Reutilization of phosphatidylglycerol and phosphatidylethanolamine by the pulmonary surfactant system in 3-day-old rabbits

172

BBA 51892

Reutilization of phosphatidylglycerol

and phosphatidylethanolamine

by the

pulmonary surfactant system in 3-day-old rabbits Harris C. Jacobs *, Alan H. Jobe, Machiko Perinatal Research Laboratories,

Lung surfactant;

and Sally Jones

Harbor- UCLA Medical Center, UCLA School of Medicine, Torrance, CA 90509 (U.S.A.)

(Revised

Key words:

Ikegami

Phospholipid

(Received manuscript

1000 W. Cmton Street,

September 5th, 1984) received December 17th, 1984)

reutilization;

Phosphatidylglycerol;

Phosphatidylethanolamine;

(Rabbit)

Developing rabbits reutilize the phosphatidylcholine of surfactant with an efficiency of about 95%. The efficiency of reutilization of other components of surfactant have not been determined. 3-day-old rabbits were injected intratracheally with [ 3H]dipalmitoylphosphatidylcholine (DPPC) mixed with unlabeled natural surfactant and either disaturated [ 32P]phosphatidylglycerol (DSPG) or [ “Cjdipalmitoylphosphatidylethanolamine (DPPE). The recovery of [ ‘H]DPPC, [ “C]DPPE, and [ 32P]DSPG in the alveolar wash was measured at different times after injection. By plotting the ratio of [ 32P]DSPG to [ 3H]DPPC or I “CIDPPE to 13H]DPPC counts/min in the alveolar wash vs. time after injection we showed that these two phospholipids are reutilized less efficiently than phosphatidylcholine. Based on other studies, several assumptions were made about the kinetics of surfactant phosphatidylethanolamine and phosphatidylglycerol. From the slopes of the semilog plots of total [ 14CIDPPE and total 1” P]DSPG counts/min in the alveolar wash vs. time and these assumptions, we determined that these two phospholipids were reutilized at an efficiency of only 79%.

Introduction The mammalian lung maintains a normal functional pool of alveolar surfactant by a process whose complexity has become more apparent from the work of several investigators [l-4]. Surfactant is stored as lamellar bodies in type II cells prior to secretion by exocytosis [5]. Following secretion, at least several components of surfactant, including phosphatidylcholine, are recycled by the type II cell [3,4,6]; in 3-day-old rabbits, the reutilization of surfactant phosphatidylcholine exceeds 90%.

* To whom correspondence should be addre :ssed at (present address): Department of Pediatrics, Yale Llniversity School of Medicine. P.O. Box 3333. 333 Cedar Strl eet, New Haven,

CT 06510, U.S.A. 0005-2760/85/$03.30

The efficiency of reutilization of other components of surfactant has not been determined. We demonstrated, also using 3-day-old rabbits, that several analogues of phosphatidylcholine are reutilized by the pulmonary surfactant system in a manner indistinguishable from the parent molecule [7]. In the present study we used the same techniques to assess the ability of 3-day-old rabbits to reutilize two other phospholipids which are normally found in surfactant. Materials and Methods Injection injection

solutions. solutions.

We prepared The

first

solution

two separate contained

[methyl-3 HIcholine labeled L-a-dipalmitoylphosphatidylcholine (DPPC) (30 Ci/mmol New Eng-

0 1985 Elsevier Science Publishers B.V. (Biomedical

Division)

173

land Nuclear Corp.) and L-a-[dipalmitoyl-l“C]phosphatidylethanolamine (DPPE) (86 mCi/ mmol New England Nuclear Corp.). These two phospholipids were mixed in chloroform, dried under N, at 50°C and diluted in distilled water. The mixture was sonicated for 4 min, using the large probe of a Fisher sonic dismembrator (model 300) at 60% maximum energy output [6,7]. To this mixture we added a small amount of unlabeled natural surfactant collected by centrifugation from the alveolar washes of 3-day-old rabbits as before ]6,71, and enough Lactated Ringers Injection (Travenol Lab, Inc.) to produce a 1 : 1 mixture of Lactated Ringers Injection and distilled water [8]. Each rabbit received, by intratracheal injection, 0.2 ml of the injection solution which contained approx. 40000 cpm of [ 3H]DPPC, 70000 cpm of [14C]DPPE and an amount of surfactant equivalent to about 5% of the endogenous alveolar pool size. The second injection solution contained [3H] DPPC and [ 32P]phosphatidylglycerol which was saturated in both acyl positions (DSPG). [ 32P]DSPG was prepared by injecting [ 32P]orthophosphate (carrier free) (New England Nuclear Corp.) into the lungs of five 3-day-old rabbits via the trachea (see below). 40 h after injection, the rabbits were killed and their lungs removed and homogenized in 0.15 M saline. The homogenate was extracted once with 3 vol. of a 2 : 1 chloroform/methanol mixture and the chloroform phase was collected. The chloroform was evaporated under N, at 50°C and the disaturated phosphatidylcholine (DSPC), was isolated according to Mason et al. [9]. This DSPC, which contains [32P]DSPC, was converted to [32P]DSPG using phospholipase D (Sigma Chemical Co.) according to the procedure of Dawson [lo]. Briefly, DSPC, glycerol and phospholipase D are incubated over night at room temperature in a mixture of chloroform and buffer (40 mM CaCl,/O.l M acetate, pH 5.6). Phospholipase D exchanges the choline head-group for glycerol with high efficiency and the [ 32P]DSPG is isolated by thin-layer chromatography on handmade silica gel H TLC plates (Supelco) in one-dimension (chloroform/ methanol/water/acetic acid, 65 : 25 : 8 : 4 : 4, v/v). [ 32P]DSPG was eluted from the silica using 2 : 1 chloroform/methanol and concentrated under N,

at 50°C. The purified [32P]DSPG had a specific activity of about 0.3 mCi/mmol. The second injection solution was prepared in a similar fashion except for substituting [ 32P]DSPG for [14C]DPPE. Each rabbit received 0.2 ml of this solution containing about 20000 cpm of [ 32P]DSPG (0.03 mmol), 35 000 cpm of [ 3H]DPPC and an amount of unlabeled surfactant equivalent to about 5% of the endogenous pool of surfactant. Animals. All rabbits were injected intratracheally as before [6,7]. A midline neck incision was made under local anesthesia and the trachea was visualized. One of the two solutions was injected into the distal trachea via a 30 gauge needle while manually occluding the trachea proximal to the injection site. The neck wound was closed with 4-O silk suture. No animal experienced more than 1 or 2 min of mild respiratory distress from the injection procedure. 27 3-day-old rabbits weighing 66.9 f 3.1 g were injected with the solution containing [‘4C]DPPE. They were killed at one of nine preset times from 2 to 60 h after injection. 54 3-day-old rabbits weighing 62.6 & 1.6 g were injected with the solution containing [ 32P]DSPG. These rabbits were killed at one of ten preset times from 2 to 72 h after injection. All rabbits were hand fed throughout the course of the experiment and all were healthy at the time of killing. Fraction isolation. After having been killed, each rabbit was subjected to a thorough alveolar wash with normal saline [3,6]. The washed lungs were homogenized in 0.32 M sucrose containing 0.01 M Tris-HCl/O.lS M NaCl/O.OOl M CaCl,/O.OOl M MgS04/0.0001 M EDTA (pH 7.4) [ll]. An aliquot of the homogenate was saved and the remainder was used to isolate lamellar bodies by a series of differential and sucrose density gradient centrifugation steps as previously described [6,11]. Lipid analysis. The lipids in each fraction were extracted according to Folch et al. [12]. Phosphatidylcholine was separated from phosphatidylethanolamine and phosphatidylglycerol by one-dimensional thin-layer chromatography on handmade silica gel H TLC plates using chloroform/ methanol/acetic acid/water (65 : 25 : 8 : 4, v/v) as the solvent. The phosphatidylcholine and the phosphatidylethanolamine-phosphatidylglycerol spots (these two phospholipids comigrate with this

174

solvent system) were scraped into separate scintillation vials. Aquasol(New England Nuclear Corp.) was added and the spots were assayed for 3H and 32P or 3H and 14C counts/mm as appropriate. These counts were used to determine the ratio of 3H to 14C or 3H to 32P counts/min in the various fractions vs. time after injection as well as total recovered counts/mm vs. time in the alveolar wash and in the residual lung homogenate. Several alveolar wash samples from rabbits given each injection solution were extracted in duplicate. The duplicate samples were chromatographed in two dimensions using chloroform/ methanol/ 58% ammonium hydroxide/ water (195 : 90 : 7.5 : 6, v/v) as a second dimension which provided additional separation of the phospholipids including separation of phosphatidylethanolamine from phosphatidylglycerol. These samples were used to determine if radioactivity accumulated in any lipid other than those administered. For both injection solutions, radioactivity was not found in any lipid other than the original lipids. Analysis. When rabbits are injected intratracheally with radiolabeled DPPC mixed with radiolabeled surfactant, there is a decrease with time in total recovered phosphatidylcholine counts/min. In a previous experiment where this was done, the change with time in the total recovered counts/min was best described by a linear sum of exponentials [6]. In that experiment, the equations describing the decay in labeled DPPC predicted the same percent reutilization as the equations describing the decay in labeled surfactant. We previously demonstrated that the reutilization of analogues of DPPC can be assessed by mixing the labeled analogues with labeled DPPC and unlabeled surfactant and injecting the mixture into rabbits intratracheally [7]. By plotting the ratio of counts/ min in the analogue to counts/min in DPPC, the metabolism of the analogue relative to DPPC was determined. Here we used the same methodology to compare the metabolism of phosphatidylglycerol and phosphatidylethanolamine relative to phosphatidylcholine. All results are expressed as the ratio of counts/ min in phosphatidylethanolamine or phosphatidylglycerol to phosphatidylcholine or as total recovered counts/min. Statistical comparisons were by a Student’s f-test.

Results We calculated the expected percent recovery in the alveoli of the [3H]DPPC given in the two protocols described here. These calculations were based on the results of a previous experiment which described the recovery in the alveolar wash of radiolabeled phosphatidylcholine given as part of natural surfactant injected into 3-day-old rabbits [6]. These predicted values were plotted on a semilog axis and the best fit line was determined by the method of least-squares (Fig. 1A). The percent recovery of [3H]DPPC in the alveolar washes of rabbits from both protocols presented here was measured and the mean values for each time of killing were calculated and plotted on a semilog axis. Linear least-squares regression was used to determine the best fit line to these means (Fig. 1 B,C). The slopes of the lines in Figs. 1 B and C were found to be not significantly different from the slope of the line in Fig. 1A. The [3H]DPPC in these experiments behaved as if it was the phosphatidylcholine of the endogenously secreted surfactant.

0

60

30

Hours Fig. 1. Predicted and measured recovery of [ 3H]DPPC in the alveolar wash following intratracheal injection. (A) Predicted recovery of [ 3H]DPPC expressed as the log of the % injected vs. time after injection. Predictions are based on the data in Ref. 6. The line shown was determined by linear least-squares regression. (B) Measured recovery of [ 3H]DPPC in the alveolar wash expressed as the log of the % of total [ ‘H]DPPC injected vs. time after injection for rabbits given [3H]DPPC and [32P]DSPG. Each point represents the mean for rabbits killed at that time. Standard errors fall within the data points. The included line was determined by linear least-squares regression. (C) Same as in Part B but for rabbits given [3H]DPPC and [ l4 CIDPPE.

175

Fig. 2A shows the mean ratio of [32P]DSPG to recovered in the alveolar washes of rabbits vs. time after injection. The best fit line, which was determined by linear least-squares regression, had a slope which differed significantly from 0 (P < 0.01). A similar plot was generated for the mean ratio of [14C]DPPE to [ 3H]DPPC recovered in the alveolar wash vs. time after injection (Fig. 2B). The slope of the line determined by linear least-squares regression on the means again differed significantly from 0 (P < 0.01). Both DPPE and DSPG were cleared from the pulmonary surfactant system faster than the DPPC with which they were injected. As expected from the plots shown in Figs. 1 and 2, the total recovered [32P]DSPG in the alveolar wash decreased with time. This also was true for the total recovered [‘4C]DPPE in the alveolar washes for rabbits injected with this phospholipid. We standardized the total recovery in the alveolar wash of each of these phospholipids by multiplying the ratio of [14C]DPPE to [3H]DPPC (or [32P]DSPG to [3H]DPPC) by the predicted recovery of [3H]DPPC for that time of killing. The mean of these adjusted values for each time of killing was calculated and plotted on a semilog

[ 3H]DPPC

0

30

60 HOlJlS

Fig. 2. Ratio of [32P]DSPG and [14C]DPPE to [3H]DPPC counts/mm in the alveolar wash. (A) The ratio of [“P]DSPG to [ 3H]DPPC counts/min in the alveolar wash of each rabbit was measured. Each point represents the mean of this ratio for all rabbits killed at each timef 1 SE. The included line was determined by linear least-squares regression on the means and has a slope which is statistically different from 0 (P < 0.05). (B) Same as Part A but for rabbits given [t4C]DPPE rather than [32P]DSPG. The slope of the regression line is statistically different from 0 (P < 0.05).

axis vs. time after injection. The best fit line was determined by linear least-squares regression (Fig. 3). The slope of this line is used to obtain the biological half-life values given in Table I (See Discussion). Standardizing the total recovered [14C]DPPE and [32P]DSPG counts/mm to the same [3H]DPPC recovery curve allows the reutilization of these three phospholipids to be compared directly. Furthermore, the standardization technique did not significantly change the slope of the semilog plot of total recovered counts/min vs. time after injection for [14C]DPPE or for [ 32P]DSPG. The calculated reutilization of phosphatidylglycerol and phosphatidylethanolamine are given in Table I. (See Discussion.) Examination of lamellar bodies produced results which were consistent with a more rapid clearance of phosphatidylglycerol and phosphatidylethanolamine from the surfactant system when compared to phosphatidylcholine. The rabbits given [3H]DPPC and [r4C]DPPE had a decrease in the ratio of [14C]DPPE to [3H]DPPC in lamellar bodies vs. time after injection. Similarly, the rabbits given [ 3H]DPPC and [ 32P]DSPG had a decreasing ratio of lamellar body [ 32P]DSPG to [3H]DPPC vs. time after injection. However, in both cases, the scatter resulted in slopes which

OJ 0

30

60 HOUM

Fig. 3. Standardized total recovered [ 32P]DSPG and [‘4C]DPPE in the alveolar wash. (A) The mean ratio of [ 32P]DSPG to [’ H]DPPC at each time was multiplied by the [ 3H]DPPC total recovered counts/nun predicted from the data in Ref. 6. Each point represents the log of the mean standardized total [32P]DSPG counts/mm recovered in the alveolar wash for rabbits killed at the indicated times. The included line was determined by linear least-squares regression on the means. (B) The same as Part A but for rabbits given [r4C]DPPE.

176 TABLE

I

REUTILIZATION OF PHOSPHATIDYLGLYCEROL AND PHOSPHATIDYLETHANOLAMINE BY THE PULMONARY SLJRFACTANT SYSTEM OF 3-DAY-OLD RABBITS The biological half-life is determined from the slope of the semilog plot of total counts/mm vs. time after intratracheal injection. This value is used to calculate the percent reutilization (see text). Phosphohpid

Biological half-life

Reutilization (S)

(h) Phosphatidylglycerol Phosphatidylethanolamine

were not statistically shown).

39.5 38.8

different

79+8 79+9

from

0 (data

not

Discussion DPPE represents, at most, a minor subspecies of this phospholipid in surfactant. Also, the DSPG used was a racemic mixture, L-cu-phosphatidylD,L-glycerol, rather than a pure isomer. These variations from normal surfactant components must be considered in extending our results to the behavior of these phospholipids in endogenously secreted surfactant. Previous studies in developing rabbits found no differences in the kinetics of secretion and reutilization of phosphatidylcholine molecules of surfactant containing different fatty acids [3,6,7]; and four analogues of the naturally occurring L-a-dipalmitoylphosphatidylcholine, including the optical isomer of this molecule, were handled by the surfactant system of 3-day-old rabbits as if they were naturally occurring forms [7]. These studies as well as those by other investigators [8,13,14] are most consistent with removal of surfactant aggregates from the alveoli by a process of bulk uptake. For these reasons as well as the finding that the percent recovery of the [3H]DPPC given to the rabbits in the present study was not different than predicted from a prior study (Fig. 1) [7], it is reasonable to assume that the [ 32P]DSPG and the [14C]DPPE with which the [3H]DPPC was mixed were treated by the surfactant system as if they were part of the endogenously secreted surfactant. All conclusions are based on this assumption.

The ratio of [‘*P]DSPG to [ ‘HJDPPC and the ratio of [ “C]DPPE to [ 3H]DPPC measured in the alveolar wash over the time-course of the experiment indicate that the reutilization of these two phospholipids is different than that of phosphatidylcholine. However, the slopes of the lines fit to these ratios do not translate directly into estimates of the efficiency of reutilization. Estimates of the efficiency of reutilization of these two phospholipids can be obtained by relating the data in the present experiment to the previously published information on the turnover and reutilization of surfactant phosphatidylcholine. The calculation of the efficiency of reutilization of these two phospholipids from the slopes of the regression lines in Fig. 3 is based on the following assumptions: (1) a precursor-product relationship exists between the phosphatidylglycerol and phosphatidylethanolamine found in the lamellar bodies and the alveolar lavage, (2) the radiolabeled molecules of phosphatidylglycerol and phosphatidylethanolamine which are reutilized are reutilized as intact molecules, and (3) the molecules not reutilized are degraded and diluted in pools sufficiently large that the radiolabeled component does not reappear in surfactant. It has been determined by various investigators that lamellar bodies contain both phosphatidylglycerol and phosphatidylethanolamine [1,15.16]. Lamellar bodies are exocytosed into the alveolar space [5], hence, they are a source of surfactant phosphatidylglycerol and phosphatidylethanolamine. The assumption of a precursor-product relationship then involves, assuming that no other source of the phospholipids for surfactant exists. It has been demonstrated that alveolar surfactant phosphatidylcholine comes only from lamellar body phosphatidylcholine [l-3]. When any radiolabeled precursor of phosphatidylcholine is injected into animals, less than 5% can be found in the surfactant phosphatidylcholine at later times, and the recovery is usually much less [17). It is unlikely that the recovered radiolabeled phosphatidylglycerol and phosphatidylethanolamine in the alveolar surfactant of the present experiment are due to breakdown and resynthesis. Alveolar phosphatidylcholine has a turnover time (time required to fill the pool if it were empty) of about 10 h in 3-day-old rabbits [3]. In

177

fact, it is easy to demonstrate that the other components of surfactant derived from lamellar bodies also have a turnover time of 10 h. For example, the time required to fill the alveolar pool with component ‘a’ if it were empty can be expressed as

where T, is the turnover time of component a, Q, is the alveolar pool size of component a, and f, is the rate at which material flows from the lamellar bodies into the alveolar space. We can express f, as L=Qi.n

(2)

where Qi is the lamellar body pool size of component a and n is the fraction of lamellar bodies which are exocytosed per unit time. Combining Eqns. 1 and 2: 7” =

e, Q:. n

A similar equation can be written for any other component, b, and we can divide T, by Tb: Q, Q:.n T, _ Qb Tb QL.n

(4)

The fraction of lamellar bodies exocytosed per unit time, n, is the same in the numerator and the denominator of Eqn. 4 so that T, _=-.

Pa

1

Tb Qb Q:/Q(,

Q,/Q, is the ratio of the alveolar pool size of component a to that of b, while Q:/Qk is the same ratio in lamellar bodies. Since the composition of lamellar bodies is the same as that of the alveolar wash [1,15,16], Pa

-=--7

Q:

Qb Qb

which means that %=I

or T,=T,

The biological half-life (the time required for the recovered counts/mm to decrease 50%) of any radiolabeled surfactant component would be expected to be 0.69 x turnover time if the component is pulse labeled in the alveolar space and if there is no reentry of that component once it leaves the system [3,6]. Since, in the present study both phosphatidylethanolamine and phosphatidylglycerol pulse labeled the alveolar space, a biological half-life longer than 6.9 h for either phospholipid would imply reutilization of that phospholipid. Both phospholipids had biological halflives longer than 30 h (Table I), indicating that both were reutilized. We can calculate the extent of reutilization, making use of the fact that the turnover time for alveolar surfactant is 10 h. Just as one can calculate the turnover time and the degree of reutilization from an analysis of the total recovered counts/‘min vs. time after intratracheal injection of labeled surfactant, one can predict the total recovered counts/mm vs. time for any combination of turnover time and percent reutilization [6]. If the total recovered counts/mm vs. time after injection could be described by a single exponential, then the time required for the recovered counts/mm to decrease by 50% (the biological half-life) would be independent of the time chosen as the starting point. Similarly, if more than one exponential is required to describe the recovered counts/mm, then this is not the case and the slope of the regression line describing the semilog plot of recovered counts/mm vs. time would depend on the times that the measurements are made. The total recovered counts/mm were obtained at slightly different times in the two experiments. Hence, using a turnover time of 10 h and taking the different times of killing into account, we generated two separate curves (one for each phospholipid), predicting reutilization as a function of the slope of the regression line fit to the semilog plot of total recovered counts/mm (Fig. 4). The predicted values are then empirically fit by polynomial regression. The percent reutilization for the measured slope is read off the curve or calculated from the associated equation. To make these two experiments compatible with each other as well as with the previous study [6] which defined the reutilization of phosphati-

178

Fig. 4. Reutilization of phosphatidylglycerol and phosphatidylethanolamine as a function of the slope of the semilog plot of total alveolar :ounts/min vs. time after injection. (A) Total recovered phosphatidylglycerol counts/mm in the alveolar wash was predicted for each time of killing for reutilization from 0 to 100%. The slope of the semilog plot for each group of predicted recoveries for a given % reutilization was determined and was plotted on the x-axis against the % reutilization on the y-axis. Note that the values on the x-axis are from 0 to -0.1. The included curve was determined by non-linear regression (see text). The slope of the semilog plot of measured counts/mm recovered in the alveolar wash vs. time was determined and the % reutilization was obtained directly from the associated equation. (B) Same as Part A, but for phosphatidylethanolamine.

we determined the recovered counts/mm of phosphatidylglycerol and phosphatidylethanolamine relative to one set of phosphatidylcholine data. We used the equation given in the previous study to predict the total recovered counts/mm of phosphatidylcholine at each time of killing used in the phosphatidylglycerol experiment in the present study. These predicted total phosphatidylcholine counts/mm were multiplied by the mean ratio of phosphatidylglycerol to phosphatidylcholine counts/mm measured at those times to give the total phosphatidylglycerol counts/mm relative to the original phosphatidylcholine data. Similar calculations were made for rabbits given phosphatidylethanolamine. This normalization procedure eliminated any small variations between the groups. We determined the slopes of the regression lines fit to the semilog plots of the total phosphatidylglycerol and phosphatidylethanolamine counts/mm and used these slopes to calculate the percent reutilization of each phospholipid. We found that while phosphatidylcholine was reutilized with 95% efficiency, phosphatidylglycerol

and phosphatidylethanolamine were reutilized with 79 + 8% and 79 f 9% efficiency, respectively. When reutilization of these two lipids was determined from the uncorrected total recovered counts/mm, the values were statistically similar to those given above. Because they are normalized. the values given above are more useful for comparative purposes. Thus, we have demonstrated that the pulmonary surfactant system of 3-day-old rabbits does not metabolize all components of surfactant at the same rate. Specifically, phosphatidylglycerol and phosphatidylethanolamine are reutilized, but at a rate significantly less than that for phosphatidylcholine. Phosphatidylcholine alone cannot duplicate the functions of surfactant in vivo [18]. Which of the other component(s) are necessary for normal surfactant function is as yet unknown. Nevertheless, results reported here support the concept that proper repackaging, possibly with replenishment of certain components of surfactant, is important to maintain normal function. Acknowledgements This work was supported by NIH Grant HD11932 from Child Health and Development, Department of Health and Human Services, by Research Career Development Award HD/ HL-00252 to A.J., and NIH Research Service Award HL06544 to H.J. References 1 Baritusso, A.G., Magoon, M.W., Goerke, J. and Clements. J.A. (1981) B&him. Biophys. Acta 666, 382-393 2 Young, S.L., Kremers, S.A., Apple, J.S., Crapo, J.D. and Brumley, G.W. (1981) J. Appl. Physiol. 51, 248-253 3 Jacobs, H., Jobe, A., Ikegami, M., Jones, S. (1982) J. Biol. Chem. 257, 1805-1810 4 Hallman, M., Epstein, B.L. and Cluck, L. (1981) J. Clin. Invest. 68, 742-751 5 Kuhn, C. (1968) Am. J. Pathol. 53, 809-833 D. 6 Jacobs, H.C., Jobe, A., Ikegami, M. and Conaway, (1983) J. Biol. Chem. 258, 4159-4165 7 Jacobs, H.C., Jobe, A., Ikegami, M., Miller, D. and Jones, S. (1984) B&him. Biophys. Acta 793, 300-309 8 Oyarzun, M.J., Clements, J.A. and Baritussio, B. (1980) Am. Rev. Resp. Dis. 121, 709-721 9 Mason, R.J., Nellenbogen, J. and Clements, J.A. (1976) J. Lipid Res. 17, 281-284 10 Dawson, R.M. (1967) B&hem. J. 102, 205-210

179 11 Jobe, A.H. (1977) B&him. Biophys. Acta 489, 440-453 12 Folch, J., Lees, M. and Sloane-Stanley, G.H. (1957) J. Biol. Chem. 226, 497-509 13 Williams, M.C. (1983) Am. Rev. Resp. Dis. 127, 271 14 Kuhn, C. (1968) Am. J. Pathol. 53, 809-833 15 Hallman, M. and Gluck, L. (1975) B&him. Biophys. Acta 409. 172-191

16 Jobe, A.H., Ikegami, M., Sarton-Miller, I., Yu, G. and Jones, S. (1981) B&him. Biophys. Acta 666, 47-57 17 Jacobs, H.C., Jobe, A.H., Ikegami, M., Jones, S. and Miller, D. (1983) B&him. Biophys. Acta 752, 178-181 18 Chu, J., Clements, J.A., Cotton, E.K., Klaus, M.H., Sweet, A.Y. and Tooley, W.H. (1967) Pediatrics 40, 709-782