The Pulmonary Surfactant System: Biochemical and Clinical Aspects

This article starts with a brief account of the history of research on pulmonary surfactant. We will then discuss the morphological aspects and composition of the pulmonary surfactant system. We describe the hydrophilic surfactant proteins A and D and the hydrophobic surfactant proteins B and C, with focus on the crucial roles of these proteins in the dynamics, metabolism, and functions of pulmonary surfactant. Next we discuss the major disorders of the surfactant system. The final part of the review will be focused on the potentials and complications of surfactant therapy in the treatment of some of these disorders. It is our belief that increased knowledge of the surfactant system and its functions will lead to a more optimal composition of the exogenous surfactants and, perhaps, widen their applicability to treatment of surfactant disorders other than neonatal respiratory distress syndrome. Accepted for publication: 8 July 1996     

Direct and Phagocyte-mediated Lipid Peroxidation of Lung Surfactant by Group B Streptococci

In newborn infants, group B streptococci (GBS) often cause pneumonia, with polymorphonuclear leukocytes (PMN) migrating into the lungs. Because surfactant therapy may be needed in such patients, we evaluated the interaction between GBS or GBS-stimulated PMN and a surfactant preparation (Curosurf) in vitro. The superoxide production of GBS strains or GBS-activated PMN was measured, using the nitroblue tetrazolium (NBT) test and the subsequent lipid peroxidation (LPO) as the content of malondialdehyde (MDA) and 4-hydroxyalkenals (4-HNE). The growth of GBS in surfactant was determined and related to the LPO. Finally, the effect of LPO on surfactant activity, caused by GBS-stimulated PMN, was assessed by measuring dynamic surface tension in a pulsating bubble surfactometer. Curosurf diminished the NBT reduction by both live GBS and GBS-stimulated PMN. Surfactant was peroxidized by reactive oxygen species (ROS) from both GBS and GBS-stimulated PMN in a time-dependent manner. Vitamin E significantly reduced the peroxidation level of surfactant in both cases. Surfactant peroxidation was associated with a reduction in the number of live bacteria. The biophysical activity of Curosurf was impaired by GBS-stimulated PMN, as reflected by increased minimum surface tension during cyclic compression. These findings indicate that Curosurf undergoes LPO by ROS produced by GBS and/or PMN. We speculate that exogenous surfactant preparations should be supplemented with vitamin E or another antioxidant, when given to infants with GBS pneumonia. Accepted for publication: 21 August 2000     

Bactericidal Activity of Alveolar Macrophages is Suppressed by V-ATPase Inhibition

Bafilomycin A₁, a selective inhibitor of V-type H⁺-translocating ATPase (V-ATPase), may be a useful adjunct in cancer chemotherapy (Altan et al. [1998] J Exp Med 187:1583–1598). Therapeutic uses of the enzyme inhibitor need to consider the agent's potential effects on normal (nontumor) cells. This study determined the effects of bafilomycin A₁ on resident alveolar macrophages (mφ). Treatment of alveolar mφ with bafilomycin A₁ (10 μM, 1 h) caused a significant decrement in cytosolic pH. This was accompanied by marked alteration of mφ bactericidal capabilities. The enzyme inhibitor caused a marginal reduction in the phagocytosis of opsonized Staphylococcus aureus and significantly suppressed intracellular killing of the phagocytosed bacteria. In keeping with the effects on intracellular killing, bafilomycin A₁ significantly reduced the production of reactive oxygen species (ROS). On the other hand, cell spreading was enhanced significantly by bafilomycin A₁. Comparable changes in ROS generation and mφ spreading were produced by altering cytosolic pH through changes in extracellular pH (pH_o) in the absence of bafilomycin A₁. These findings suggest that the agent's effects on ROS production and mφ spreading were related to the accompanying changes in cytosolic pH. The enzyme inhibitor also altered mφ morphology, leading to the shortening of microvilli and focal loss of surface ruffles. These morphologic effects differed from those produced by altering cytosolic pH by changes in pH_o. The results demonstrate that V-ATPase activity is an important determinant of mφ functioning and structure. Therapeutic use of V-ATPase inhibitors might be expected to compromise the bactericidal activity of alveolar mφ. Accepted for publication: 20 January 2000     

Lipid Peroxidation of Lung Surfactant by Bacteria

The epithelium of the lung is lined with extracellular pulmonary surfactant. This is the surface that invading bacteria first come into contact with when they enter the alveoli. As bacteria become established and interact with this layer, various characteristics of surfactant may become altered. We studied free radical production by three bacterial species, group B streptococci, Escherichia coli, and Pseudomonas aeruginosa, as well as the effect of two concentrations of lung surfactant (Curosurf at 0.04 and 0.4 mg/ml) on this production estimated by the nitro blue tetrazolium reduction test. We also measured the lipid peroxidation of surfactant at various incubation times (0–20 h), using a LPO-586 test kit. In addition, the effect of vitamin E as an antioxidant in a concentration of 0.5 μM was determined by the lipid peroxidation test. We found that the nitro blue tetrazolium reduction by the three bacterial species and lipid peroxidation of lung surfactant increased with time. Vitamin E reduced the lipid peroxidation of this surfactant. By measuring bacterial growth at various incubation times we showed that lung surfactant was bactericidal to group B streptococcal and E. coli strains and that P. aeruginosa strains were resistant to surfactant. We conclude that bacteria, probably by their production of reactive oxygen species, cause lipid peroxidation of lung surfactant. Accepted for publication: 5 October 1998     

Mediators and Oxygen Radicals in Hyperpnea-Induced Airway Constriction of Guinea Pigs

Leukotrienes (LTs), tachykinins (TKs), and oxygen radicals have been suggested to be important modulating factors for the hyperpnea-induced bronchoconstriction (HIB) of guinea pigs. In this study, we tested the hypothesis that LTs and oxygen radicals modulate HIB by triggering TK release. Eighty-five Hartley guinea pigs were divided into four groups: control, dimethylthiourea (DMTU), FPL 55712, and A63162. DMTU is the scavenger for hydroxyl radical. FPL 55712 is an antagonist of LT receptor, whereas A63162 is an inhibitor of lipoxygenase. Each group was further divided into three subgroups: baseline, hyperpnea, and recovery. Each animal was anesthetized, cannulated, paralyzed, and artificially ventilated. We measured dynamic respiratory compliance (Crs), maximal expiratory flow at 50% total lung capacity (V_max50), and forced expiratory volume in 0.1 s (FEV_0.1) during the baseline and recovery periods. Hyperpnea caused significant decreases in Crs, V_max50, and FEV_0.1, indicating HIB in the control group. Pretreatment with DMTU, FPL 5712, or A63162 attenuated HIB. Plasma substance P (SP) levels increased progressively during the experiment in all groups. However, both FPL 55712 and A63162, but not DMTU, significantly decreased SP levels. Similarly, lung malondialdehyde (MDA) contents increased progressively during the experiment in the control group. Neither FPL 55712 nor A63162 significantly affected the increase. On the contrary, DMTU significantly attenuated the increase in MDA during the recovery period. These results suggest that inhibition of LTs leads to suppression at SP levels and HIB, whereas DMTU attenuates HIB by means of other mechanisms. Accepted for publication: 25 April 2000     

Influence of Phospholipid Composition on the Properties of Reconstituted Surfactants

The influence of phospholipids on the ultrastructure and metabolism of reconstituted surfactants has not been well defined. The aim of this study was to determine if changes in the phospholipid composition of reconstituted surfactants altered their biophysical properties, ultrastructure, and conversion to light subtype by cycling. We prepared various surfactants containing radiolabeled dipalmitoylphosphatidylcholine ([¹⁴C]DPPC). The addition of phosphatidylglycerol (PG) or dipalmitoylphosphatidic acid (PA) to DPPC increased conversion to light subtype. In contrast, the addition of dipalmitoylphosphatidylglycerol (DPPG) to DPPC markedly reduced conversion to light subtype on cycling. DPPC and DPPC+PG produced large liposomes (∼1,000 nm), whereas DPPC+PA or DPPC+DPPG formed multilamellar membranes. Mixtures of DPPC and PA were highly surface active in vitro, whereas the surface activity of DPPC+DPPG was similar to that of DPPC. In conclusion, the ultrastructure, metabolism, and surface active properties of DPPC+PG mixtures were influenced markedly by alterations in the fatty acid composition or polar head group of PG. Accepted for publication: 10 October 1998     

Toxic oxidant species and their impact on the pulmonary surfactant system

In this review the effects of oxidant inhalation on the pulmonary surfactant system of laboratory animals are discussed. Oxidant lung injury is a complex phenomenon with many aspects. Inhaled oxidants interact primarily with the epithelial lining fluid (ELF), a thin layer covering the epithelial cells of the lung which contains surfactant and antioxidants. In the upper airways this layer is thick and contains high levels of antioxidants. Therefore oxidant injury in this area is rare and is more common in the lower airways where the ELF is thin and contains fewer antioxidants. In the ELF oxidants can react with antioxidants or biomolecules, resulting in inactivation of the biomolecules or in the formation of even more reactive agents. Oxidation of extracellular surfactant constituents may impair its function and affect breathing. Oxidized ELF constituents may promote inflammation and edema, which will impair the surfactant system further. Animal species differences in respiratory tract anatomy, ventilatory rate, and antioxidant levels influence susceptibility to oxidants. The oxidant exposure dose dictates injury, subsequent repair processes, and tolerance induction. Accepted for publication: 26 September 1996     

Characterization of Surfactant Subtypes of Beractant and a Synthetic Peptide Containing Surfactant KL4 following Surface Area Cycling and Addition of Fibrinogen

Surfactant is not a homogeneous material and can be separated into subtypes. Subtype conversion is clinically important because it is thought to occur naturally and because surface activity varies depending on the subtype. Fibrinogen, a naturally occurring serum protein, is known to affect this conversion. In this study we studied two surfactants, beractant and KL₄, to examine their subtype characteristics. Surface area cycling, an in vitro method, was used in conjunction with sucrose gradient ultracentrifugation to separate subtypes in both surfactants. Activity, expressed as minimum surface tension of these subtypes, was measured using a pulsating bubble surfactometer. The effect of fibrinogen on subtype conversion and subsequent change in activity was elucidated. Our results indicate that following surface area cycling, beractant and KL₄ have different subtypes and different responses to fibrinogen. Cycling of beractant resulted in two bands, representing a heavy and a light subtype. In the presence of fibrinogen, cycling resulted in two separate heavy subtypes. Cycling of KL₄ surfactant also yielded light and heavy subtypes. However, in the presence of fibrinogen, cycling of KL₄ resulted in ultraheavy subtypes. These ultraheavy subtypes retained minimum surface tension comparable to that of native KL₄ surfactant. We conclude that these two surfactant preparations have different subtype conversions when subjected to surface area cycling and in the presence of fibrinogen. These conversions result in different activities toward lowering surface tension. We speculate that endogenous fibrinogen will also affect these two surfactants differently in vivo and thus affect their clinical effectiveness. Accepted for publication: 14 November 1996     

Serum Surfactant Protein A Levels in Healthy Individuals Are Increased in Smokers

Serum levels of surfactant protein A (SP-A) were studied in 237 healthy subjects in relation to sex, age, and smoking habits. SP-A values in male smokers were significantly higher than those in male nonsmokers (p < 0.001). The amount of cigarette smoking did not correlate significantly with SP-A values, however. SP-A values in young nonsmoking males and females were somewhat lower than those in older, but without significant difference. No significant difference in values was found between the sexes. We conclude that (1) smoking increases serum levels of SP-A, and (2) SP-A serum levels are not affected by age and sex. Accepted for publication: 6 March 1998     

Developmental Regulation of the Effects of Surfactant Protein A on Phospholipid Uptake by Fetal Rat Type II Pneumocytes

Surfactant protein A (SP-A) enhances the uptake of phospholipid by type II cells derived from adult and late gestation fetal rat lung. The present study was performed to examine more fully the developmental biology of the effects of SP-A on phosphatidylcholine (PC) uptake, to determine the effect of SP-A on the cellular location of bound and internalized phospholipid and on the metabolism of internalized phospholipid by morphologically undifferentiated (18-day) and morphologically differentiated (19-day) fetal type II cells. SP-A enhanced uptake almost twofold in a dose-dependent manner in 19-day fetal cells, but it had no effect on uptake by 18-day fetal cells at any concentration. Stimulation of uptake by 19-day fetal cells was saturable at concentrations above 1 μg/ml SP-A. Maximal uptake was 1.12 nmol of PC/mg of protein, and the effective concentration that yields 50% maximal response, K_Φ, was 58.9 ng/ml (84.1 pM). The effect of SP-A on uptake by 19-day fetal cells was detectable as early as 1 min of exposure. Uptake correlated significantly with time both in the absence (r= 0.98, p < 0.001) and presence of 5 μg/ml SP-A (r= 0.979, p < 0.001). The rate of uptake in the presence of SP-A (0.019 ± 0.002 nmol of PC/mg of protein/min) was twice the rate of uptake in controls (0.009 ± 0.001 nmol of PC/mg of protein/min). SP-A had no effect on binding to plasma membranes and uptake of phospholipid into lamellar bodies by 18-day fetal cells. On the other hand, SP-A significantly enhanced binding of dipalmitoyl phosphatidylcholine to plasma membranes (two- to threefold) and uptake into lamellar bodies (threefold) of 19-day fetal cells. SP-A caused a significant reduction in the degradation of internalized phospholipid by differentiated fetal type II cells. Based on the lack of effect of exogenous SP-A on 18-day fetal cells, we conclude that the response to SP-A is under developmental control. SP-A enhances the initial binding to the plasma membranes of fetal type II cells and subsequent internalization into the lamellar bodies. This effect is associated with a protection of internalized phospholipid from metabolic degradation. Both of these processes are developmentally regulated during the transition from the canalicular to the saccular phase of lung development. Accepted for publication: 15 May 1997     

Lung Function and Bacterial Proliferation in Experimental Neonatal Pneumonia in Ventilated Rabbits Exposed to Monoclonal Antibody to Surfactant Protein A

Surfactant protein A (SP-A) increases the resistance of surfactant to inhibition by plasma and other proteins. In a previous study we found that a monoclonal anti-SP-A antibody (R 5) increased the sensitivity of surfactant to inhibition by fibrinogen in vivo and in vitro. SP-A has been shown to stimulate microbial phagocytosis and killing by alveolar macrophages. We hypothesized that using R 5 to inactivate SP-A in an animal model mimicking congenital group B streptococcal (GBS) pneumonia might result in increased bacterial proliferation and a deterioration in lung function. Newborn near term rabbits were delivered by Cesarean section, anesthetized, tracheotomized, and ventilated for 5 h in a plethysmograph system allowing measurement of dynamic lung-thorax compliance. Postnatally the animals received one intratracheal injection (5 ml/kg) of R 5, nonspecific IgG, or normal saline. At 30 min all animals received a standard dose of an encapsulated GBS strain by intratracheal injection. The number of bacteria (mean log₁₀ CFU/g lung ± S.D.; CFU = colony forming unit) was evaluated in lung homogenates. Histologic lung sections were judged by light microscopy. Bacterial proliferation was similar in rabbits treated with the monoclonal antibody (9.33 ± 0.39; n= 14) and in control animals receiving saline (9.16 ± 0.35; n= 14) or nonspecific IgG (9.26 ± 0.31; n= 11). No significant differences were noted on the histologic analysis or in measurements of lung function. We conclude that intratracheal instillation of a monoclonal anti-SP-A antibody did not increase bacterial proliferation in GBS-infected newborn rabbits. These findings suggest that SP-A does not play an important role in protection against encapsulated GBS strains in the neonatal period. Accepted for publication: 20 June 1997     

The arterial site of action of nitric oxide in the neonatal pig lung determined by microfocal angiography

To determine the site of action of inhaled nitric oxide (iNO) in the newborn pig lung, lungs were isolated and perfused at constant flow for microfocal x-ray angiography. Measurements of pulmonary arterial diameters were made on arteries in the 100–2500 μm diameter range under control conditions, during vasoconstriction caused by hypoxia (decreasing PO₂ from ∼120 to ∼50 Torr), or N^ω-nitro-L-arginine methylester (L-NAME 10⁻⁴ M) administration, with or without vasodilation induced by iNO (40 ppm) or by the NO donor S-nitroso-N-acetylpenicillamine (SNAP 5 × 10⁻⁶ M) given intravascularly. Hypoxia caused constriction only in smaller arteries whereas L-NAME constricted arteries throughout the size range studied. iNO dilated the smaller arteries more than the larger arteries under all study conditions. SNAP was used to provide an intravascular source of NO for comparison to iNO. SNAP also dilated smaller arteries more than larger arteries, but it had a significantly greater effect on the large arteries than did iNO. This suggests that differential accessibility of the vascular smooth muscle to NO between sources, air and blood, is a factor in the diameter dependence of the responses. Accepted for publication: 13 March 2001     

Dose Dependence and Time Course of Smoke Inhalation Injury in a Rabbit Model

The dose dependence and time course of smoke inhalation injury were determined in a rabbit model. Animals were insufflated with 18–90 breaths of cotton smoke or room air (control) at a rate of 18 breaths/min and tidal volume of 12 ml/kg. Smoke-exposed animals exhibited dose-related histologic effects with progressive deterioration of respiratory function during the postexposure period of observation (96 h). The smoke-exposed rabbits had reproducible injuries to both airway mucosa and lung parenchyma, manifested by disruption and sloughing of airway and alveolar epithelia, and exudation of protein-rich fluid and leukocytes into the airway and alveolar spaces. Significant effects were evident by 24 h postexposure. Smoke inhalation also affected the respiratory burst of alveolar macrophages. Generation of superoxide anions by alveolar macrophages at 48 h postexposure was increased significantly after smoke inhalation (54 breaths). The present rabbit model should be useful for studying the interactions between pulmonary epithelial cells and leukocytes after smoke inhalation and for determining the role that abnormal functioning of alveolar macrophages plays in the development of smoke inhalation injury. Accepted for publication: 8 October 1998     

Ebselen Decreases Ozone-Induced Pulmonary Inflammation in Rats

We studied the effects of ebselen on rat lung inflammatory responses against ozone exposure. Rats were treated with ebselen every 12 h from 1 h before a single 4-h exposure to 2 ppm ozone. Treatment with ebselen (10 mg/kg) significantly decreased pulmonary inflammation as indicated by the albumin concentration and the number of neutrophils in the bronchoalveolar lavage fluid 18 h after the ozone exposure. Although treatment with ebselen did not alter the macrophage expression of inducible nitric oxide synthase after the ozone exposure, it did markedly inhibit the nitration reaction of tyrosine residues, suggesting that ebselen scavenges peroxynitrite during ozone-induced pulmonary inflammation. Treatment with ebselen also enhanced the pulmonary expression of both copper, zinc, and manganous superoxide dismutases at the same time point. These enzymes may also contribute to a decrease in the formation of peroxynitrite by lowering the concentration of superoxide. Thus, ebselen represents a useful compound for protecting against certain acute lung injuries by modulating the oxidant-related inflammatory process. Accepted for publication: 8 May 2000     

Effect of Silica Inhalation on the Pulmonary Clearance of a Bacterial Pathogen in Fischer 344 Rats

Silica inhalation predisposes workers to bacterial infection and impairments in pulmonary defense function. In this study, we evaluated the effect of pre-exposure to silica on lung defense mechanisms by use of a rat pulmonary Listeria monocytogenes infection model. Male Fischer 344 rats were exposed by inhalation to filtered air or silica (15 mg/m³× 6 h/day × 5 days/wk). After 21 or 59 days of silica exposure, the rats were inoculated intratracheally with 5 × 10³ L. monocytogenes. At 0 (noninfected controls), 3, and 7 days after infection, the left lungs were removed, homogenized, and the number of viable L. monocytogenes was counted after an overnight culture at 37° C. Bronchoalveolar lavage (BAL) was performed on the right lungs. Alveolar macrophages (AM) were collected, and the AM production of chemiluminescence (CL), an index of reactive oxygen species generation, was measured. The number of lavagable neutrophils (PMNs) and acellular BAL lactate dehydrogenase (LDH) activity were determined as indices of inflammation and injury, respectively. Pre-exposure to silica for 59 days caused substantial increases in PMN number and LDH activity compared with the air controls, whereas silica inhalation for both 21 and 59 days significantly enhanced the pulmonary clearance of L. monocytogenes compared with air controls. Dramatic elevations were also observed in zymosan- and phorbol myristate acetate (PMA)–stimulated CL production by lung phagocytes recovered from rats pre-exposed to silica for 59 days. These results demonstrate that short-term exposure to inhaled silica particles activates lung phagocytes, as evidenced by increases in reactive oxygen species. This up-regulation in the production of antimicrobial oxidants is likely responsible for the enhancement in pulmonary clearance of L. monocytogenes observed with short-term silica inhalation. Accepted for publication: 2 October 2000     

The Importance of Polymorphonuclear Leukocytes in Lipopolysaccharide-Induced Superoxide Anion Production and Lung Injury: Ex Vivo Observation in Rat Lungs

The purpose of this study is to determine if the polymorphonuclear leukocyte (PMN) is a major causative agent for lipopolysaccharide (LPS)-induced lung injury and responsible for the excess production of superoxide anion in the lung. We measured superoxide anion production from the lung and pulmonary capillary permeability in rats with and without PMN depletion. The superoxide anion production from the lung was measured using a purpose-built ex vivo chemiluminescence apparatus. Pulmonary capillary permeability was evaluated by the Evans blue dye extravasation method. PMN sequestration was determined by counting PMNs in histologic tissue specimens using microscopy. All rats received 3 mg/kg LPS intravenously. Examinations were undertaken at 2, 6, and 12 h after the LPS injection. The PMN-depleted group received cyclophosphamide 4 days before the LPS injection, which resulted in a PMN count of less than 200 cells/μl. In rats without PMN depletion, Evans blue dye extravasation increased significantly at 12 h after the LPS injection; PMN sequestration increased at 2, 6, and 12 h after the LPS injection; and superoxide anion production increased at 6 h and remained elevated at 12 h after the LPS injection. The increased permeability, PMN sequestration, and superoxide anion production were not seen in the PMN-depleted group. The contribution of the xanthine/xanthine oxidase system and alveolar macrophages to the observed superoxide anion production was negligible. We conclude that, in rats, the PMN is a major causative agent in LPS-induced lung injury and is responsible for the excess production of superoxide anion in the lung. Accepted for publication: 3 March 1997     

Alveolo-Capillary Protein Permeability and Lung Function in Patients with Late Pulmonary Complications after Allogeneic Bone Marrow Transplantation

A prospective study was performed to identify markers predictive for the development of pulmonary complications in the early (<50 days) and late (>50 days) phase after bone marrow transplantation (BMT). The characterization of BMT patients with early or late pulmonary complications revealed clear-cut differences. Early and long term increase of alveolo-capillary protein permeability was associated with smoking and was found in 20 patients developing pulmonary complications within 50 days after BMT (group 1). The 22 patients who developed such complications thereafter (group 2) had more acute graft vs host disease than 66 patients who remained free of these complications for a minimum of 1 year. Concentrations of bronchoalveolar lavage (BAL) fluid albumin (alb) and serum β₂-microglobulin (S-β₂m) were determined 10 days before BMT, on days 1, 30, and 40 after BMT, whereas lung function tests were performed before BMT, after discharge from the hospital, and 6 months as well 1 year after BMT. Using cut-off values for BAL fluid alb (>2.3 mg/dl) and S-β₂m (>0.8 mg/liter) we could significantly discriminate 12 patients out of 19 group 1 patients (early pulmonary complications) as well as 9 out of 21 group 2 patients (late pulmonary complications) from 12 out of 64 group 3 patients (without such complications) 1 day after BMT. Our results demonstrate that early increased alveolo-capillary protein permeability defines a patient population at risk to develop pulmonary complications later than 50 days after BMT with up to 1 year significantly decreased lung volumes (FEV1, 73% predicted, VC, 85% predicted). Accepted for publication: 12 June 1997     

Inhibitory Effects of Oxyradicals on Surfactant Function: Utilizing in Vitro Fenton Reaction

The inhibitory effects of reactive oxygen species (ROS) on the surface tension-lowering abilities of three surfactants were compared: natural lung surfactant (NLS), KL₄ surfactant containing synthetic peptide resembling the hydrophobic/hydrophilic domains of SP-B in an aqueous dispersion of phospholipids, and Survanta? (SUR) containing SP-B and SP-C. The inhibitory concentrations of Fenton reactants (i.e. 0.65 mM FeCl₂, 0.65 mM EDTA, 30 mM H₂O₂), deduced from dose-response plots of FeCl₂ on minimum surface tension (MST) of SUR, were used to assess the Fenton effect on biophysical properties of various surfactants. Neither H₂O₂ (30 mM) nor FeCl₂ with EDTA (both 0.65 mM) alone affected surfactant function, but when mixed together significantly increased (p < 0.01) the MST of SUR compared with KL₄ (p < 0.05) in a FeCl₂ concentration-dependent manner. This effect on NLS was not significant (p= 0.05) at similar phospholipid concentrations. Also, the range of increases in surface adsorption in mN/m at equilibrium surface tension (EST) was 27–40 for SUR, 36–44 for KL₄, and 24–25 for NLS. We speculate that the presence of SP-A and the catalase content in NLS may have protective effects on inactivation of NLS by ROS. We conclude that the in vitro Fenton effect could be a valuable test system for comparing the inactivation range of surfactants by oxyradicals. Accepted for publication: 20 May 1997     

Effects of Ventilation on Hyaluronan and Protein Concentration in Pleural Liquid of Anesthetized and Conscious Rabbits

The hypothesis of this study is that pleural lubrication is enhanced by hyaluronan acting as a boundary lubricant in pleural liquid and by pleural filtration as reflected in changes in protein concentration with ventilation. Anesthetized rabbits were injected intravenously with Evans blue dye and ventilated with 100% O₂ at either of two levels of ventilation for 6 h. Postmortem values of hyaluronan, total protein, and Evans blue-dyed albumin (EBA) concentrations in pleural liquid were greater at the higher ventilation, consistent with increases in boundary lubrication, pleural membrane permeability, and pleural filtration. To determine whether these effects were caused by hyperoxia or anesthesia, conscious rabbits were ventilated with either 3% CO₂ or room air in a box for 6, 12, or 24 h. Similar to the anesthetized rabbits, pleural liquid hyaluronan concentration after 24 h was higher in the conscious rabbits with the hypercapnic-induced greater ventilation. By contrast, the time course of total protein and EBA in pleural liquid was similar in both groups of conscious rabbits, indicating no effect of ventilation on pleural permeability. The increase in pleural liquid hyaluronan concentration might be the result of mesothelial cell stimulation by a ventilation-induced increase in pleural liquid shear stress. Accepted for publication: 30 January 1998