 0.55 mm, respectively, by a precision pulsator that moved liquid in and out of the sample chamber. The pressure drop across the air-liquid interface (ΔP) was measured with a transducer, and surface tension (γ) was calculated from the Laplace equation: Δp = 2γ/radius. Surface tension at maximum bubble radius (minimum surface tension, γmin) was recorded during continuous cycling (20 cycles/min) as a primary measure of overall surface tension-lowering ability. Calculations were not corrected for bubble deviations from sphericity at low surface tension, because even severe deformations results in small absolute errors in calculated surface tension on this apparatus(38).

Hypophase exchange studies. A custom hypophase exchange system was used with the pulsating bubble apparatus to allow mechanistic experiments on adsorbed surfactant films. This system incorporated two stainless steel tubes (22 gauge) that formed channels leading to the lumen of a specially designed sample chamber(28,29). One of these, the inlet channel, opened at the top of the chamber near the site where the bubble was ultimately pulsated, whereas the other channel was a drainage from the lower part of the chamber. The system allowed hypophase exchange at a constant bubble size by regulating the inflow and outflow rates to be identical. To accomplish this, the inlet and outlet channels were connected to a continuous silastic tube divided by a migrating barrier (a spring-loaded motor-driven compressing wheel) and enlarged one section of the tube at the expense of the other while total volume was constant. The two valves connecting the silastic tubing to the sample chamber opened and closed concurrently. At the start of the exchange process, the valves were closed so that the inlet tube communicated at one end with a syringe containing the liquid to form the new hypophase and the other end to drainage. The compressing wheel was then disengaged, moved to a position of maximum volume for the inlet tube, and re-engaged after an external water level adjustment to equalize tube pressures on the two sides of the barrier. The valves were then opened, and the new hypophase passed through the bubble chamber in a volume 10 times that of the original hypophase, while the bubble and surfactant film were maintained at equilibrium. The bubble was monitored continuously through a microscope during the exchange to verify constant size. Hypophase exchange allowed films to be formed by adsorption from a surfactant-containing subphase and then studied on a new (exchanged) subphase containing inhibitor molecules rather than surfactant(28,29). A CLSE film with a γmin < 1 mN/m was initially formed during bubble pulsation for 5 min in a dispersion of concentration 5 mg CLSE phospholipid/mL. This film was isolated by an initial hypophase exchange with buffered saline (150 mM NaCl, 10 mM HEPES, 5 mM CaCl2 at pH 7.0), followed by dynamic cycling for 5 min to verify a stable film with γmin < 1 mN/m. A second exchange with a hypophase containing LPC, albumin, or buffer (control) was then done, and γmin was again measured during dynamic cycling for 5 min to define the effects of inhibitors.

Adsorption studies. Adsorption was measured at 37°C in a Teflon dish containing a 70-mL buffered saline subphase (150 mM NaCl, 10 mM HEPES, 5 mM CaCl2 at pH 7.0) that was stirred continuously with a Teflon-coated magnetic bar and magnetic stirrer to minimize the mass transfer resistance from diffusion30. Surface pressure (π), defined as the decrease in surface tension below that of the pure subphase at the experimental temperature, was measured as a function of time with a hanging platinum slide (see Wilhelmy balance methods). Adsorption experiments were initiated at time zero by injecting surfactant or surfactant:inhibitor mixtures dispersed in 10 mL of identical buffer into the stirred 70 mL subphase.

Wilhelmy balance. Interfacial films spread from organic solvent (9:1 v/v hexane:ethanol) were studied in a custom-designed Wilhelmy balance with a continuous Teflon ribbon barrier to minimize leakage at low surface tension(39). The balance trough containing the subphase and surface film was milled from a solid block of Teflon and was supported by a rigid aluminum frame. The balance compression ratio (maximum area:minimum area) during cycling was 4.36:1, and the subphase volume was 800 mL. Wilhelmy balance results were expressed as surface pressure-surface area (π-A) isotherms during cycling at either 23°C or 37°C. Respreading was assessed by comparing calculated π-A isotherm areas between designated compression curves during continuous cycling(8). Surface pressure was measured by a force transducer attached to a sandblasted 1 × 2 cm platinum slide dipped into the air-water interface, and the area A (Å2/molecule) was determined from known balance areas and the amount of surfactant initially spread. A 10-min pause after initial film spreading was allowed for solvent evaporation before cycling at a rate of 1.5 or 10 min/complete cycle (0.75 or 5 min/compression).

RESULTS

Both LPC and albumin impaired the ability of CLSE to lower surface tension under dynamic cycling on the pulsating bubble apparatus, but the overall pattern of their inhibitory effects was very different (Figs. 1 and 2). In the absence of inhibitors, control dispersions of CLSE had γmin < 1 mN/m at all concentrations investigated (1, 3, or 6 mg surfactant phospholipid/mL). Mixtures of CLSE and LPC and increased by γmin as function of LPC-content: 14 wt % LPC raised γmin to 5 mN/m, 25-50 wt % LPC raised γmin to 15 mN/m, and 67 wt% LPC raised γmin to >20 mN/m, even at high CLSE concentrations, 3 and 6 mg/mL (Fig. 1). In contrast, the inhibitory effects of albumin on CLSE were not stoichiometric and were reversed by high surfactant concentration (Fig. 2). Albumin (3 mg/mL) raised γmin to >20 mN/m at a low CLSE concentration of 1 mg/mL, but inhibition was abolished at a CLSE concentration of 3 mg/mL even when 8-fold larger amounts of albumin were present (Fig. 2).

Figure 1
figure 1

Inhibitory effect of LPC on the minimum surface tension of CLSE in the pulsating surfactometer. (A) minimum surface tension as a function of LPC concentration in dispersion; (B) minimum surface tension as a function of % LPC content (weight percent of LPC in the total mixture with CLSE). Minimum surface tension is after 5 min of cycling on the pulsating bubble surfactometer (37°C, 20 cycles/min) for CLSE and LPC dispersed in buffer at the concentration or content indicated. Data are mean ± SEM for n = 4-5.

Full size image
Figure 2
figure 2

Inhibitory effect of albumin on the minimum surface tension of CLSE in the pulsating bubble surfactometer. (A) minimum surface tension as a function of albumin concentration in dispersion; (B) minimum surface tension as a function of % albumin content (percent by weight of albumin in the total mixture with CLSE). Other details as in legend to Figure 1. Data are mean ± SEM for n = 4-5.

Full size image

In hypophase exchange studies in the bubble apparatus, LPC was able to penetrate surface films of CLSE and inhibit surface tension lowering during dynamic compression (Fig. 3). An interfacial film of CLSE was initially formed during pulsation of an air bubble in a CLSE dispersion held in a specially designed sample chamber (see "Hypophase exchange methods"). This CLSE film had a stable γmin < 1 mN/m within 5 min of dynamic cycling at 20 cycles/min (Fig. 3, Pre-Exchange). An initial hypophase exchange with saline was performed to isolate the CLSE film, and the bubble was pulsated again to verify low γmin during dynamic cycling (Fig. 3, 1st Exchange). A second exchange with a hypophase containing LPC at 3 or 6 mg/mL was then done, and γmin was raised to ∼10 mN/m after 5 min of cycling in the presence of this inhibitor (Fig. 3, 2nd Exchange). In control experiments, the isolated CLSE film subjected to a second hypophase exchange with saline still exhibited γmin < 1 mN/m (Fig. 3).

Figure 3
figure 3

Inhibitory effect of LPC on minimum surface tension in performed lung surfactant films under dynamic conditions. A specialized hypophase exchange system ("Methods") isolated a CLSE film with stable γmin < 1 mN/m (1st Exchange), and the ability of LPC in hypophase concentrations of 3 and 6 mg/mL to penetrate this film and raise γmin is shown (2nd Exchange). See text for details.

Full size image

In analogous hypophase exchange experiments (Fig. 4). albumin was not able to penetrate preformed CLSE films. Interfacial films of CLSE with a stable γmin < 1 mN/m after 5 min of initial cycling were again isolated (Fig. 4, 1st Exchange). However, a second exchange with a hypophase containing albumin at 3 or 6 mg/mL did not lead to increased γmin during 5 min of dynamic cycling (Fig. 4, 2nd Exchange). Instead, CLSE films cycled on albumin-containing subphases had the same γmin < 1 mN/m exhibited by control CLSE films cycled on an exchanged subphase of buffer alone (Fig. 4).

Figure 4
figure 4

Inhibitory effect of albumin on minimum surface tension in preformed lung surfactant films under dynamic conditions. Hypophase exchange studies are analogous to those in Figure 3. A CLSE film isolated by hypophase exchange retains γmin < 1 mN/m when cycled on a hypophase containing 3 or 6 mg albumin/mL (2nd Exchange). Sex text for details.

Full size image

To supplement the dynamic surface activity results described above, additional experiments examined the interactions of LPC and albumin with CLSE during absorption. Both CLSE and LPC adsorbed rapidly when dispersed in buffer and injected beneath the surface of a stirred subphase at 37°C. CLSE alone adsorbed to a high equilibrium surface pressure of 47-48 mN/m, whereas LPC adsorbed to a lower but still significant equilibrium surface pressure of 38 mN/m (Fig. 5). The surface pressure-time adsorption isotherm for pure LPC changed by very little if its subphase concentration was varied from 0.016 to 0.056 mg/mL, all of which are well above its critical micelle concentration of 7 × 10-6 M (0.0035 mg/mL)(40). Mixtures of CLSE:LPC had adsorption isotherms that varied between those of the two pure components in a content-dependent fashion (Fig. 5). As LPC content increased, there was a successive decrease in surface pressure from that of CLSE alone. However, equilibrium surface pressures were still substantial because of the good adsorption exhibited by the lysolipid alone.

Figure 5
figure 5

Inhibitory effect of LPC on the adsorption of CLSE. Surface pressure (amount by which surface tension is lowered below the pure subphase value of 70 mN/m at 37°C) is plotted as a function of time during adsorption. CLSE or LPC alone, or CLSE plus LPC at the weight percents indicated, were combined and injected at time zero beneath the surface of a well-stirred buffered subphase ("Methods"). Final subphase concentration of CLSE was constant at 0.084 mg/mL. Curve for LPC alone is at a final subphase concentration of 0.028 mg LPC/mL. Data are mean ± SEM for n = 3-4 experiments.

Full size image

Albumin exhibited several differences from LPC in its inhibitory effects on CLSE during adsorption. When injected beneath the surface of a stirred subphase, albumin adsorbed to an equilibrium surface pressure of only 18-19 mN/m (Fig. 6), much less than that of CLSE or LPC. When albumin and CLSE were combined and injected together, there was a substantial reduction in both adsorption rate and final equilibrium surface pressure compared with CLSE alone (Fig. 6). At the low surfactant concentration of 0.084 mg phospholipid/mL shown, albumin gave a much larger reduction than LPC in the equilibrium surface pressure of CLSE (Fig. 6 vs. Fig. 5). However, consistent with our previous work (e.g.9,11,13), the inhibitory effects of albumin on CLSE adsorption could be abolished if surfactant concentration was raised. We found that albumin in amounts from 1.25 to 8 mg/mL (contents of 80-96% by weight) had essentially no inhibitory effect on simultaneously adsorbing CLSE at a higher surfactant concentration of 0.3 mg phospholipid/mL (data not shown).

Figure 6
figure 6

Inhibitory effect of albumin on the adsorption of CLSE. CLSE and albumin alone, or combined at the weight percents indicated, were injected beneath the surface of stirred subphase in analogy with Figure 5. Final CLSE subphase concentration was fixed at 0.084 mg/mL. Curve for albumin alone is for a final subphase concentration of 1.25 mg albumin/mL. Data are mean ± SEM for n = 3-4 experiments.

Full size image

The differing characteristics with which LPC and albumin inhibited the surface activity of CLSE were studied further in adsorption experiments examining film penetration (Figs. 7 and 8). LPC was found to be able to adsorb and penetrate CLSE surface films preformed by solvent spreading at the air-water interface (Fig. 7). Injection of LPC beneath a CLSE film preformed at a surface pressure of 20 mN/m resulted in increased surface pressures that were higher than those found for LPC adsorbing alone at a clean surface (Fig. 7A). Conversely, injection of LPC beneath a CLSE film preformed at a high surface pressure of 48 mN/m resulted in decreased surface pressures that exceeded those found for LPC adsorbing alone to a clean surface (Fig. 7B). These results indicated that LPC was able to adsorb and form mixed surface films at an interface originally occupied by CLSE film at high or low surface pressure. (Although LPC could penetrate preformed films of CLSE, the converse was less true. In studies analogous to those in Figure 7, injection of CLSE beneath an LPC film preformed by solvent spreading to a surface pressure of 20 mN/m resulted in a rise in surface pressure to only 26 mN/m after 30 min and to 36 mN/m after 50 min (data not shown). This indicated that although CLSE or some of its components could penetrate an LPC surface film, the process was much slower than for LPC penetration of preformed CLSE films).

Figure 7
figure 7

Effect of LPC adsorption on preformed CLSE surface films. (A) LPC injected beneath a CLSE film preformed at surface pressure 20 mN/m; (B) LPC injected beneath a CLSE film preformed at surface pressure 48 mN/m. CLSE films were preformed by spreading in hexane:ethanol (9:1 v:v) until surface pressure was stable at the desired value for 2 min. LPC dispersed in buffer was then injected into the stirred subphase beneath the film, and surface pressure was monitored as a function of time. Final LPC concentration was 0.028 mg/mL. Data are mean ± SEM, n = 3-4.

Full size image
Figure 8
figure 8

Effect of albumin adsorption on preformed CLSE surface films. (A) albumin injected beneath a CLSE film preformed at surface pressure 20 mN/m. (B) albumin injected beneath a CLSE film preformed at surface pressure 48 mN/m. Details are as in legend to Figure 7, except that the inhibitor was albumin rather than LPC. Final albumin concentration was 1.25 mg/mL. Data are mean ± SEM, n = 3-4.

Full size image

Although LPC could readily adsorb and penetrate into an interfacial film of CLSE, albumin could not. Surface pressures were essentially unchanged when albumin was injected beneath CLSE films preformed at surface pressures of either 20 mN/m or 48 mN/m (Fig. 8). These results contrast markedly with those shown earlier, that albumin could significantly decrease the surface pressures reached by CLSE when the two were allowed to adsorb simultaneously to an initially clean interface (Fig. 8 vs. Fig. 6). The difference is that in the case of the results in Figure 8, albumin cannot penetrate and occupy the interface in significant concentrations, because the surface is already occupied by a CLSE film. Results similar to those shown in Figure 8 have been reported for albumin in our prior work.(13).

The ability of LPC and DPPC, a major phospholipid constituent in CLSE, to mix and interact directly in surface films was also investigated through Wilhelmy surface balance studies of solvent-spread monolayers (Fig. 9). Figure 9A shows pressure-area isotherms for mixed monolayers of LPC and DPPC spread to a dilute initial concentration of 120 Å2/molecule and compressed at 10 min/cycle at 23°C. Additivity plots based on calculations from these data at surface pressures of 5, 10, and 30 mN/m are given in Figure 9B. Ideal molecular areas for noninteracting components are given by the straight lines representing the mol fraction-weighted sum of the pure component areas at each surface pressure. The deviations from additivity shown in Figure 9B indicate that DPPC and LPC form at least partially miscible films in which the components interact to affect surface pressure-area behavior. The ability of LPC to interact in cycled interfacial films with CLSE was also demonstrated at body temperature in Wilhelmy balance studies (Tables 1 and 2). Solvent-spread mixed films of CLSE:LPC had decreased maximum surface pressures (increased minimum surface tensions) compared with CLSE alone (Table 1), and the ability of CLSE:LPC films to respread on successive cycles of compression-expansion was also decreased (Table 2).

Figure 9
figure 9

Variation of mean molecular area in binary films of DPPC and LPC compressed at slow rate in a Wilhelmy balance. (A) surface pressure-area (π-A) isotherms for the first compression of binary films of DPPC:LPC initially spread to 120 Å/molecule and compressed at a rate of 10 min/cycle at 23°C. (B) additivity plots showing actual and ideal molecular areas versus composition in mixed DPPC:LPC films at the three surface pressures indicated in panel A. Solid lines: actual molecular areas at selected surface pressures. Dashed lines: ideal average molecular areas if no molecular interactions. See text for details. Data are mean ± SEM, n = 4.

Full size image
Table 1 Effect of LPC on maximum surface pressure in CLSE films at 37°C
Full size table
Table 2 The effect of LPC on respreading in CLSE films during cycling at 37°C
Full size table

DISCUSSION

This study used complementary interfacial methods to determine how LPC and albumin inhibited the surface activity of lung surfactant in vitro. Mixtures of CLSE:LPC studied during dynamic cycling in a pulsating apparatus had impaired surface tension lowering as a function of LPC content, and this inhibition was present even at high surfactant concentrations up to 6 mg phospholipid/mL (Fig. 1). Albumin inhibited CLSE with a different pattern where surface activity was reduced at low surfactant concentrations, but normal values of γmin < 1 mN/m were found at higher surfactant concentrations (Fig. 2). Several lines of evidence suggested that these different patterns of inhibition were due to LPC and albumin acting, at least in part, through different biophysical mechanisms. In particular, a major part of the inhibitory effects of LPC on the surface activity of lung surfactant came through penetrating and mixing in the interfacial film itself, whereas albumin acted primarily through competitive adsorption and blocking of the air-water interface.

The inhibition studies of this paper were done with CLSE prepared by chloroform:methanol extraction from lavaged alveolar surfactant, rather than with whole surfactant itself. The hydrophilic protein SP-A is removed by the organic solvent extraction step that generates CLSE, and SP-D is not present. However, the mechanistic interpretations of LPC and albumin inhibition based on interactions with surfactant extract films in our work should be directly applicable to lung surfactant in vivo. CLSE contains all of the hydrophobic constituents of alveolar surfactant in the natural ratio and with minimal contamination from cell and tissue-derived materials. These hydrophobic constituents dominate the composition of whole surfactant and the interfacial films that is forms.

Our pulsating bubble results demonstrated a content-dependent inhibitory effect of LPC on the surface tension-lowering ability of CLSE that persisted even at high surfactant concentration: 14 wt % LPC raised γmin to 5 mN/m, 25-50 wt % LPC raised γmin to 15 mN/m, and 67 wt % LPC raised γmin to >20 mN/m at CLSE concentrations as high as 6 mg/mL (Fig. 1). Cockshutt et al.(15) have previously reported that addition of 10% by weight LPC into a lipid extract of bovine lung surfactant increased γmin from ∼4.5 to ∼15 mN/m on a pulsating bubble surfactometer. Holm et al.(28) have also reported that a significant component of surfactant inactivation induced by phospholipase A2 in vitro was related to the effect of LPC on the interfacial film, although fatty acids produced by high phospholipase A2 concentrations also had a detrimental effect on biophysical function(28).

The custom-designed hypophase (subphase) exchange system used in the present study facilitated mechanistic assessments of inhibitor effects. Measurements of γmin on a standard pulsating bubble apparatus reflect contributions from multiple phenomena including dynamic film compression and expansion, respreading, and adsorption. Whereas such measurements provide an extremely useful and physiologically relevant overall activity assessment for lung surfactants(8), mechanistic interpretations are hindered by this phenomenological complexity. Hypophase exchange allowed an interfacial film of lung surfactant to be created by adsorption and cycling on one subphase, and then be isolated and studied during rapid dynamic cycling in the presence of inhibitors introduced in a new subphase. The ability of LPC but not albumin to penetrate interfacial films of CLSE and compromise surface tension lowering at a rapid physiologic cycling rate was directly verified in hypophase exchange experiments here (Figs. 3 and 4). LPC but not albumin was also able to adsorb and alter equilibrium surface pressures in preformed interfacial films of CLSE in the absence of dynamic compression (Figs. 7 and 8). The ability of LPC to incorporate in surface films of DPPC and CLSE and affect surface tension-area behavior during compression was also demonstrated in Wilhelmy balance studies (Fig. 9; Tables 1 and 2). Analogous Wilhelmy balance studies on solvent-spread films of albumin and CLSE or DPPC were not feasible, because albumin is not soluble in hexane:ethanol or similar water-immiscible organic solvents. However, Tabak and Notter(41) have shown that the direct addition of albumin into surface films of DPPC by spreading down a glass rod at the air-water interface results in only minor decreases in maximum surface pressure (increases in minimum surface tension) during subsequent dynamic cycling.

Our results for albumin are consistent with previous studies indicating that it and other blood proteins such as fibrinogen and Hb inhibit the surface activity of lung surfactant primarily through competitive adsorption(13). Because plasma proteins cannot readily penetrate an existing interfacial film of lung surfactant, their inhibitory effects are most substantial if they reach the surface when it is not fully occupied by surfactant molecules. This is the case when plasma proteins adsorb simultaneously with lung surfactant into a clean interface when the subphase surfactant concentration is relatively low. Under such conditions, albumin gave a substantial reduction in lung surfactant adsorption that exceeded the detrimental effects of LPC on this surface property (Figs. 5 and 6). However, as the concentration of surfactant in the subphase increases, plasma proteins are less effective in competing for the interface, mitigating the magnitude of their inhibitory effects (e.g.9–14). Plasma proteins may also impair lung surfactant adsorption by associating with surfactant aggregates in the subphase during lung injury(42), although such associations can be so loose as to be reversed by simple centrifugation(43).

In contrast to albumin, most of the inhibitory action of LPC correlated with its ability to penetrate and fluidize lung surfactant films, compromising the ability to reach low surface tension during dynamic compression (Figs. 1,7,9 and Table 1). Fluid FFA, particularly oleic acid, have also been shown to penetrate and mix with surfactant films to generate similar effects(18). The effects of LPC on lung surfactant activity were not limited, however, to the surface film alone. Like albumin, LPC interfered competitively with lung surfactant during adsorption (Fig. 5), and it may have interacted directly with lung surfactant aggregates in the subphase. The addition of LPC to lung surfactant mixtures has previously been reported to cause the appearance to change from milky to clear, suggesting an alteration in vesicle structure and size(44). It is possible that LPC facilitated the removal of lung surfactant components from the subphase or interface by incorporation into mixed micelles, reflecting a detergent-like action. Nonetheless, the interpretation that the major detrimental effects of LPC come from penetration and interaction in the surface film itself is consistent with the molecular characteristics of this lysolipid.

Amphipathic lipid molecules with hydrophilic and hydrophobic regions are capable of exerting deleterious effects on cellular membranes either by inserting into or perturbing components of the membrane lipids through detergent-like actions(45). Weltzien et al.(46) have suggested that insertion of the wedge-shaped LPC molecule may induce a curvature into a normally planar structure, directly destabilizing the phospholipid bilayer. Lung surfactant phosphatidylcholines, particularly those with disaturated chains, have a cylindrical molecular cross-section that allows them to pack into solid, condensed films at high degrees of compression. The conical or wedge-shaped molecular cross-section of LPC could interfere with the regular packing of such molecules, compromising their ability to reach very low surface tensions characteristic of normal lung surfactant. The amphipathic character of LPC and its conical shape also account for its detergent-like properties, which enable it to form micelles and interact with lung surfactant components and aggregates in the subphase.

Much evidence has been accumulated to show that respiratory failure associated with diseases such as pancreatitis correlates well increased phospholipase A2 activity in bronchoalveolar lavage(47–49). The deleterious effect of phospholipase A2 on lung surfactant function has been proved to be related to one of its major hydrolysis products, LPC(47,50,51). Biochemical analyses of pulmonary surfactant recovered by lung lavage from animal models or patients with acute pancreatitis or ARDS have a decreased content of lung surfactant phosphatidylcholine and an increased level of LPC(47,50,51). Although concentrations of alveolar LPC in ARDS-related lung injury are not known precisely, it is likely that inhibitory levels of lysolipids are generated from the action of phospholipase A2 and other phospholipase on lung surfactant and cell membrane phospholipids in some diseases.

The ability of LPC to penetrate existing lung surfactant films and compromise dynamic surface tension lowering even at high surfactant concentrations heightens its inhibitory potential in vivo. This may also apply to other cell-membrane lipids, inasmuch as this class of compounds has been found to be highly detrimental to lung surfactant activity in other studies(11). The ability of an exogenous surfactant preparation to overcome inactivation as its concentration is raised, despite the continued presence of inhibitor compounds, is crucial for therapeutic applications to ARDS. We have shown here that CLSE is able to overcome inhibition by LPC as well as albumin, although surfactant concentration had to be raised high enough to lower the content of LPC below 14% by weight (Fig. 1). Previous studies have demonstrated that CLSE can also overcome inhibition by red blood cell membrane lipids(11,52), FFA(18,52), and plasma proteins(9,11,13,52).

In conclusion, LPC and albumin differed in the pattern with which they inhibited the surface activity of extracted calf lung surfactant, reflecting different primary mechanisms of action. LPC was found to reduce surface activity in a content-dependent pattern that persisted even at high concentrations of lung surfactant where inhibition by albumin was abolished. LPC had detrimental effects on both the adsorption and dynamic film behavior of lung surfactant, but the latter effects were most pronounced. Prominent in the actions of LPC was its ability to penetrate and interact in interfacial films with surfactant phospholipids, interfering with surface tension lowering during dynamic compression. LPC also interacted with lung surfactant components and aggregates in the subphase during adsorption, reflecting its detergent-like properties. In contrast to LPC, albumin did not penetrate well into preformed CLSE films, and reduced surface activity primarily by competitively inhibiting adsorption. The film-penetrating ability of LPC and its stoichiometric content-dependent pattern of surface activity reduction that persists at high surfactant concentrations increase its potential detrimental effects if it is present in substantial levels in the alveoli during acute lung injury.