INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Surfactant therapy is now used worldwide, with remarkable success, for premature infants with respiratory distress syndrome (1). However, in a minority of cases the effect is transitory or minimal (4). A variety of explanations account for failure of surfactant therapy, including inactivation of surfactant in situ by a number of substances normally absent from alveolar spaces. Inactivation of pulmonary surfactant is an incompletely understood phenomenon that may have a role in the pathogenesis of a number of human lung diseases. Inactivation has been demonstrated in vitro and in vivo with a number of substances, including blood, plasma and serum proteins, lipids, antibodies, lectins, charge perturbations, bilirubin, oxidants, enzymes, and meconium (5). Both the reversible and irreversible forms of inactivation that occur may involve a number of mechanisms. Increasing the ratio of surfactant to inactivating substance has been found to reduce some forms of inactivation. Proteins contained in surfactant (principally surfactant proteins A and B [SP-A and SP-B]) reduce inactivation caused by a number of substances, but the mechanism for this effect is not known (18). Several clinical trials have found that the severity of meconium aspiration pneumonia, a disease of newborn infants, is reduced by treatment with surfactant (22, 23).

In preliminary work, we showed that SP-A reduced inactivation of surfactant not only by serum proteins (as found by others) but also with meconium (24). The possibility that the carbohydrate component of SP-A might be important for the reduction of surfactant inactivation led us to explore the effects of simple sugars, sugar polymers, and other nonionic polymers in mixtures of pulmonary surfactants in the presence and absence of meconium and other inactivating substances.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Dog pulmonary surfactant and SP-A were the gifts of S. Hawgood (University of California-San Francisco) and were prepared according to published techniques (21). From 0.2 to 4 µg of SP-A was run under reducing conditions on polyacrylamide gels and stained with Coomassie blue. Protein bands were seen at the expected molecular weights, with no evidence of contaminating proteins.

First-passed meconium from term infants was used for these experiments, although in pilot experiments little difference was found among fresh meconium, frozen meconium, or lyophilized meconium. Meconium was added to surfactant preparations that were diluted in 5 mM HEPES buffer with 0.9% NaCl, pH 7.0-7.2. Permission for use of discarded human material was obtained from the University of California-San Francisco (San Francisco, CA) Committee on Human Research. The samples were mixed by Vortex (Scientific Industries, Bohemia, NY) at room temperature for up to 2 min, until uniformly coffee beige in color. Particulates of > 1 mm were usually visible and were excluded from the sample chambers by drawing and expelling mixtures through a 25-gauge needle. In some experiments lysophosphatidylcholine (Sigma, St. Louis, MO) or human serum (obtained from healthy adults) was added to surfactant mixtures in the same way meconium was added, to study their inactivating effects.

Polyethylene glycols (PEGs), dextrans, polyvinylpyrrolidones (PVPs) of various molecular weights, or simple sugars were obtained from Sigma and used without further purification. These were added to surfactant and then mixed by Vortex for 2 min at room temperature, which dissolved the polymers or sugars readily. The mixtures were tested within minutes to hours.

Survanta (Ross Laboratories, Columbus, OH) was obtained from our nursery (excess from treated infants) and refrigerated for up to 12 mo. Rare samples of Survanta that did not lower surface tension to less than 10 mN/m on film area reduction were discarded and not used in these experiments. In some experiments, SP-A in buffered saline was added to Survanta mixtures before mixing with meconium.

Biochemicals were obtained from Sigma. Bovine serum albumin was prepared from fraction V and listed as > 99% albumin (radioimmunoassay [RIA] grade). Lysophosphatidylcholine was > 99% pure and isolated from bovine liver. These reagents were used without further purification. Ultrapure water was obtained by filtration, reverse osmosis, and deionization (18 M · cm).

The pulsating bubble surfactometer (Electronetics, Amherst, NY) as modified by Putz and coworkers (25) was used to measure surface tension. Samples with surfactant concentrations of 1.25 mg/ml (unless otherwise indicated) were placed in new 25-µl chambers. A 0.27-µl bubble was drawn and surface tension measured over 10 s; the bubble was then cycled at a rate of 20 times per minute between a volume of 0.27 and 0.70 µl while pressure was measured. Surface tension was estimated by the Laplace formula (P = 2ST/r, where ST is surface tension, P is the inflating pressure, and r is the bubble radius), assuming a spherical bubble shape. (This assumption is invalid at low surface tensions owing to bubble buoyancy and deformation [26].) Sample chambers were maintained at 37° C during the measurements by means of a temperature-controlled water jacket. The device was calibrated by water manometer and by use of pure fluids with known surface tension. The minimum surface tension after the tenth bubble deflation (30 s after the start of cycling) was selected for comparison of samples.

White Sprague-Dawley rats (250-375 g) of either sex were studied. The animals were killed by intraperitoneal injections of pentobarbital. A tracheostomy was performed using a 14-gauge catheter, the abdomen was incised, and the diaphragm was opened bilaterally. The lungs were degassed by placing the rat with a patent airway in a Bell jar and lowering barometric pressure to near water vapor pressure. Air inflation and deflation were achieved by using a 30-ml syringe attached to the tracheal cannula and pressure changes were measured at the trachea with a GM3 pressure transducer and WindowGraf chart recorder (Gould Electronics, Oxnard, CA). Total lung capacity was defined as the volume of air in the lungs at an inflating pressure of 30-35 cm H₂O after a period of stabilization (~ 45 s, or when < 0.1 ml of air entered in 10 s). Deflation pressure-volume measurements were made by withdrawing air from the syringe to reduce pressure in 5-cm H₂O decrements, allowing 20 s for pressure to stabilize at each step. Volumes were corrected for compression of gas in the apparatus (27). Control pressure-volume curves were carried out on all animals. Repeat degassing was carried out if the lungs did not appear airless at the end of the control pressure-volume maneuver. A mixture of Survanta (25 mg/ml; 4 ml/kg body weight) with 5% lyophilized meconium with or without 5% PEG (MW 10,000) was then instilled. The lungs were inflated five times to distribute the mixture and the pressure-volume curve was repeated. Lung volume results are reported as milliliters per gram estimated lung weight. We weighed lungs from eight normal rats and found that lung weight was 0.8 ± 0.4% of body weight (mean ± SEM). We therefore assumed a lung weight of 0.8% of body weight to normalize lung volumes. Pressure-volume relationships were evaluated using analysis of variance for repeated measures (28).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Major findings are shown in Figure 1 and Table 1. (Data represent means ± standard errors of the mean for a minimum of four or more replicates in all tables unless otherwise noted.) Minimum surface tensions are compared for a variety of mixtures in Table 1. Dextran, PEG, and to a lesser extent PVP and SP-A, when added to Survanta, prevent an increase in surface tension after meconium has been added to the mixturethat is to say, the additives prevent meconium inactivation of Survanta. Results with surfactant isolated from dog lung are shown for comparison. Neither dextran, PEG, nor meconium in the absence of surfactant lowers surface tension to values of < 40 mN/m.

View larger version (18K): [in this window] [in a new window]   Figure 1.   Representative bubble area/surface tension isotherms are shown for four samples using the same sample of Survanta. (A) Survanta alone (1.25 mg/ml); (B) Survanta plus 5% PEG; (C ) Survanta plus 3% meconium; (D) Survanta plus PEG plus meconium. Note that Survanta-PEG mixtures achieve a low surface tension with less reduction in surface area than does Survanta alone, that is to say less surface film compression is required (compare [B] and [A]). Meconium mixed with Survanta prevents the mixture from achieving low surface tension (inactivation) (C ), although inactivation in this particular sample of Survanta was less than usual for this concentration of meconium. PEG added to the Survanta-meconium mixtures prevents the inactivation (D).

Table 2 shows surface tension results of dilutions of Survanta in the presence or absence of PEG or dextran. In these experiments, polymers enhance the ability of dilute suspensions of Survanta to lower surface tension.

Table 3 shows results from experiments in which polymers of various molecular weights are used. All molecular sizes studied in excess of 3.3 kD were effective in preventing meconium inactivation.

Table 4 shows the results when the concentrations of the polymers are altered. Variation is greatest with concentrations of 1% polymer. The data suggest that concentrations of > 1% (especially for PVP) are required for optimal demonstration of the effect.

In additional experiments, meconium, serum, or lysophosphatidylcholine was mixed with Survanta, and then the polymer was added and the surface tension measured. These experiments tested whether inactivation of Survanta by these substances was reversible by adding polymer. Results are shown in Table 5 and clearly indicate that polymers reverse inactivation of Survanta by meconium, serum proteins, and lysophosphatidylcholine.

A number of simple sugars (5-10% solutions of glucose, mannose, maltose, and galactose) were added to Survanta before mixing with 0.1-3% meconium. The best results were found with mannose in concentrations of 10%. These results were inconsistent, however, unlike the results with polymers. The mannose-Survanta mixtures lowered surface tension to < 12 mN/m in 69% of trials in the presence of 0.2% lyophilized meconium (², p < 0.01).

Results of rat experiments are shown in Figure 2. The purpose of the rat experiments was to determine whether mechanical properties of the lung were improved when Survanta-PEG mixtures were compared with Survanta alone in the presence of meconium. This design is a simple test of in situ effect. Nine animals were used for each experimental group for measurement of pressure-volume relationships, and 12 animals were used in each experimental group for determination of total lung capacity. The pretreatment (control) pressure-volume measurements were not different for the two groups of animals, and these are combined in Figure 2. Volume is shown in milliliters per gram of estimated lung weight. Lung filling occurred at 30 cm H₂O for the pretreatment (normal) measures and required 35 cm H₂O after instillation of the test mixtures. No air leaks occurred. Deflation pressure-volume relationships were compared using analysis of variance with repeated measures. Both the control curve and the curve representing the group treated with PEG-meconium-Survanta differed from the curve representing the group treated with meconium-Survanta (used as the reference) with a probability < 0.01. Using paired values (experimental/control), we found that total lung capacity was reduced to 70 ± 3% (mean ± SEM) of control values in the group treated with Survanta- meconium. Total lung capacity was reduced to 88 ± 3% of control values in the group treated with Survanta-meconium- PEG. The difference between the Survanta-meconium group and the Survanta-meconium-PEG group is significant (Student t test, p < 0.05).

View larger version (10K): [in this window] [in a new window]   Figure 2.   Average deflation pressure-volume curves for three groups of rats. The top curve is the control curve. The middle curve represents the surfactant-meconium-PEG group. The bottom curve represents the surfactant-meconium group. ANOVA with repeated measures for the two experimental groups indicates differences between the curves with a probability of 0.01. Means and standard errors are shown for n = 9 rats in each of the experimental groups.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nonionic polymers, particularly polyoxyethylene polymers, added to Survanta permit the mixtures to behave in the presence of meconium, serum, or lysophosphatidylcholine as well or better than whole dog surfactant, or Survanta to which SP-A has been added. The effect is not specific to one polymer, it is not highly dependent on molecular weight, and it requires a relatively low concentration of polymer (> 1% and < 10%). The inactivation of Survanta by meconium, serum, or lysophosphatidylcholine is reversed in vitro by adding polymer to mixtures of inactivating substance and Survanta.

Composition of meconium has not been well studied (29). Some time ago, Shwachman and co-workers found increased protein content and decreased water in meconium from newborns with meconium ileus (cystic fibrosis) (30). Meconium from normal newborn infants is approximately 80% water. About 20% of the dry weight is extractable by organic solvent extraction. While both fractions contain inhibitory activity, more (10×) is found in the organic solvent extract (6, 31). About 12% of the dry weight of meconium is protein, and probably more than half of this is high molecular weight mucoproteins. Meconium also contains bile acids and salts, bilirubin and its derivatives, cholesterol, triglycerides, and free fatty acids (32).

We chose these particular nonionic polymers for study because of their differing structure and because they are biocompatible. Dextran is a branched polymer of glucose. PEG is a linear neutral polyether with the formula H(OCH₂CH₂)_nOH. PVP contains five membered heterocyclic pyrrolidone rings linked by vinyl groups (CH₂CH₂). PVP has been used as a plasma substitute. PEG is soluble in both water and some organic solvents. It is not toxic and is approved by the United States Food and Drug Administration for internal consumption. It forms a two-phase system with some other polymers in aqueous solutions. Covalent binding of PEG to proteins generally does not affect the biological activity of the protein, but it does reduce renal clearance of the protein. PEG binding to proteins can reduce their immunogenicity (33).

Survanta was used in this study primarily because, until recently, it has been the only animal-derived surfactant available for human use in the United States, and because we found that polymers enhanced its effects. Other surfactants for therapeutic use have greater concentrations of SP-B, and both SP-B and SP-A convey resistance to inactivation (10, 18, 34, 35). Therefore other surfactants (like the natural dog surfactant used in this study) may not have function improved to the same degree as Survanta.

Inactivation of surfactant may be defined by the failure of surfactant to achieve a low surface tension in the presence of an added substance. There are at least three common mechanisms involved with inactivation. First, certain lipids may compete with surfactant lipids for the air-liquid interface and form surface films that collapse at relatively high surface tension. Fatty acids and lysophosphatidylcholines are examples of this form of inactivation (11, 13, 32). Second, binding or aggregation of surfactant constituents may delay formation of a surface film. Inactivation with some proteins such as albumin and fibrin provide examples (36). Third, surfactant degradation either by enzyme action (lipases, proteases) or by biophysical processes has been described (37, 38).

Since the inactivation of surfactant by meconium is reversible by adding polymer to the mixtures, it is unlikely that breakdown of surfactant to non-surface-active constituents is a major explanation. On the other hand, the observations we have made in this study do not clearly establish or eliminate the other possible mechanisms for inactivation.

How do polymers prevent or reduce inactivation of Survanta by meconium? The answer to this question is by no means clear. Even in simple systems, polymer-surfactant interactions are complex (39). Because addition of polymer also reverses inactivation of Survanta by lysophosphatidylcholine and serum proteins, as well as meconium, polymers may prevent inactivation by more than one process. Although both SP-A and SP-B reduce inactivation when added to surfactant lipids, their structures differ markedly, and they differ structurally from the polymers used in this study. Therefore polymers may not prevent inactivation in ways similar to surfactant proteins. At low concentrations (less than 5% for PEG), the structure of the PEG molecules is randomly disordered coils. At 5% and above, the polymers become entangled and their effective molecular volume grows markedly, presumably creating depletion forces that exclude polymer from the hydration shells of the lipid. Concomitantly, lipids aggregate in a smaller aqueous volume, permitting a greater concentration of surfactant lipids to be available for adsorption at the air-liquid interface (40, 41). This explanation implies that meconium constituents that are inactivating are reduced in the "lipid space" and perhaps are concentrated in the polymer-rich phase. That is to say, a separation of meconium and surfactant components may occur in the presence of polymers.

There are a number of limitations to our study. Although a beneficial effect is seen in whole lung experiments, the size of the effect is modest and we have not yet studied effects in lungs from living animals. Our in vitro studies have been limited largely to a specific surfactant because of its beneficial effects on meconium aspiration pneumonia. Results in this study are limited to one means of assessing surface activity of surfactant mixtures (pulsating bubble surfactometer). Effects of meconium-surfactant-polymer mixtures have been studied under limited conditions (e.g., mixing, time, temperature, pH). The data do not permit definitive explanation of the effects at the molecular level. Several different species are represented in these experiments (surfactant from dog or cow, meconium from humans, and lung experiments in rats).

Nonetheless, if these results are confirmed and amplified, a number of clinical implications are obvious. Foremost is the possibility that existing formulations of therapeutic surfactants can be simply modified in order to improve efficacy. If inactivation of surfactant plays a major role in various pulmonary diseases, then polymers and/or surfactants containing polymers may prove useful.