INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

SURFACTANT is a complex mixture of phospholipids (80-90% wt/wt), neutral lipids (5-10%), and proteins (5-10%) that lines the alveolar surface in a monomolecular film and principally serves to reduce the work of breathing by lowering the surface tension of the alveolus and distal conducting airways (12). Synthesis, secretion, and reuptake of surfactant is controlled primarily by alveolar type II epithelial cells. Phosphatidylcholine (PC) is the most abundant phospholipid (80%), and the fatty acid acyl composition of PC is predominantly saturated palmitic acid (65%). The surfactant-associated apoproteins [surfactant protein (SP) A-D] markedly enhance the adsorption, spreading, and maintenance of the phospholipid monolayer (27, 44).

Disease states that result in a deficiency of surfactant are characterized by severe dysfunction of the lungs, including reduced gas exchange, decreased lung compliance, and increased airway resistance. Premature infants with insufficient surfactant synthesis by type II epithelial cells develop respiratory distress syndrome (RDS) (32). Children and adults can also develop an acute RDS (ARDS) that is similar to RDS but is the result of a severe systemic inflammatory insult (16, 19). Similarly, in asthma, the function of surfactant to maintain the patency of conducting airways is inhibited by inflammation (21, 29). Several mechanisms for inflammation-mediated inhibition of surfactant have been suggested, including 1) direct surfactant injury by hydrolytic enzymes, proteases, and metabolic by-products, 2) disruption of the surfactant monolayer by serum proteins that leak into the alveolus, and 3) reduction in surfactant synthesis by direct injury to type II epithelial cells (28).

One specific mechanism for surfactant injury is hydrolysis of surfactant-associated phospholipids by phospholipases A₂ (22, 25). Intratracheal administration of phospholipases A₂ to adult rats results in severe lung injury and serves as an in vitro model of ARDS (13). Phospholipases A₂ have been categorized into at least 10 groups (I-X) based on amino acid sequence data (10, 11). Many of these enzymes are secreted extracellularly and are commonly referred to as secretory phospholipases A₂. The secretory phospholipases A₂ share many important characteristics including 1) small size (13-18 kDa), 2) stability in acidic pH, 3) millimolar calcium requirement for activity, and 4) minimal preference for any specific fatty acid in the sn-2 position of the phospholipid. The secretory phospholipase A₂ (sPLA₂) groups are differentiated by their structural configuration, including amino acid sequence and the number and location of disulfide bonds. However, the biological activity of the enzymes between groups can differ significantly as subtle differences in the sequences regulate interfacial substrate binding and catalysis (41).

The first mammalian sPLA₂ (group I) has been isolated and purified from porcine, bovine, equine, and human pancreas, with 75-80% homology of the amino acid sequences between the species (39). Although the principal role of the pancreatic sPLA₂ is in digestion, the protein and mRNA for human group I sPLA₂ are also present in human lung (35). The mammalian group I sPLA₂ preferentially hydrolyzes PC and can hydrolyze surfactants in vitro (22, 37). Mammalian group II sPLA₂ was initially isolated and purified from inflamed peritoneal and synovial fluids (6, 20). Expression of group II sPLA₂ has been demonstrated in several inflamed and noninflamed tissues, including the lung (25). The principal roles of the group II sPLA₂ are believed to be in host defense through potent antibacterial effects and in inflammation through signal transduction and generation of arachidonic acid metabolites (36, 42). The group II secretory phospholipases A₂ prefer phosphatidylethanolamine and phosphatidylserine and hydrolyze PC less efficiently than group I sPLA₂ (6, 20). New mammalian low-molecular-weight phospholipases A₂ have been identified, groups V and X, but their principal roles are not fully understood (1, 10).

The bronchoalveolar lavage (BAL) fluid (BALF) from humans with ARDS contains less total phospholipid (including PC and phosphatidylglycerol), higher amounts of lysophosphatidylcholine (lysoPC), and increased sPLA₂ activity (16). Similarly, BALF from asthmatic patients who are challenged with endobronchial antigen instillation demonstrates increased amounts of sPLA₂ activity and 1-palmitoyl-lysoPC (5, 7). In combination, the BALF characteristics of patients with ARDS and asthma strongly support the presence of secretory phospholipases A₂. The group and cellular origins of the secretory phospholipases A₂ responsible for the changes in phospholipid composition are unknown. Examination of BALF from ARDS patients with antibodies to group II and heparin affinity confirms the presence of a group II sPLA₂ and a second, non-group II sPLA₂ (25). Furthermore, the proenzyme of group I sPLA₂ is present in the serum of patients with ARDS (30).

In this study, we examined the capacity of the mammalian group I (pancreatic) and group II (inflammation) enzymes to 1) release free fatty acids, 2) increase the formation of lysophospholipids, and 3) cause surfactant dysfunction. Our results identify significant differences in the hydrolysis of surfactant-associated PC by these enzymes and provide important information in understanding the role of sPLA₂-mediated hydrolysis of surfactant in inflammatory lung diseases.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Phospholipid and lysophospholipid standards were purchased from Avanti Polar Lipids (Alabaster, AL). Arachidonic acid was purchased from Cayman Chemical (Ann Arbor, MI). The 17:0 fatty acid standard was purchased from NuPrep (Elysian, MN). Radiolabeled [2-palmitoyl-9,10-³H(N)]dipalmitoyl-L--phosphatidylcholine ([³H]DPPC) was purchased from DuPont NEN (Boston, MA). COS-1 cells were obtained from the American Type Culture Collection (Manassas, VA). All solvents were purchased from Fisher Scientific (Pittsburgh, PA). Diethylaminoethyl dextran was purchased from Pharmacia (Piscataway, NJ). All additional chemicals were purchased from Sigma (St. Louis, MO), including the following phospholipases: recombinant porcine pancreas (group I), Naja naja (snake venom group I), and Crotalus atrox (snake venom group II).

Surfactants. Natural porcine surfactant (NPS) was isolated from juvenile pigs (10-15 kg) that were euthanized with intravenous Pentothal Sodium. Repetitive saline lavage (60-80 ml/kg) was performed via an endotracheal tube. Within 30 min, cells were centrifuged (600 g for 30 min) from the lavage fluid. A surfactant pellet was isolated from the cell-free supernatant by ultracentrifugation (15,000 g for 60 min). The pellet was resuspended in saline and washed three times with normal saline and repeat ultracentrifugation. After the surfactant pellet was washed, it was resuspended in saline and the phosphorus content was determined with the method of Bartlett (2). The surfactant suspension was separated into aliquots, which were stored at 70°C. Survanta (Ross Laboratories, St. Louis, MO) samples were stored at 0-5°C. Individual aliquots of the surfactants were thawed on the day of each experiment.

Preparation of human group II phospholipases. Recombinant human group II (rhGpII) sPLA₂ was obtained by transfecting COS-1 cells with a pCMV-5 plasmid containing an sPLA₂ gene construct with the method of Wong et al. (43). rhGpII sPLA₂ was collected after 3 days of transfection and partially purified by overnight extraction in 0.18 M H₂SO₄. A second example of group II sPLA₂ was obtained by partial purification of human synovial fluid (HuSF) from the inflamed joints of patients with rheumatoid arthritis by acid extraction. All acid extracts were dialyzed against a buffer (pH = 7.40) containing 0.05 M Tris and 0.05 M NaCl and stored as aliquots at 70°C. Aliquots were thawed and used fresh daily for each experiment. The protein content of the rhGpII and HuSF sPLA₂ was measured with Bradford's Coomassie blue reagent (Pierce, Rockford, IL).

Distribution of [³H]DPPC in surfactant. A stock solution of labeled surfactant was prepared fresh daily. Aliquots of [³H]DPPC (10 µCi/ml in 1:1 toluene-ethanol, 0.1 µCi/incubation) were transferred to microcentrifuge tubes and dried under N₂ gas. NPS or Survanta was added and diluted to 1.0 mg phospholipid/ml in saline buffered (pH = 7.4) with Tris (5.0 mM) and CaCl₂ (5.0 mM) and mixed three times with a Branson sonicator (Heat Systems-Ultrasonics, Farmingdale, NY) with a stepped microtip set at ~40 W for 15 s. Sonicated samples were loaded at the bottom of a discontinuous sucrose gradient (in H₂O): 0 M (3 ml), 0.25 M (11 ml), 0.35 M (11 ml), and 0.60 M (11 ml including sample). The gradient was ultracentrifuged (64,000 g for 60 min at 4°C), and fractions were removed and analyzed for radioactivity and phosphorus. Data are expressed as the percentage of total radioactivity and the absolute phosphorus (in nmol) in each fraction.

Hydrolysis of [³H]oleate-labeled Escherichia coli. [³H]oleate-labeled E. coli was prepared with a modification of the method of Kramer and Pepinsky (26). Labeled E. coli (400 pmol lipid/sample, 0.1 µCi/ml) was then exposed to secretory phospholipases A₂ for 1 h at 37°C. The incubation was terminated by lipid extraction of the phospholipids and fatty acids with the method of Bligh and Dyer (4). The phospholipids and fatty acids were separated by thin-layer chromatography (TLC) with Silica G plates (Analtech, Newark, DE) and a mobile solvent phase of hexane, ethyl ether, and formic acid (90:60:6 by vol). Phospholipid and free fatty acid fractions were visualized on the TLC plates with I₂ vapor and scraped and analyzed for radioactivity with a scintillation counter. Before TLC, each sample was supplemented with unlabeled arachidonic acid (40 µg) to enhance free fatty acid staining by the I₂ vapor. The data are expressed as the percentage of the total radioactivity recovered in the free fatty acid fraction.

Hydrolysis of [³H]DPPC-labeled surfactant. Aliquots (200 µl) of the sonicated, [³H]DPPC-labeled surfactant stock were diluted to a final phospholipid concentration of 0.5 mg/ml with saline containing 5 mM Tris (pH 7.4) and 5 mM CaCl₂ and incubated in the presence or absence of secretory phospholipases A₂ for 2 h at 37°C. Samples were then processed and analyzed as outlined in Hydrolysis of [³H]oleate-labeled Escherichia coli.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

A model was established for measuring hydrolysis of surfactant phospholipids by secretory phospholipases A₂. A trace amount of [³H]DPPC was mixed with surfactant by sonication as described in MATERIALS AND METHODS such that after exposure of the mixture to an sPLA₂, the release of the [³H]palmitate from the sn-2 position of the surfactant phospholipid could be quantitated by TLC analysis of the free fatty acids. To confirm that the [³H]DPPC was homogeneously mixed into the surfactant, the mixture was fractionated over sucrose density gradients, and the colocalization of lipid phosphorus and [³H]DPPC was established (Fig. 1). Free [³H]DPPC in the absence of surfactant migrated in the early gradient fractions (fractions 1-3). However, [³H]DPPC comigrated with the major surfactant phospholipid peak (fractions 5-10) when the two were mixed by sonication before gradient separation. Thus trace-labeled surfactant appears to be a uniformly mixed substrate.

View larger version (26K): [in this window] [in a new window]   View larger version (21K): [in this window] [in a new window]   Fig. 1.   Incorporation of [2-palmitoyl-9,10-3H(N)]dipalmitoyl-L--phosphatidylcholine ([3H]DPPC) in natural porcine surfactant (NPS). NPS was mixed and sonicated with 10 µCi/ml of [3H]DPPC. Samples were loaded at bottom of a discontinuous sucrose gradient. Data are means ± SD.

Labeled surfactant was then subjected to hydrolysis by sPLA₂. Porcine pancreatic sPLA₂, a group I enzyme, readily hydrolyzed the labeled E. coli and surfactant in a dose-dependent manner (Fig. 2). Hydrolysis increased over the range of 10-1,000 U/ml of enzyme. To compare the ability of the various secretory phospholipases A₂ to hydrolyze surfactant-associated phospholipids, we performed E. coli hydrolysis experiments. The hydrolysis of E. coli serves as the standard for comparison. In all experiments, we defined one unit of enzyme activity as the amount required to result in 50% of the maximum E. coli hydrolysis. The group I sPLA₂ also demonstrated significant hydrolysis of the labeled E. coli.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Porcine pancreatic group I secretory phospholipase A2 (sPLA2) hydrolysis of [3H]DPPC-labeled surfactant and [3H]oleate-labeled Escherichia coli. Porcine pancreatic sPLA2 was incubated at 37°C with either [3H]DPPC-labeled surfactant (0.5 mg of phospholipid/ml) for 2 h or [3H]oleate-labeled E. coli (10 µg of phospholipid/ml) for 1 h. Concentration of porcine pancreatic sPLA2 ([Porcine sPLA2]) is expressed in U/ml where 1 unit is equivalent to amount of enzyme required to achieve 50% of maximum E. coli hydrolysis. Phospholipids and fatty acids were separated with thin-layer chromatography (TLC). Data are means ± SD of %total radioactivity in each free fatty acid fraction. * P  0.02 compared with control without group I sPLA2 (data not shown).

rhGpII sPLA₂ was examined for its ability to hydrolyze surfactant phospholipids. The recombinant protein expressed in COS-1 cells demonstrated significant hydrolysis of [³H]oleate-labeled E. coli (Fig. 3) but demonstrated only minimal hydrolysis of [³H]DPPC-labeled surfactant despite use of concentrations of up to 10,000 U/ml. In these experiments, we again defined one unit of enzyme activity as the amount required to result in 50% of the maximum E. coli hydrolysis. These results suggest that the mammalian group II sPLA₂ is significantly less able to hydrolyze [³H]DPPC than the group I sPLA₂. We considered the possibility that the rhGpII sPLA₂ expressed by the COS-1 cells might be an incomplete or altered product and therefore might artificially demonstrate lower rates of surfactant hydrolysis than the native enzyme. To investigate that possibility, we performed identical experiments with the use of HuSF sPLA₂ obtained from inflamed synovial fluid, and the rates of hydrolysis by HuSF for both substrates, [³H]DPPC-labeled surfactant and E. coli, were identical to the rates of hydrolysis with the rhGpII sPLA₂ (data not shown). Despite the use of higher concentrations than were used for the group I porcine pancreas sPLA₂, the hydrolysis of surfactant by both group II enzymes was minimal. To examine the effects of inhibitors (i.e., Clara cell protein) (23), which might potentially be present in our NPS preparations, we studied the activity of both group II enzymes against [³H]DPPC-labeled Survanta. The rates of hydrolysis in the Survanta preparations were identical to those in the labeled NPS preparations (data not shown).

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Recombinant human group II (rhGpII) sPLA2 hydrolysis of [3H]DPPC-labeled surfactant and [3H]oleate-labeled E. coli. rhGpII sPLA2 was incubated at 37°C with either [3H]DPPC-labeled surfactant (0.5 mg phospholipid/ml) for 2 h or [3H]oleate-labeled E. coli (10 µg phospholipid/ml) for 1 h. Concentration of rhGpII sPLA2 ([rhGpII sPLA2]) is expressed in U/ml where 1 unit is equivalent to amount of enzyme required to achieve 50% of maximum E. coli hydrolysis. Phospholipids and fatty acids were separated with TLC. Data are means ± SD of %total radioactivity in each free fatty acid fraction. Significant difference compared with control without rhGpII sPLA2 (data not shown): * P < 0.01;  P < 0.05.

The hydrolysis of surfactant phospholipids by secretory phospholipases A₂ results in the production of a lysophospholipid and a free fatty acid. Consequently, the group I porcine pancreatic sPLA₂ resulted in an increase in the lysoPC content and a decrease in the PC content (Fig. 4) over the same range as our [³H]DPPC-labeled surfactant experiments. The formation of lysoPC after exposure of surfactant to the maximum concentration of rhGpII sPLA₂ (10,000 U/ml) was low and comparable to the levels of [³H]palmitate released as demonstrated in Fig. 3 (data not shown).

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Alterations in surfactant phospholipids by porcine pancreatic group I sPLA2. NPS (0.5 mg of phospholipid/ml) was incubated in presence and absence of porcine pancreatic sPLA2 (0-1,000 U/ml) at 37°C for 2 h. Phospholipids were extracted and isolated with TLC, and each fraction was analyzed for phosphorus content. Data are means ± SD of %total phosphorus in phosphatidylcholine (PC) and lysophosphatidylcholine (LPC) fractions. * P  0.001 compared with control without group I sPLA2.

The profile of fatty acids released from surfactant was also directly determined after hydrolysis of unlabeled surfactant with the porcine pancreatic group I sPLA₂ (1,000 U/ml). As expected, palmitic acid was the predominant free fatty acid released, accounting for ~59% of the total (Fig. 5). In addition, a broad profile of released free fatty acids was seen, including myristic (14:0), palmitoleic (16:1), oleic (18:1), and linoleic (18:2). This profile of fatty acids released after hydrolysis of our juvenile NPS is comparable to the relative percentages estimated for the fatty acid compositions in the sn-2 position reported for calf surfactant PC (24). However, small differences between our NPS and calf PC do exist, including a smaller 16:1 fraction and a larger 18:2 fraction. These differences likely reflect differences in the composition of PC between the two species, and sn-2 fatty acids are released as a result of hydrolysis of other non-PC phospholipids (i.e., phosphatidylglycerol and phosphatidylethanolamine). A similar but smaller release of palmitic acid (P = 0.13) was seen in experiments in which the maximum concentration of rhGpII sPLA₂ (10,000 U/ml) was used. This result is consistent with the lack of specificity for individual fatty acids in the sn-2 position typically demonstrated by group I and group II sPLA₂, which contrasts with the preference of the high-molecular-weight cytosolic PLA₂ for arachidonic acid (8).

View larger version (30K): [in this window] [in a new window]   Fig. 5.   Fatty acids released from surfactant phospholipids by porcine pancreatic group I sPLA2. NPS (0.5 mg of phospholipid/ml) was incubated in presence and absence of porcine pancreatic sPLA2 at 37°C for 2 h. Free fatty acids were separated with TLC, HCl methylated, extracted, and measured with gas chromatography with an internal standard. Data are means ± SD of concentrations of each fatty acid. * P  0.01 compared with control without group I sPLA2.

The impact of sPLA₂-mediated hydrolysis of PC on surfactant function is demonstrated in Fig. 6. Increase in concentrations of the group I porcine pancreatic enzyme (100-1,000 U/ml) results in an increase in the _min, which reflects a decline in surfactant function. Although the hydrolysis of PC increases steadily over this dose range, the change in  does not. Despite a further increase in hydrolysis between 500 and 1,000 U/ml, _min does not increase further and appears to reach a plateau. This result suggests that the relationship between hydrolysis of PC and surfactant function may depend on a critical "threshold" of degradation that limits optimal packing of the phospholipid monolayer at the air-liquid interface of the bubble. In contrast, exposure of labeled NPS to the maximum concentration of rhGpII sPLA₂ (10,000 U/ml) did not result in an increase in _min (data not shown).

View larger version (25K): [in this window] [in a new window]   Fig. 6.   Comparison of porcine group I sPLA2-mediated hydrolysis and inhibition of surfactant function. [3H]DPPC-labeled surfactant (0.5 mg of phospholipid/ml) was incubated with porcine pancreatic sPLA2 (0-1,000 U/ml) at 37°C for 2 h in buffered saline with CaCl2 (5 mM). Data are means ± SD of minimum surface tensions recorded over 10 min on a pulsating bubble surfactometer and of %total radioactivity in free fatty acid fraction. Significant difference compared with control without group I sPLA2: * P < 0.001;  P < 0.02.

To further compare the activity of secretory phospholipases A₂ within and between groups I and II, we studied the ability of snake venom secretory phospholipases A₂ to cause surfactant dysfunction (Fig. 7). In hydrolysis experiments, both of the snake venom secretory phospholipases A₂, N. naja (group I) and C. atrox (group II), readily hydrolyze E. coli and labeled NPS, with activity for the [³H]DPPC-labeled NPS similar to that with the porcine pancreatic sPLA₂ based on the comparable hydrolytic rates of E. coli (data not shown). Similarly, both snake venom secretory phospholipases A₂, N. naja (100 U/ml) and C. atrox (100 U/ml), result in significant increases in _min. The ability of the mammalian and snake venom group I enzymes to cause surfactant dysfunction is remarkably similar. In contrast, the group II snake venom causes surfactant dysfunction, whereas the rhGpII sPLA₂ does not.

View larger version (18K): [in this window] [in a new window]   Fig. 7.   Comparison of mammalian and snake venom sPLA2. NPS (0.5 mg phospholipid/ml) was incubated at 37°C for 2 h in buffered saline with CaCl2 (5 mM) with the following sPLA2: group I [porcine pancreatic (100 U/ml) and Naja naja (N. naja; 100 U/ml)] and group II [rhGpII (10,000 U/ml) and Crotalus atrox (C. atrox; 100 U/ml)]. Samples were then analyzed for function with pulsating bubble surfactometer. Data are means ± SD. * P < 0.001 compared with control without sPLA2.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Our data support the hypothesis that secretory phospholipases A₂ may play a critical role in inflammatory lung diseases like ARDS and asthma. The most abundant surfactant-associated phospholipid, PC, is hydrolyzed by the mammalian group I sPLA₂ (Fig. 2). This interaction results in the formation of lysoPC (Fig. 4) and free fatty acids (Fig. 5) and results in decreasing overall surfactant function (Fig. 6). The mammalian group II sPLA₂ hydrolyzes surfactant-associated PC but with significantly less activity than the group I sPLA₂ and does not increase _min. This difference in the mammalian enzymes is not solely explained by the class differences between group I and group II enzymes because the snake venom group II C. atrox resulted in significant PC hydrolysis and caused surfactant dysfunction.

Surprisingly, the group II sPLA₂ rhGpII did not demonstrate significant hydrolysis of PC. The literature suggests that the group II enzyme would be the most likely candidate enzyme because of its release by phagocytes, including neutrophils and alveolar macrophages, in inflammatory states (3, 40). The increased levels of sPLA₂ activity in patients with septic shock have been attributed to the release of group II enzymes (17). In addition, a BAL study (25) of patients with ARDS demonstrates an increase in the activity of sPLA₂, which coelutes with purified group II standards on heparin column fractionation (25). However, this same BAL study also demonstrated an additional heparin column fraction, with significant sPLA₂ activity from an unclear group or source. From our data, the second sPLA₂ could be group I because its affinity for PC is substantially greater than that of the group II sPLA₂ based on the relative activity for each enzyme toward E. coli as a substrate. The role for group I enzymes is supported by the propensity for ARDS in patients with pancreatitis and by models of pancreatitis where high levels of serum sPLA₂ activity have been demonstrated (34). Although pancreatitis is an uncommon cause of ARDS and the pancreas is an unlikely source for increased sPLA₂ activity in patients with asthma or in the majority of ARDS patients, there are additional sources of group I sPLA₂. Expression of group I sPLA₂ mRNA has been reported from nonpancreatic tissues including the human lung, and activated human granulocytes release the group I proenzyme (31, 35). Our data favor the group I sPLA₂ as a more likely candidate enzyme than the group II sPLA₂ for causing surfactant damage in ARDS and asthma.

In addition, recently identified secretory phospholipases A₂ might also contribute to surfactant damage. A group V enzyme has been identified from a human cDNA library; it may be present in lung and appears to have a greater affinity for hydrolysis of PC than the mammalian group II enzyme (11). In addition, mRNA for a group X enzyme has been isolated from human lung tissue, but activity of this enzyme against PC or other surfactant-associated phospholipids is unknown (10). Studies similar to those reported here are needed to define the role for those secretory phospholipases A₂ in surfactant dysfunction.

There are variations and uncertainties in the functional activity of native and recombinant group II enzymes that make it impossible to conclude that all group II enzymes cannot contribute to the hydrolysis of surfactant-associated PC. Both group II secretory phospholipases A₂ (rhGpII and HuSF) demonstrated excellent activity on the bacterial substrate E. coli, which suggests that the intact protein with proper folding was present. Various compounds can serve to modify activation of secretory phospholipases A₂, including the inhibitory low-molecular-weight Clara cell protein (23), and could therefore impact our results if present. The similarity in our results for the hydrolysis of NPS and Survanta, which is commercially purified and prepared, suggests that contaminating inhibitors like the Clara cell protein were not a factor in our experiments. In addition, our studies do not exclude that higher rates of hydrolysis for the labeled surfactant might not be seen if higher concentrations of rhGpII sPLA₂ (>10,000 U/ml) were used. Our comparisons of group I and II activities are based on the relative activity of each toward E. coli as a substrate and not the absolute levels of enzyme in surfactant. If levels of group II sPLA₂ within the inflamed alveolar microenvironment in ARDS or asthma increased beyond those used in this study, significant hydrolysis could occur.

There are also important variables in the physical state of the substrate of sPLA₂-mediated hydrolysis that may have contributed to our results. The group I secretory phospholipases A₂ have a higher affinity for PC than group II secretory phospholipases A₂ and might be expected to demonstrate the results we have shown. Group II secretory phospholipases A₂ preferentially hydrolyze phosphatidylethanolamine and phosphatidylglycerol as substrates (20, 38). We did not examine the ability of the group I or II enzymes to hydrolyze surfactant-associated phospholipids other than PC. In our experimental incubations, the air-liquid interface at the top of the microcentrifuge tube was small. As a result, the greatest percentage of the phospholipid was not in a monomolecular film but more likely in vesicles. We cannot exclude the possibility that the group II sPLA₂ could more readily hydrolyze surfactant-associated PC in a monomolecular film as found in the alveolus. Another potential variable within the surfactant mixture that could lead to altered presentation of the substrate to the enzyme is the SPs. The similar activity of group I and II secretory phospholipases A₂ on NPS and Survanta suggests that SP-A and -D, which are not present in Survanta, play little or no role in determining enzymatic activity. Low levels of SP-B and -C are present in Survanta, and, therefore, we cannot evaluate their role in regulation of sPLA₂-mediated hydrolysis.

The mechanism of phospholipase-induced dysfunction of surfactant is likely to be multifactorial. Because the enzymatic reaction results in a reduction in the native phospholipids and an increase in lysophospholipids and free fatty acids, both have been shown to inhibit in vitro surface tension lowering of surfactant activity (9, 18). Any one of these changes in the surrounding milieu of the alveolar monolayer and subphase could potentially interfere with the ability of the surfactant layer to pack tightly, associate with the SPs, and subsequently lower . In all likelihood, it is a combination of these events that explains the subsequent dysfunction, and clarification of this intricate relationship warrants further investigation. From the data in Fig. 6, it is intriguing to note the sharp contrast in function shown between 100 and 500 U/ml of the group I pancreatic enzyme despite hydrolysis steadily increasing within the same concentration range. These data suggest that there may be a critical concentration or relationship between the ratio of intact phospholipids, lysophospholipids, and fatty acids that determines the function of the surfactant film.

In summary, our data demonstrate that there are significant differences between the mammalian group I and group II secretory phospholipases A₂ and their ability to hydrolyze surfactant-associated PC and lead to surfactant dysfunction. These findings have important implications in the role of sPLA₂ in inflammatory lung conditions like ARDS and asthma.