Technical Field

The invention relates to natural pulmonary surfactants obtained from mammals, and more particularly, to foams comprising a natural surfactant, lipid extracts thereof, and processes of producing the same.

Backgro nd Art The major cause of perinatal morbidity and mortality in developed nations is the Neonatal Respiratory Distress Syndrome (NRDS) which arises mainly because infants delivered prematurely do not have sufficient pulmonary surfactant stores to stabilize their lungs. Pulmonary surfactant maintains normal lung function by reducing the surface tension at the air-liquid interface of the alveoli in the terminal air spaces of the lungs, thereby preventing collapse of the alveoli and bronchioles. Pulmonary surfactant effectively lowers the surface tension of the liquid film which bathes the entire cellular covering of the alveolar walls to low values sufficient to maintain alveolar inflation during all phases of the respiratory cycle.

It has long been known that pulmonary surfactant, being by definition a surface active material, can readily produce foam. The observation that exposure of animals to phosgene gas resulted in an acute efflux of foam into their major airways led Pattle to the rediscovery of the surfactant system of the lung (Pattle Nature 175:1125-1126, 1955). Acute oedema foam can also be produced by administering mixtures of oxygen and ammonia or by infusing adrenaline into animals in vivo (Pattle 1955, op cit . ; Pattle J Path Bact 72:203-209, 1956). Foam can also be obtained by infusing saline into the lungs of anaesthetized animals (Pattle 1956, op cit . ) . However, the amounts of surfactant obtained by these methods are very low. On the basis of the weight of the foam, Pattle calculated that he was able to recover less than 0.25 mg of

surfactant from an anaesthetized adult rabbit by his saline- infusion method.

Other prior art processes for producing foams comprising natural surfactant include agitating bronchoalveolar lavages, and passing air or nitrogen through bronchoalveolar lavages or fetal pulmonary fluid (Pattle 1955, op cit. ; Enhorning et al. J Exp Med 84:250-255, 1964). Pulmonary surfactant foam can also be produced by subjecting small pieces of lung to reduced pressure and from isolated lungs by perfusing the pulmonary vasculature with saline while compressing and expanding the lungs in a vacuum flask using positive and negative pressures (Klaus et al. Proc Natl Acad Sci 47:1858-1859, 1961; Bondurant et al. J Appl Physiol 37:911-917, 1962). Though foaming methods were used in the 1950's and

1960's for obtaining small samples of surfactant to study its biophysical or surface tension reducing activity, such prior art methods did not provide sufficient surfactant to allow extensive chemical characterization of surfactant using available methods or physiological testing of surfactant. Furthermore, the foam produced by such prior art methods was not in a convenient form for handling, though it could be dried to a solid and resuspended. For these reasons, the conventional methods for obtaining surfactant for conventional chemical testing and later, physiological testing, were either (1) bronchoalveolar lavage, according to which method saline was infused in the trachea, recovered and then centrifuged; or, (2) lung mincing, according to which method surfactant was obtained from saline extracts of minced lung by a series of differential centrifugation steps.

In current usage, "natural surfactant" typically refers to pulmonary surfactants recovered from alveolar washes or from amniotic fluid by simple centrifugation. A variety of such natural surfactants have been used in clinical trials to prevent and treat NRDS in infants delivered prematurely.

Natural surfactant is primarily composed of phospholipids, neutral lipids and three major protein components. Natural surfactant comprises from 80 to 95 per cent phospholipid (85- 95% [w/v]), from two to ten per cent neutral lipid (2-10% [w/v]), and from five to ten per cent protein (5-10% [w/v]). The phospholipid and neutral lipid compositions of natural surfactant are relatively constant across mammalian species. The major lipid components of natural surfactant are dipalmitoylphosphatidylcholine (DPPC), unsaturated phosphatidylcholine (PC), phosphatidylglycerol (PG), and phosphatidylinositol (PI). The ratio of the acidic phospholipids, phophatidylinositol and phosphatidylglycerol, changes during perinatal development.

The composition of bovine natural surfactant has been reported to be 35.6% dipalmitoylphosphatidylcholine (DPPC), 32.5% unsaturated phosphatidycholine (PC), 10% phosphatidylglycerol (PG) , 1.5% phosphatidylinositol (PI), 3.0% phosphatidylethanolamine (PE) , 1% lyso-Jbis-phosphatidic acid, 2.5% sphingomyelin, 3.0% neutral lipids and 10% protein (Yu et al. Liβids 18:522-529, 1983).

Natural surfactant comprises at least three proteins which have been designated surfactant-associated protein A (SP-A), SP-B, and SP-C (Possmayer Am Rev Resp Pis 138:990-998. 1988) . Surfactant-associated protein A is a sialoglycoprotein having a molecular mass of 640 kDa, and is composed of 18 monomers each having a molecular mass of approximately 35 kDa. Surfactant-associated proteins B and C are hydrophobic and lipophilic proteins of low molecular mass, and are essential for the biophysical activity of surfactant. Surfactant- associated protein B is a dimer of a molecular mass of approximately 15 kDa which migrates at approximately 5 kDa after reduction. Surfactant-associated protein C has been demonstrated to be present in isolated surfactant as a monomer of 3.5 kDa, and to a lesser extent, as an apparent dimer of 7 kDa.

One problem intrinsic to natural surfactants recovered by alveolar wash procedures is that they contain nonspecific, predominantly plasma proteins as contaminants, and may contain microbial contaminants such as bacteria or viruses. The fact that natural surfactants contain relatively high concentrations of protein, both surfactant-associated protein and also non-surfactant protein, presents problems when preparations comprising natural surfacants are administered clinically. Such preparations are potentially immunogenic due to the presence of significant amounts of protein therein. Immunogenic surfactant preparations derived from any source could lead to sensitization to these foreign proteins among infants to whom the surfactant had been administered. Natural surfactants obtained from non-human mammals are more likely to stimulate adverse immunological responses in human neonates than are those obtained from humans. Theoretically, even preparations comprising human natural surfactant could cause immunogenic complications due to immune responses to non-self human antigens based on minor genetic variations in protein structure. Though natural human surfactant has been used clinically to treat NRDS, viral or bacterial contamination of preparations comprising human natural surfactant remain a problem because natural surfactant can not be autoclaved or sterilized by other means known to the applicants such as irradiation without also losing its biophysical (i.e. surface-tension reducing) activity.

Modified natural surfactants are a second type of surfactant. Modified natural surfactants generally are prepared by extraction of lipids from natural surfactant obtained from lung minces or alveolar lavage, followed in some cases by selective addition (or removal) of certain compounds, and suspension procedures designed to restore the desired surface properties. Early studies demonstrated that protein- depleted lipid extracts of natural surfactant have the ability to reduce the surface tension of a pulsating bubble to near 0 mN/m at minimum bubble size and to promote lung expansion

and survival of surfactant-deficient prematurely delivered animals. Modified natural surfactants can prove to be biophysically ineffective (i.e. unable to reduce the surface tension of alveoli) depending upon the nature of the means used to finally resuspend the surfactant lipids for administration clinically. In developing modified natural surfactants, the objectives have been to decrease protein content, to achieve sterility, and to standardize, and improve surface properties. In the development of modified natural surfactants, problems relating to sterilization techniques, suspension techniques, reproducibility of preparation and surface properties have been considerable (Jobe et al. Am Rev Resp Pis 136:1256-1275, 1987). The requirements for multiple skills to not only prepare the surfactants but also to test them both in vivo and in vitro have tested the ingenuity of the investigators (Notter et al. J Appl Physiol 57:1613-1624, 1984).

It is well known that natural surfactant can be extracted with organic solvents such as chloroform:methanol to yield protein-depleted lipid extracts which retain those biophysical and physiological properties required for clinical use (Metcalfe et al. J Appl Physiol 49:34-41, 1980; Fujiwara et al. Lancet i:55-59, 1980; Tanaka et al. J Jap Med Soc Biol Interface 13:87-94, 1982). Organic solvent extracts of natural surfactant, also known as lipid extract surfactants, have been used extensively in prophylactic trials to prevent NRDS, and also in so-called "rescue" trials to treat the established disease (Jobe et al. Am Rev Resp Pis 1365:1256- 1275, 1987; Robertson et al. Exp Lunσ Res 14:279-310, 1988). Such prior art lipid extract surfactant preparations have been prepared from natural surfactant collected either by bronchoalveolar lavage with salt solutions or by salt solution extraction of minced lung. In such prior art methods, the saline solutions used to collect natural surfactant are subjected to a series of differential and gradient centrifugation steps in order to obtain crudely purified

natural surfactant for lipid extraction. The natural surfactant obtained in this manner is subsequently extracted with organic solvents to yield a lipid extract surfactant. Both the collection and purification of natural surfactant using such prior art methods are labourious and time-consuming. Hence, possible bacterial or viral contamination of natural surfactants derived from unsterilized natural surfactants is a deficiency intrinsic to modified natural surfactants of the prior art. Because several hours are required to complete collection and purification of natural surfactant using prior art processes, it is possible for preexisting microbial contaminants of the natural surfactant to contaminate lipid extract surfactants produced therefrom, thereby rendering them unsuitable for clinical use. For example, under suitable conditions using prior art processes, bacterial contaminants may (1) secrete lipid soluble toxins co-extractable with the lipid extract surfactant; (2) secrete phospholipases capable of degrading constituent lipid elements of natural surfactant; and (3) proliferate under suitable conditions, thereby increasing the bioburden of any resulting lipid extract surfactant. In the cases of such examples, the resulting lipid extract surfactants would be rendered unsuitable for clinical use. For instance, administration of preparations containing lipid extract surfactant contaminated by lipid soluble toxins to neonates could lead to toxicological complications posing serious health risks. In addition, preparations containing lipid extract surfactant contaminated with non-surfactant lipids (e.g. bacterial or mammalian cell membrane lipids) or having an abnormal phospholipid composition due to degradation of phospholipids by bacterial phospholipases may be biophysically inactive, and thus therapeutically ineffective. Because all contaminated lots of modified surfactants must be rejected as unsuitable for clinical use, even a low rate of contamination of lipid

extract surfactants produced using prior art collection and purification processes can prove to be very costly.

Artificial surfactants synthesized from mixtures of synthetic compounds that may or may not be constituents of natural surfactant are a third type of surfactant. See for example United States patent 4,826,821 for a review of the prior art (Clements). Artificial surfactants are attractive commercially because all potential complications relating to immunogenicity and sterility are avoidable theoretically. However, when compared with modified surfactants such as lipid extracts of natural surfactant, the record for artificial surfactants has generally not been encouraging. A deficiency intrinsic to prior art artificial surfactants stems from the fact that the biophysical activity of artificial surfactants varies with the different suspension techniques used in their preparation and their surface properties. The biophysical activity in vivo of artificial surfactants having equivalent or very similar phospholipid compositions is not readily predictable, and must be evaluated individually (Jobe et al. Am Rev Resp Pis 136:1256-1275, 1987). Routine testing in vitro of the surface properties of artificial surfactants may not accurately predict their biophysical activity in vivo .

Synthetic natural surfactants comprise the fourth and final type of surfactant known in the prior art, and are not particularly relevant to the invention. Synthetic natural surfactants are reconstructed in vitro from surfactant- specific proteins synthesized using recombinant PNA and molecular biology techniques and from mixtures of phospholipids and neutral lipids. See for example World Patent Number 8904326 (Benson et al.) and British Patent Number 2,181,138 (Schilling et al.) for summaries of this branch of the prior art.

Presently, there exist large stores of natural pulmonary surfactant which are untapped due to the limitations of prior art methods of collecting and purifying natural surfactant. A multiplicity of mammals are slaughtered

routinely to produce meat for human consumption. These animals are inspected and certified to be healthy and fit for human consumption, and their lungs contain natural surfactant suitable for processing to provide modified natural surfactants. The lungs of these mammals are either processed without first recovering the surfactant lining the lungs, or simply discarded. It is desireable to have available a method for collecting natural pulmonary surfactant quickly and easily from such slaughtered mammals on site, and for further purifying natural surfactant collected from these mammals for administration into mammalian alveolar spaces.

Pisclosure of Invention

The invention provides a means for obtaining natural pulmonary surfactant from mammals in quantity and under conditions which obviate purification steps required for natural surfactants obtained by prior art bronchoalveoloar lavage and differential centrifugation methods. The invention involves instilling a saline solution into the lungs of a freshly slaughtered mammal in situ to suspend the natural pulmonary surfactant therein and recovering the surfactant- saline suspension as butchering of the mammal proceeds. When the surfactant-saline suspension is withdrawn from the lungs, it is sufficiently agitated that the surfactant forms a foam in situ which is collected from the mammal in combination with the expelled suspension.

The invention further comprises a lipid extract surfactant separated from the foam of the invention and a process for producing the same. Lipid extract surfactant of the invention possesses excellent biophysical activity and has proven effective for treating NRDS in clinical trials.

Pesireable attributes of the invention include features such as the increased yield of natural surfactant collected (per animal) by the method disclosed herein, a significant reduction in the length of time necessary to complete the collection process enabling rapid extraction of

natural surfactant into an organic solvent mixture and therefore rapid sterilization of the natural surfactant, and simplification of the process of preparing lipid extracts of natural surfactant as compared to conventional processes of the prior art. Sequestering the natural surfactant in a foam which can be skimmed from the expelled saline obviates conventional purification steps of prior art processes.

That larger amounts of natural surfactant can be obtained from a foam produced in situ by use of the process disclosed herein than can be obtained using the conventional bronchoalveolar lavage and centrifugation processes was a novel and unexpected finding. The collection and extraction processes disclosed herein enable natural surfactant to be collected and extracted into an organic solvent mixture in a time span of minutes rather than many hours. The rapid processes of the invention for collecting and extracting natural surfactant are advantageous, and a significant improvement over prior art processes. The invention provides rapid extraction of natural surfactant into an organic solvent mixture which destroys microbial and pathogenic viral contaminants of the natural surfactant, and thereby limits the potential bioburden of lipid extract surfactant produced therefrom. Hence, the improved efficiency of the collection and extraction processes of the invention provide an important quality control feature in addition to commercial benefit.

The processes described herein for obtaining bovine natural pulmonary surfactant are merely for purposes of illustration and are typical of those that might be used. Clearly, other mammals and other procedures may also be employed, as is understood in the art.

Further features of the invention will be described or will become apparent in the course of the following detailed description.

Brief Pescription of the Prawinσs

Figure 1 shows a flow chart of the process for making lipid extract surfactant;

Figure 2 shows a flow chart of the process for preparing a lipid extract surfactant suspension.

Best Mode For Carrying Out The Invention

The processes described herein for sequestering bovine natural surfactant in a foam readily collectable from a mammal via its trachea and of preparing a lipid extract surfactant are merely for purposes of illustration, though typical of those that may be used. Mammals other than cattle and other procedures may also be employed, as is understood in the art. For instance, natural surfactant foam may be similarly obtained from mammals such as horses, sheep, rabbits, pigs, dogs and cats.

Preparation of Bovine Surfactant Foam in situ

In the preferred embodiment of the invention, a bovine animal is killed. The animal can be bled. The animal's trachea is then exposed without damaging the lungs.

(Pamage to the lungs may make the animal unsuitable for use.)

To permit collection of natural surfactant from the lungs of the mammal, a length of plastic tubing having an outer diameter sized to fit snugly within the animal's trachea (e.g. of 3.75 cm [1.5 inches]) is inserted into the trachea so that a first end of the tubing lies close to the bifurcation of the animal's bronchi. The tubing is then secured within the trachea, either manually or with a length of butcher's cord or the like. After the tubing has been secured, approximately ten to fifteen litres of an ice-cold saline "working solution" (0.15 M NaCl, 10 mM CaCl ₂ , 8 mM MgCl ₂ ) are introduced into the trachea manually or by pump means. For example, a vessel such as a jerrycan containing working solution is elevated in communication with the trachea so that the required volume of saline flows into the animal's

lungs. Natural surfactant lining the lungs is thereby suspended in the saline solution in situ . After the saline solution has been instilled into the lungs, the resulting surfactant-saline suspension is withdrawn from the lungs and returned to the stock working solution, either by gravity or pump means. Next, the lungs are again refilled with surfactant-saline suspension from the stock working solution. This final volume can be allowed to remain in the lungs until the animal's posterior end or hind quarters are elevated by a hoist or other means, so that the animal's lungs are one- half to one meter below its posterior end. The surfactant- saline suspension in the lungs is then allowed to flow back into the jerrycan by gravity. Under the force of gravity, the surfactant-saline suspension is expelled from the lungs and trachea with force sufficient to cause the natural surfactant to foam in the lungs in situ , with a copious white foam being expelled in combination with the remaining suspension. Preferably, the jerrycan is lowered to collect as much of the suspension/foam combination as possible. Preferably, the animal's viscera are then removed and the lungs are exposed. Preferably, surfaces of the lungs are massaged until no more foam is observed flowing into the tubing, to collect the last of the foam. Massaging of the lungs is best achieved by applying firm pressure with the palm of a hand onto the lung surface. (Any preparations containing significant amounts of blood must be discarded.)

Once the last of the foam has been collected, the surfactant-saline suspension in combination with foam is transferred into a large container and let stand for a few minutes to allow the foam to rise to the surface. The copious white foam is then skimmed off into a fresh container.

Preparation of Lipid Extract Surfactant

Referring to Figure 1, the foam is strained after collection of foam from a number of animals is complete.

Approximately five litres of foam are strained at a time.

Each five litre aliquot of foam is strained through five layers of cheesecloth using a Buchner funnel and collected in a side-arm flask containing 100 ml chloroform:methanol (1:1 [v/v]). (Suction is applied via the side-arm with a water aspirator.) The foam dissolves into the organic solvents virtually instantaneously. Any microbial contaminants of the foam will be killed by this step.

The final volume of the filtrate is measured and sufficient 10% potassium chloride (10% [w/v] KCl) is added to give a chloroform:methanol:1.0% KCl ratio of 1.0:1.0:0.9 (Bligh et al. Can J Biochem Physiol 37:911-917, 1959). The combined fluids are mixed by gentle rotation and inversion to extract all lipids into the chloroform. The resulting mixtures are then centrifuged in glass flasks at 800 g at 4°C for 20 minutes to produce two phases; namely, a bottom phase containing chloroform and lipids, and an upper phase containing water and methanol. A "pad" of pelleted protein is collected at the interface between the two phases. Both phases of the clear supernatant are decanted into a fresh vessel, leaving the protein pellet in the centrifuge tube. The mixture is swirled briefly and recentrifuged as above for 10 minutes. It is also possible to store these tubes without centrifugation until the two phases separate.

After a suitable interval, the upper aqueous phase and any remaining precipitated protein are removed with a water aspirator. Alternatively, the upper aqueous and lower organic solvent phases may be isolated using a separatory funnel. The organic solvent layer is then evaporated; preferably using a rotary evaporator under reduced pressure. A 500 ml boiling flask is used for 2.5 g of lipid.

To remove any trace amounts of precipitated protein, the resulting viscous oil is re-extracted with chloroform:methanol (1:1) and left at 4°C for several hours. The solution is then centrifuged at 400 g for 15 minutes. The resulting supernatant is decanted, and then evaporated under reduced pressure to dryness on a rotary evaporator at least

twice. The resulting lipid extract is then taken up with a small volume of chloroform (to give a lipid concentration of approximately 150 mg/ml) and transferred to a 250 ml centrifuge tube. The lipid extract is dried under nitrogen until an oily residue is formed on the walls of the tube, precipitated with 20 volumes (typically 250 ml) of cold acetone, stored for several hours at -20°C, and then centrifuged at 800 g at 4°C for 20 minutes. The resulting precipitate is resuspended in chloroform and the acetone precipitation step is repeated.

The final precipitate may be dissolved and stored in a small volume of chloroform:methanol (1:1) (approximately 250 mg/ml) at -20°C indefinitely, or as shown in Figure 2, suspended in saline solution for tests of biophysical activity or administration into mammalian alveolar spaces.

Referring to Figure 2, to prepare a suspension of lipid extract surfactant suitable for testing the biophysical activity of the lipid extract surfactant purified from the foam or for clinical application, 500 mg of lipid in chloroform:methanol (1:1) is evaporated under reduced pressure in a small boiling flask. Typically, the required volume to give 500 mg is approximately 2 ml. Next, the lipid is redissolved in chloroform:methanol and centrifuged at 800 g for 20 minutes. The resulting liquid fraction is transferred to a 50 ml round bottom screw-cap centrifuge bottle and the lipid is dried under nitrogen onto the walls of the tube, so that the lipid forms an even coating on the lower surface of the centrifuge tube.

After drying, the lipid-coated tubes are lyophilized for 20 minutes to remove any traces of solvent. The lipid is then resuspended by vortexing and bath sonication at 35 to 40°C in a solution of 0.15 M NaCl, 1.5 mM CaCl ₂ , to give a final lipid concentration of 25 mg/ml. Preparations of resuspended surfactant may be pooled, mixed with a stirring bar, and aliquots thereof may be pipetted into sterile serum vials, stoppered with slotted stoppers, and the solvent

evaporated from the vials evacuated under gentle vacuum. Preferably, the vials should be capped with aluminium caps and crimped. Preferably, the vials are autoclaved at 121°C for 15 to 20 minutes to sterilize the aliquots, cooled, and stored frozen at -20°C.

Characterization of lipid extract surfactant Biophysical activity

Pispersions of lipid extract surfactant in saline- 1.5 mM CaCl ₂ share the ability of natural surfactant to reduce the surface tension of a pulsating bubble to near 0 mN/m at minimum bubble radius. To test the biophysical activity of the lipid extract surfactant a 0.2 ml sample is withdrawn from a vial and diluted to 0.5 ml with saline containing 1.5 μM CaCl ₂ (i.e. to a concentration of 10.0 mg lipid extract surfactant per ml). The biophysical activity is determined with a pulsating bubble surfactometer (Electronetrics Corporation, Amerεt, NY, USA). This assay monitors both the adsorption of the surfactant phospholipids to the air-saline interface to form a surface-active monolayer and the squeeze- out of unsaturated phospholipids, leaving a monolayer enriched in dipalmitoylphosphatidylcholine (DPPC) which can reduce the surface tension to low values during the dynamic compression produced during the reduction of bubble surface area (Enhorning J Appl Physiol 43:198-203, 1977; Yu et al. Lipids 18:522-529, 1983; Weber et al. Biochim Biophys Acta 796:83-91, 1984).

With the pulsating bubble surfactometer technique, a bubble communicating with ambient air is created in a small chamber. The bubble is pulsated between radius of 0.4 to 0.55 mm at 20 cycles per minute at 37°C, and acts as a single artificial alveolus. The pressure across the bubble is monitored with a pressure transducer. Surface tension is calculated according to the Law of Young and Laplace which states that the difference in pressure across the bubble is equal to two times the surface tension divided by the radius.

Surface tensions at maximum bubble radius and minimum bubble radius are calculated. To have acceptable levels of biophysical activity, lipid extract surfactant preparations must reduce the surface tensions to 30 ± 5.0 mN/m at maximum bubble radius, and 2.5 ± 2.5 mN/m at minimum bubble radius within 50 pulsations. Any preparations not meeting these criteria should be abandoned.

Biochemical characterization The composition of lipid extract surfactant prepared by the processes disclosed herein is of a high degree of purity, and is very similar to that reported for highly purified bovine pulmonary surfactant obtained by bronchoalveolar lavage (Yu et al., Lipids 18:522-529, 1983; Weber et al. Biochim Biophys Acta 796:83-91, 1984). Tables 1 and 2 show the lipid compositions of bovine pulmonary surfactant and that of the bovine lipid extract surfactant separated from the foam described herein, respectively.

Table 1

Lipid Composition of Bovine Pulmonary Surfactant (From Yu et al., Lipids 18:522-529, 1983)

Phospholipid % Total Phosphorus

(n = 4)

Phosphatidylcholine 79.2 ± 1.6

Phosphatidylglycerol 11.3 ± 0.5

Phosphatidylinositol 1.8 ± 0.3 Phosphatidylethanolamine 3.5 ± 0.5

Lyso-J is-phosphatidic acid 1.5 ± 0.4

Sphingomyelin 2.6 ± 0.5

Table 2

Phospholipid Composition of Bovine Surfactant isolated by foaming in situ

Phospholipid % Total Phosphorous

(n = 3)

Lyso-phosphatidylcholine 0.2 ± 0.03

Sphingomyelin 2.0 ± 0.35 Phosphatidylcholine 79.2 ± 0.65

Phosphatidylinositol 1.0 ± 0.03

Phosphatidylethanolamine 3.0 ± 0.18

Phosphatidylglycerol 14.4 ± 0.51

Lyso-Jbis-phosphatidic acid Trace

The phospholipid content of the lipid extract surfactant is determined by the assay of Rouser et al. fLipids 5:769-775, 1970). The method of Puck-Chong fLipids 14:492- 497, 1979) provides identical values. A 200 μl sample of lipid extract surfactant is diluted to 500 μl and 50 μl aliquots are spotted on Whatman 5P plates. Chloroform/etha- nol/water/triethylamine (30:34:8:35) is used to develop the plates according to the method of Touchstone et al. (Lipids 15:61-62, 1980). After development, each plate is dried and sprayed with a phosphate spray as described in Pittmer and Lester (J Lipid Res 5:126-127, 1964). Lipid extract surfactant of the invention contains 75 to 85% phosphatidylcholine, 8 to 16% acidic phospholipids such as phosphatidylglycerol and phosphatidylinositol, 2.5 to 7.5% phosphatidylethanolamine, 0.1 to 3% lyso-jbis-phosphatidic acid and < 5.0% sphingomyelin. Lipid extract surfactants not conforming to the above lipid profile should be discarded. Similarly, if a lipid extract surfactant were to contain 3.0%

lysophosphatidylcholine or more by weight, that preparation should be discarded.

The composition of the lipid extract surfactant of the invention contains lower levels of lysophosphatidyl- choline, cholesterol and cholesterol esters than reported for prior art lipid extract surfactants (U.S. Patent Number 4,338,301, issued July 6, 1982; Notter et al. J Appl Physiol 57:1613-1624, 1984; Shelly etal. Lung 160:195-206, 1982; U.S. Patent No. 4,397,839, issued August 9, 1983; Berggren et al. Exp Lunσ Res 8:29-51, 1985; King et al. Handbook of Physiology. The Respiratory System [Fishman AP and Fisher AB, eds.], Washington: American Physiological Society, 1:309-336, 1985). When present in excess, these compounds can inhibit the biophysical activity of lipid extract surfactant. It appears that the higher levels of lysophosphatidylcholine and cholesterol present in some prior art preparations arise from contamination of pulmonary surfactant by cellular membranes (Rooney et al. Biochim Biophys Acta 431:447-458, 1976; Holm et al. J Appl Physiol. 1987). In addition, the lipid extract surfactant of the invention contains relatively low levels of another membrane lipid, namely, sphingomyelin. (The lecithin/sphingomyelin (L/S) ratio is used as an indicator of the relative amounts of surfactant to non-surfactant lipids in surfactant isolated from a niotic fluid and other sources [Gluck et al. Ped Res 1:237-246, 1971].) Because collection of natural surfactant by sequestering it in a foam in the lungs in situ as opposed to mincing avoids damage to the lungs and trachea of the mammal, the foam production and surfactant collection process of the invention precludes the possibility of contamination of the natural surfactant with cellular membranes.

The protein content of the lipid extract surfactant of the invention is measured using the method of Lowry et al. (J Biol Chem 132:265-275, 1951) in the presence of sodium dodecylsulphate as shown in Possmayer et al. fCan J Biochem 55:609-617, 1977) using bovine serum albumin as a standard.

The lipid does not interfere with samples having 2.0 mg lipid or less in a final volume of 5.0 ml. By this method, the protein content of the lipid extract surfactant is 12.5 ± 6.0 μg protein per mg phospholipid. Natural surfactant obtained by the standard lavage/centrifugation prior art processes comprises proteins other than SP-A, SP-B and SP-C. These are presumably serum contaminants; albumin is one of these proteins. Electrophoresis of lipid extract surfactant on polyacrylamide gels with sodium dodecylsulphate shows decreased amounts of serum proteins, no SP-A, only SP-B and SP-C (Possmayer Am Rev Resp Pis 138:990-998, 1988). The depletion of non-surfactant proteins characteristic of the lipid extract surfactant of the invention reduces the antigen load of the lipid extract surfactant, and advantageously reduces its potential immunogenicity. Because the gene sequences encoding SP-B and SP-C are highly conserved, and both proteins are small and hydrophobic, they are not likely to be very immunogenic. Hence, lipid extract surfactants derived from a natural surfactant obtained from mammals as disclosed herein are unlikely to pose problems of immunogenicity in clinical use.

It will be appreciated that the above description relates to the preferred embodiment by way of example only.

Many variations on the invention will be obvious to those knowledgeable in the field, and such obvious variations are within the scope of the invention as described and claimed, whether or not expressly described.

For instance, organic solvent extractions of the natural surfactant foam are normally performed using chloroform:methanol, but in principle, any extraction method which removes a plurality of the potentially immunogenic non- surfactant proteins and SP-A, but retains the low molecular weight hydrophobic proteins SP-B and SP-C essential for biophysical activity could be used for this purpose.

Industrial Applicability

Organic solvent extraction of the foam of the invention provides a lipid extract surfactant with biophysical and physiological activities appropriate for administration into mammalian alveolar spaces for clinical use to prevent and treat neonatal respiratory distress syndrome, to prevent or treat adult respiratory distress syndromes (e.g. shock lung, pneumonia) , or for clinical use in connection with lung transplants.