Pulmonary SP-A enhances adsorption and appears to induce surface sorting of lipid extract surfactant SAMUEL STANLEY

SCHURCH, FRED POSSMAYER, CHENG, AND AMANDA M. COCKSHUTT

Departments of Medical Physiology and Medicine, The University of Calgary, Alberta T2N 4Nl; and the Departments of Biochemistry and Obstetrics and Gynaecology, The University of Western Ontario, London, Ontario N6A 5A5, Canada Schurch, Samuel, Fred Possmayer, Stanley Cheng, and Amanda M. Cockshutt. Pulmonary SP-A enhances adsorption and appears to induce surface sorting of lipid extract surfactant. Am. J. Physiol. 263 (Lung Cell. Mol. Physiol. 7): L210-L218, 1992.-The effect of surfactant concentration and supplementation with surfactant-associated protein A (SP-A) on the surface activity of lipid extract surfactant (LES) was examined using a captive bubble technique. Adsorption of LES is strongly concentration dependent over the range of 50-1,000 pg/ml. Addition of SP-A to LES at low concentrations in the presence of calcium dramatically increases the rate of adsorption. In quasistatic cycling experiments, samples containing SP-A require less compression to achieve low surface tensions even during the first compression cycle. Calculated film compressibilities at 15 mN/m indicate that SP-A alters the surfactant monolayer behavior such that in a small number of cycles the compressibility is indistinguishable from that of pure DPPC. Furthermore, SP-A reduces the incidence of bubble “clicking,” suggesting a stabilization of the monolayer at low surface tensions. In dynamic cycling experiments, SP-A reduces compression of the film area required to achieve a low surface tension of -1 mN/m. SP-A eliminated the plateau just below 25 mN/m normally observed during the compression phase with low concentrations of LES and the shoulder observed at -35 mN/m during expansion. In the presence of SP-A and, to a lesser extent with high concentrations of LES, there is a marked lag in the increase in surface tension during the initial part of the dynamic expansion loop, with surface tensions remaining near 1 mN/m for -10% of the increase in bubble area. The results indicate that SP-A enhances phospholipid adsorption during dynamic cycling and may enhance elimination of nonDPPC lipids during cycling. The absence of a rapid increase in surface tension during the initial reexpansion suggests the formation of a multilayer configuration of unknown structure, which can maintain very low surface tensions during the initial expansion phase. Finally, experiments with EDTA imply that the SP-A mediated effects are not absolutely dependent on but are greatly enhanced by calcium. surface tension; surfactant tidylcholine; film stability

monolayer;

dipalmitoylphospha-

SURFACTANT is a complex mixture of lipids and proteins secreted from the alveolar type II epithelial cell, which reduces the surface tension at the air-liquid interface of the alveolus. Despite the complex nature of the mixture, it is thought that a monomolecular layer enriched in the saturated phospholipid dipalmitoylphosphatidylcholine (DPPC) is responsible for the reduction of surface tension to very low values on expiration (6, IO, 22). Other surfactant components, which include unsaturated and negatively charged phospholipids and the surfactant-associated proteins, appear to be required to generate and maintain this monolayer ‘probably highly enriched in DPPC. Although surfactant composition varies depending on PULMONARY

L210

lOdO-0605/92 $2.00 Copyright

0

the species, the method of isolation, and the purification procedure, all surfactants contain large proportions of phosphatidylcholine and a relatively large amount of an anionic phospholipid, either phosphatidylglycerol or phosphatidylinositol (30, 37). Preparations of natural surfactant also contain at least three surfactant-associated proteins denoted SP-A, SP-B, and SP-C (24, 35). Clarification of the roles of the various surfactant components in surface tension reduction has been the subject of intense research. These experiments have demonstrated that the low-molecular-weight, hydrophobic proteins SP-B and SP-C substantially increase surface activity, regardless of the assay system used (13,23,39). SP-C is important for rapid adsorption (39), whereas SP-B promotes both adsorption and the removal of phosphatidylglycerol from the monolayer (39,40). SP-A has little effect on its own, but enhances adsorption in conjunction with SP-B (13, 23, 40). Measurements of surface activity can be made in systems that incorporate the lungs of animals, either in situ or excised (25). Although these experiments may give an accurate picture of the physiological situation, they are often too complex to elaborate precise data on molecular mechanisms. Many researchers have turned to in vitro techniques to address these questions. A wealth of information has been obtained using the LangmuirWilhelmy surface balance and the pulsating bubble surfactometer. In these systems, an air-liquid interface is generated, either on the surface of a trough or a bubble in a suspension, and adsorption of surfactant lipids and surface tension during dynamic cycling can be measured (9,25). Although these systems mimic many of the characteristics of the lung, large area compressions are required to obtain low surface tensions. Furthermore, these monolayers are much less stable than those found in the lung, in that the surface tension will rise quickly when a compressed monolayer is held at a fixed area (10, 29). This is presumably due to film collapse and/or leakage, which results in a loss of material from the monolayer (10). Schurch et al. (29) recently described a new in vitro method for the assessment of the surface activity of films adsorbed from aqueous solutions. In this captive bubble technique, a bubble of diameter up to 8 mm is analyzed in a leak-proof chamber. Surfactants examined in this manner mimic closely the in situ behavior observed in the lung. In particular, the surface films are very stable, even at very low surface tensions, and relatively little compression (reduction of surface area) is required to reduce the surface tension to 1 mN/m. It was suggested by Clements (6) that the stability and

1992 the American

Physiological

Society

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

SP-A

ENHANCES

ADSORPTION

low surface tensions observed in the lung were due to the exclusion of components other than DPPC from the alveolar lining during expiration. This was supported by the finding that monolayers of pure DPPC spread from organic solvents are far more stable than those of pulmonary surfactant (11, 15, 22). Experiments with pulmonary surfactant and model mixtures of surfactant containing DPPC and phosphatidylglycerol, using a variety of techniques, strongly support the view that non-DPPC lipids can be removed from the monolayer by mechanical compression, leaving a film which is enriched in DPPC and consequently is capable of withstanding high surface pressures (2,5,8). Furthermore, it was demonstrated using the pulsating bubble surfactometer that repeated cycling was required to obtain low surface tension on compression, suggesting that the process of repeated expansion and compression may result in an alteration of monolayer composition and/or configuration (7). Quasi-static cycling with the captive bubble has also demonstrated that considerably less reduction of surface area is required to achieve surface tensions 30 mN/m. A rationale and discussionof the useof this method for surface tension calculation is given elsewhere(29). Bubble urea. The formula of Malcolm and Elliot (21) allowed us to calculate bubble surface tension but not surfacearea. We therefore useda similar approach to calculate area from maximum bubble diameter, D, and a polynomial in H/D, the ratio of maximum bubble height to diameter (W. M. Schoel, unpublished observations).Briefly, we derived the coefficients for the area polynomial by fitting it to areasobtained from figures of revolution of the digitized contours of over 100 test bubbles formed from the suspensionsof 200 pg/ml LES. We selected these test bubblesto cover a wide range of both H/D (0.1-0.9)

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

L212

SP-A

ENHANCES

ADSORPTION

and diameter (2-8 mm) and substituted a graded series of precision ball bearings for the bubble to calibrate the system. The resulting function was found to yield bubble areas that were valid for the whole range of sizes and shapes tested. The calculated values agreed to within 1% with those obtained by the method of Rotenberg et al. (26) as discussed in our previous article (29). Adsorption measurements. To measurethe adsorption of surfactant onto the monolayer a small atmospheric bubble (2-3 mm diam) wasrapidly expandedto a maximum sizeof 7-8 mm, resulting in an -lo-fold increasein bubble surfacearea in an interval of 0.3 t 0.05 s (meant 1 SE, n = 10). The interval of 0.3 s was determined by counting the video frames that appeared on the television screen during the expansion period. The chambercontents were stirred during the adsorption period by a small magnetic bar. The bubble shapeis recorded continuously on a video recorder and surfacetension is calculatedasin

AND

SURFACE

-70

SORTING

A

5 Z

E 60 V

z v, 50 Z W

40 e 2 CK 30

-

L

I

0

5

I

I

10 ADSORPTION

15 TIME

I

I

1

20 (min )

25

30

60

5

Z

E 50

Calculation of surface tension. Quasi-static isotherms. A small atmosphericbubble was rapidly expanded as describedin Adsorption measurements. After

fs z 40 E I-

adsorption to -25 mN/m the film was compressedstepwise, waiting at each step until the bubble shapeno longer changed noticeably within 20-30 s (this correspondsto a changein surface tension of ~0.5 mN/m in that period). Between 10 and 15 stepswere taken for eachcompressionand expansionpart of the cycle. Dynamic hysteresis loops. A bubble is formed and cycled dynamically at 20 cycles per minute (cpm). The compressionresults in a reduction of surfacearea to -20% of the original area for the low concentration (200 pg/ml) LES and to 40-50% for the higher concentrations (1,000hg/ml) of LES and the samples that contained SP-A. The bubble is recorded continuously throughout the cycling process. Calculation of film compressibility. The compressibility of a film is a useful parameter for the interpretation of compression data. The film compressibility at a surfacetension of 15 mN/m is reported here for comparisonto pure DPPC films. It is calculated from the equation, Cl5 = (l/A)(dA/d$, where A is the relative area from the start of compressionand y is the surface tension. This value is obtained from the isotherm by determining the relative area at 15 mN/m and the slopeof the tangent to the curve at this surfacetension. The Cl5 for a pure DPPC film is 0.005 (mN/m)-l calculated from previous data (IO, 29).

w 30 0 i2 5 20 cn

-

l

I

0

5

I

I

10 ADSORPTION

15 TIME

I

I

J

20

25

30

(set

>

Fig. 1. Time courses for adsorption of lipid extract surfactant (LES) different concentrations. A: surface tension (mN/m) is plotted vs. sorption time (min) for LES at 50 pg/ml. Samples are suspended 0.9% NaCl, 1.5 m&I CaC12. B: surface tension (mN/m) is plotted adsorption time (s) for LES 100 (a), 200 (*), 400 (m), 800 (A), 1,000 (A) pg/ml. Samples suspended as in A. Values are means k n = 4. Note different scales of axes in A and B (minutes as opposed seconds).

at adin vs. and SE, to

RESULTS

The captive bubble technique was used to examine the concentration dependence of the adsorption of surfactant lipids onto the air-liquid interface. Figure 1 shows the time course of adsorption of LES at different concentrations after a rapid expansion of surface area. The process is strongly concentration dependent, such that samples at 50 pg/ml require more than 30 min to reach equilibrium, whereas samples at 800 pg/ml or greater require ~5 s to reach equilibrium surface tensions of ~25 mN/m. The initial, or first measurable, surface tension value following bubble expansion, is also an indicator of adsorption rate, since it reflects the amount of adsorption that occurs during the expansion period of ~0.3 s. Suspensions with very slow adsorption rates will have initial surface tension values approaching 70 mN/m, the surface tension of water at 37OC. Indeed, LES at 50 pg/ml has initial values >60 mN/m. Addition of SP-A to LES at low concentrations in the presence of calcium results in a dramatic increase in the rate of surfactant lipid adsorption. Figure 2 demon-

I

I

I

0

1

2 ADSORPTION

Fig. 2. Time courses for pg/ml supplemented with 4.0%(A). Surface tension Samples are suspended in t SE, n = 4.

I

3 TIME (set)

I

1

4

5

adsorption of lipid extract surfactant at 200 different amounts of SP-A, 0.5(o), 1.0(e), and (mN/m) is plotted vs. adsorption time (s). 0.9% NaCl, 1.5 mM CaCl,. Values are means

strates the effect of addition of increasing amounts of SP-A to LES at 200 pg/ml (expressed as % wt/wt of the surfactant lipid concentration, such that 1.0% SP-A = 2 pg/ml in this case). In the absence of SP-A, adsorption to equilibrium takes -20 s and initial surface tensions >45 mN/m are observed (see Fig. 1B). With the addition of 0.5% SP-A the adsorption period is decreased to ~3-5 s, and initial surface tensions ~35 mN/m are observed. The addition of larger amount of SP-A, 1.0 and 4.0%, leads to adsorption to equilibrium within 1 s of bubble expansion. At lower phospholipid concentrations, (Fig. 3,

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

SP-A ENHANCES

1

I 1

0

I”

0

‘1

20

ADSORPTION

1

1

1

1

2 3 ADSORPTION

4

5

6

I

‘1

40 60 ADSORPTION

TIME

SORTING

L213

(set)

‘1

1’

80 TIME

AND SURFACE

(set

’

100 >

1

120

Fig. 3. Time course for adsorption of lipid extract surfactant of concentration of 100 pg/ml (open symbols) and at 50 pg/ml (filled symbols). Top: 4% SP-A added; bottom: 1% SP-A added. Values are means + SE, n = 4.

top), with

the addition of 4% SP-A, adsorption to takes 3 s for 100 pg/ml and 5 s for 50 pg/ml. With the addition of 1% SP-A the adsorption time is -10 s for 100 pg/ml and 120 s for 50 pg/ml (Fig. 3,

Relative Area Fig. 4. Quasistatic isotherms of lipid extract surfactant (LES). Surface tension (mN/m) is plotted vs. relative area for LES at 200 pg/ml. Open symbols, measurements made during compression; filled symbols, measurements made during expansion. Samples are suspended in 0.9% NaCl, 1.5 mM CaCl,. Values are means & SE, n = 4. A: first cycle, B, second cycle; C: fourth quasistatic cycle. Gap between last compression value and first expansion value represents a bubble “click,” indicated with 1,2.

25 mN/m bottom).

After adsorption of the surfactant to equilibrium, the bubble can be cycled in a quasistatic fashion. The isotherms generated from such an experiment with a LES concentration of 200 pg/ml are shown in Fig. 4. Clearly, the compression and expansion parts of the cycle do not overlap, i.e., there is hysteresis. As can be noted on these isotherms, there is a “click” observed at low surface tensions, which results in a spontaneous decrease in surface area and concomitant increase in surface tension [an indepth discussion of the clicking phenomenon is given elsewhere (29)]. Sequential isotherms of this nature (Fig. 4A is the first cycle, Fig. 4B is the second, and Fig. 4C is the fourth) result in an increasing ability of the bubble to attain low surface tensions with smaller reductions in surface area. As well, there is much less hysteresis in the fourth cycle than in the first. The addition of SP-A changes the appearance of these isotherms as can be seen in Fig. 5. When 0.5% SP-A is added (Fig. 5, A, B, and C), less hysteresis is observed in

the first cycle and compression of only 21% instead of 29% results in surface tensions of 1 mN/m. In this case clicking was observed only in the first cycle. By the fourth cycle a compression of 14% results in these very low surface tensions. Higher concentrations of SP-A, such as 1.0% (Fig. 5, D, E,and F), result in further enhancement of surface activity such that clicking is eliminated and 12% compression results in low surface tensions in the fourth cycle. Increasing the SP-A concentration to 4.0% did not lead to further enhancement for phospolipid concentrations at and above 200 pg/ml (data not shown). The results of the quasistatic isotherms are summarized in Table 1, and the calculated film compressibility at 15 mN/m is included. Bubbles formed in the captive bubble apparatus can also be cycled dynamically (at 20 cycles/min). The results of such experiments are shown in Fig. 6, where the data from four consecutive cycles centering on the 20th cycle is plotted. When LES at 200 pg/ml is cycled in this fashion (Fig. 6A) several interesting features are observed. Compressions of -80% of the surface area are required to

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

L214

SP-A ENHANCES

ADSORPTION

AND SURFACE

SORTING

Fig. 5. Quasi-static isotherms of lipid extract surfactant supplemented with 2 different concentrations of SP-A. Surface tension (mN/m) is plotted vs. relative area for lipid extract surfactant (LES) at 200 pg/ml. Samples are suspended in 0.9% NaCl, 1.5 mM CaCl,. Open symbols, measurements made during compression; filled symbols, measurements made during expansion. Values are means t SE, n = 4. A, B, and C represent addition of 0.5% SP-A on first, second, and fourth cycles, respectively. D, E, and F represent addition of 1.0% SP-A on first, second, and fourth cycles, respectively. 252015105I Ok+

Rela

twe

’ 0 60

Area

070

Rela

of LES rf: SP-A Cycle

max, mN/m

ymin, mN/m

AA, %

I 090

I 100

I 1 10

t we Area

Table 1. Summary of quasi-static isotherms SEA, %

a

1 080

C (rnN;L)-l

0

First 26.0 2.0 28.7 0.015 Second 26.0 1.8 21.5 0.010 Fourth 26.0 1.0 17.5 0.008 0.5 First 25.0 1.0 20.7 0.015 Second 25.0 1.0 18.0 0.007 Fourth 25.0 1.0 14.5 0.006 1.0 First 25.0 1.0 21.5 0.009 Second 25.0 1.0 13.7 0.006 Fourth 25.0 1.0 12.0 0.005 Values are means, n = 4. LES, lipid extract surfactant. C1, is the film compressibility at y of 15 mN/m = (l/A)(dA/d$, Cl5 for pure diphosphatidylcholine film is 0.005 (mN/m)+ AA, total area change.

reach low surface tensions. Even after 20 cycles compressibility at 15 mN/m is 0.042 (mN/m)-l, which is almost 10 times the value observed for pure DPPC. In the compression segment of the loop a large plateau just below 25 mN/m is observed. Similarly, in the expansion part of the cycle, at a surface tension of -35 mN/m, a shoulder is present. This shoulder may represent the adsorption of surfactant lipids. At higher surfactant concentrations the loops have a very different shape (Fig. 6B). At 1,000 pg/ml LES the plateau and the expansion shoulder essentially disappear. Furthermore, a compression of only 32% is required to achieve low surface tensions. It should also be noted that the surface tension at maximum bubble area is ~35 mN/m compared with the -45 mN/m observed at the lower surfactant concentration. Film compressibility at 15 mN/m is 0.013 (mN/m)? LES at 200 pg/ml supplemented with 4.0% SP-A (Fig. 6C) behaves similarly to

the higher surfactant concentration (Fig. 6B). In this case no plateau or shoulder is observed. Maximum surface tensions of ~35 mN/m are observed and compression of ~30% is required to obtain low surface tensions. Compressibilities of these samples at 15 mN/m are 0.011 (mN/m)-I. Interestingly, in these samples, and to a lesser extent with high concentrations of LES in the absence of SP-A, the shape of the expansion segment of the loop at low surface tension is less steep. The surface tension remains below 1 mN/m for a short period as the area begins to increase on expansion. This is in noticeable contrast to what is observed in the quasistatic experiments (compare Fig. 6C to the isotherms in Fig. 5). Previous experiments have demonstrated that the effects of SP-A depend on the presence of calcium for optimal activity (3, 7, 13, 16). Adsorption and cycling experiments were performed in the presence of EDTA to assess the role of calcium in this system. Figure 7 shows the effect of incubation of samples with EDTA (12 h at 37°C) on adsorption time courses. LES alone adsorbs much slower in the absence of calcium. Addition of SP-A does increase the rate of adsorption; however, times of 3 min compared with 1 s are required to achieve equilibrium surface tension at 4.0% added SP-A. Incubation with EDTA also altered the behavior of the samples during quasistatic and dynamic cycling. During dynamic cycling even with 4.0% SP-A a large plateau is observed and compression of ~50% is required to reduce the surface tension to low values (data not shown). The results of quasistatic cycling of LES at 200 pg/ml are summarized in Table 2. In the presence of EDTA, LES alone does not reduce the surface tension below 18 mN/m, even with compression of 80% of the surface area. Addition of SP-A does result in the achievement of lower surface tensions,

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

SP-A ENHANCES 50

0 0

0

O0

Incubation

:.v

RO

0

30

L215

SORTING

0”

0

-0 l o

40

AND SURFACE

Table 2. Effect of EDTA on quasi-static cycling of LES

00 .o.aJ Ofi

i A -

SP-A, %

ymax, mN/m

ymin, mN/m

AA, %

AAz5, %

C&,, (mN/m)-l

00

0 m

I 20

ADSORPTION

0.

1

l

t:.

1.5 mM EDTA

O0 O@” Ood,

ooO~Ooo

0 1.0 4.0

Controls 1.5 mM CaC& 1.5 mM 3 mM

0 1.0 1.0

46.7 32.5 25.2

18.0 3.0 2.5

67.5 44.8

79.8 69.3 62.5 44.8

0.023

26.3 24.6

3.6 1.7 2.3

43.5 32.5 32.1

40.6 32.5 32.1

0.028 0.022 0.014

0.009

EDTA + 24.9 CaCl, next day 1.5 mM CaCl, 1.0 25.6 1.5 30.3 30.3 0.013 SP-A added next day Values are for the first quasi-static cycle. LES, lipid extract surfactant. All samples incubated overnight (16-18 h) in buffers indicated. AA, total area change, AAZ5, area change from 25 mN/m to minimum surface tension.

(compare values in Table 1 and Table 2), the effect of EDTA is still readily apparent. DISCUSSION

I

0

02

I

1

1

I

I

1

04

06

08

10

12

14

Relat

we

Area

Fig. 6. Effect of concentration and SP-A on dynamic hysteresis loops of lipid extract surfactant (LES). Surface tension (mN/m) is plotted vs. relative area. Plots are the summaries of 4 consecutive dynamic cycles centering on 20th cycle. Open symbols, measurements made during compression; filled symbols, measurements made during expansion. Samples are suspended in 0.9% NaCl, 1.5 mM CaC12. A: LES at 200 pg/ml. B: LES at 1,000 pg/ml. C: LES at 200 pg/ml supplemented with 4.0% SP-A. 60r A 55 E \ z 50 -1E z

f+5-

0 (7T z

40-

2 w 0 i? cr

3530-

2

25I

I

0

5

I

I

I

I

I

10 15 20 25 30 ADSORPTION TIME ( min )

I

35

1

40

I

I

45

50

Fig. 7. Time courses for adsorption of lipid extract surfactant (LES) in presence and absence of SP-A O(O), l(o), and 4%(a) after overnight incubation in EDTA. Surface tension (mN/m) is plotted vs. adsorption time (min) for LES at 200 fig/ml. Samples are suspended in 0.9% NaCl, 1.5 mM EDTA and incubated overnight in presence of 0.01% sodium azide. Values are means t SE, n = 4.

.however, compressions of 4565% are required, as opposed to -30% when calcium is present and the samples are incubated similarly. Controls are included in this table to demonstrate that although the surfactant is somewhat compromised by the overnight incubation at 37OC

The results of the first series of studies performed using the captive bubble method indicated differences between rabbit natural surfactant and bovine lipid extract surfactant with respect to adsorption kinetics and monolayer stability (29). These data suggested that SP-A was responsible for the enhancement of adsorption and the stabilization of compressed monolayers. In the present investigation these suggestions have been verified using samples of bovine lipid extract surfactant supplemented with SP-A. Results of adsorption studies indicate that in the presence of calcium, SP-A dramatically enhances the adsorption of LES at low surfactant concentrations. Although at and above a phospholipid concentration of 200 pg/ml the addition of more than 1% SP-A does not appear to substantially increase the rate of adsorption, this is different at lower phospholipid concentrations. Increasing the SP-A concentration from 1 to 4% decreases the adsorption time of LES at 50 pg/ml from 120 to 5 s. The strong concentration dependence of adsorption in the absence of SP-A suggests that the process is largely limited by diffusion of the lipids in suspension to the air-liquid interface and the subsequent conversion from bilayer to monolayer configuration. SP-A is therefore behaving as a catalyst for adsorption, somehow overcoming the diffusion and conversion limitations. SP-A causes the aggregation of phospholipids in the presence of calcium (16), therefore it is unlikely that SP-A actually increases the rate of diffusion. A model that involves the restructuring of the lipids in the aqueous subphase leading to facilitated entry onto the monolayer is more likely (3, 16). The addition of the low-molecular-weight surfactant proteins SP-B and SP-C to lipid mixtures containing phosphatidylcholine and phosphatidylglycerol leads to a large increase in the rate of phospholipid adsorption when measured in a surface balance or pulsating bubble surfactometer (13, 39). The addition of SP-A further accelerates the adsorption process, although it has very little effect on its own that is on the adsorption of above lipid mixture (13, 23, 40). The interaction of SP-A with SP-B appears to be particularly important for adsorption

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

L216

SP-A

ENHANCES

ADSORPTION

efficiency. When combined with the recent findings that tubular myelin also requires phosphatidylcholine, phosphatidylglycerol, SP-B, and SP-A (31), the obvious interpretation is that the formation of tubular myelin or a related structure correlates with increased adsorptivity of the surfactant lipids. Such correlation has been made previously in studies by Benson et al. (3). Conclusive evidence for tubular myelin formation in the present system would have to be confirmed by electron microscopy. If the effect of SP-A on LES were confined to a simple enhancement of surfactant lipid adsorption, the area-surface tension characteristics observed during the first compression following adsorption to equilibrium would be expected to be unchanged by the addition of SP-A. Results of quasistatic cycling experiments indicate that the addition of SP-A does in fact alter the behavior of the monolayer on compression. When the first quasistatic cycles are compared in the presence and absence of SP-A, it is clear that SP-A limits the hysteresis area, lowers the amount of compression necessary to achieve low surface tensions, reduces bubble clicking, and leads to lower minimum surface tensions. When the film compressibilities at 15 mN/m are compared it appears as though monolayers formed in the presence of SP-A are already enriched in DPPC before or during the first compression cycle. This implies that SP-A either causes the selective adsorption of DPPC onto the interface or has a compressionindependent mechanism for removal of non-DPPC lipid. A related possibility is that there is a phase separation of non-DPPC lipids such that they are patched or organized in the monolayer so that only a very small compression is required to extrude them from the monolayer. However, since LES is only ~50% DPPC (38), it appears unlikely that a compression of 21% could result in the squeeze-out of all unsaturated phospholipids from the monolayer. The loss of the clicking phenomenon and the achievement of lower surface tensions in the presence of SP-A also indicate a monolayer stabilizing function. Simple enrichment of DPPC would have this result (22). Quasistatic cycling involves a series of small discrete alterations in bubble area where the surface monolayer is allowed to partially “relax” during the compression-expansion processes. With dynamic cycling, where the bubble volume is continuously altered at 20 cpm, there is less opportunity for adsorption during the expansion phase and a greater possibility of overcompression (collapse at minimum tension) during the compression phase. In our case this collapse plateau represents -35% of the overall area change for the hysteresis loops of Fig. 6, B and C. This results in very different isotherms, particularly at low LES concentrations (compare Figs. 4 and 5 with Fig.6). With all conditions studied, dynamic compressions require larger area reductions to achieve low surface tensions than quasistatic compressions after repeated cycles. A number of interesting features can be observed, particularly at low concentrations of LES. During dynamic compression with low LES concentrations (200 pg/ml), a distinct plateau occurs below 25 mN/m where a decrease in surface area of 30% results in only a small reduction in surface tension. Previous studies with the Langmuir-Wilhelmy surface balance have termed this

AND

SURFACE

SORTING

phenomenon the “purification” or “squeeze-out” plateau, since it appears as though non-DPPC lipids are being removed as opposed to the simple packing of monolayer lipids which would result in the reduction of surface tension below the equilibrium surface tension of 25 mN/m (4, 8, 10, 14). It has been suggested that squeeze-out corresponds to the exclusion of the lipid-expanded phase, resulting in a partial enrichment in DPPC (4). This would account for the eventual lowering of the surface tension to low values with large area reductions. The purification plateau disappears on addition of SP-A and at high concentrations of LES. Another feature of the dynamic loops at low concentrations of LES is a shoulder just above 35 mN/m in the expansion portion of the curve. In this phase of the expansion, increases of surface area of IO-20% result in only modest elevations in surface tension after which surface tension increases more rapidly to 45 mN/m. Since adsorption from the bulk suspension is very slow at this lipid concentration, adsorption during this phase must represent adsorption from a distinct pool of lipids that we refer to as the “surface reservoir.” The nature of this surface reservoir is unknown. With higher surfactant lipid concentrations or when SP-A is present, the surface tension increases to 30 mN/m and then remains constant until the bubble attains its maximum volume. This suggests that rapid adsorption masks or eliminates this surface reservoir phenomenon. A particularly interesting feature of the loops performed in the presence of SP-A, and to a lesser extent those performed at high LES concentrations, is the change in slope of the expansion curve of highly compressed monolayers. Rather than the surface tension increasing quickly as expansion begins (as is observed during quasistatic cycling), the surface tension actually remains at 1 mN/m, even though the surface area is increased by as much as 10%. This observation suggests a structure, perhaps a multi-layer of nearly pure DPPC, is formed in this overcompression situation which can expand by unfolding on expansion maintaining near zero surface tension. A similarity can be noted between the dynamic cycling isotherms obtained with high LES concentrations and SP-A supplemented low LES concentrations (Fig. 6, B and C). These curves resemble the corresponding quasistatic loops more closely than the dynamic isotherm obtained with low concentrations of LES (Fig. 6A). Although the mechanisms involved are still not completely understood, the following interpretation attempts to reconcile observations on dynamic cycles with the present and previously reported (28, 29) quasistatic results. The difference between the dynamic isotherms with low and high LES is attributed to the ability of the higher bulk lipid concentration to drive phospholipid insertion into the less densely packed regions of the surface during that portion of the expansion curve where surface tension arises above the equilibrium value. This limits the maximal surface tension during expansion to 30 mN/m as opposed to the 45 mN/m observed with low concentrations of LES. Since no purification plateau is observed

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

SP-A

ENHANCES

ADSORPTION

with high LES, we conclude the actual amounts of phospholipid incorporated into the monolayer during expansion is relatively small so that the non-DPPC components can be readily excluded during the compression phase. Overall compressibility is consistent with a monolayer highly enriched in DPPC during the previous 19 could expansion- compression cycles. These observations also be explained by selective adsorption of DPPC during expansion, but we know of no obvious explanation for such a mechanism. The similarity between the dynamic isotherms generated with low LES plus SP-A and high LES concentrations can be readily explained by SP-A’s ability to enhance adsorption to equilibrium. SP-A could also contribute by promoting selective adsorption of DPPC during expansion or by enhancing selective desorption of non-DPPC components, as concluded for the quasistatic experiments. Clearly, the differences between the quasistatic and dynamic cycling observed with LES at low concentration demonstrate that the nature of the cycling procedure affects the monolayer structures, even when performed with the same apparatus. Much greater compression is requi red during dynamic cycling than quasistatic, implying that overcompression and rapid compression may lead to monolayer collapse or instability causing poor adsorption, such that subsequent cycles continue to require large compressions. These findings may explain why, in addition to leakage, measurements made on the surface balance, the pulsating bubble and the captive bubble differ, since the former two methods involve film area compression s of 50% or more. The dependence of the SP-A-mediated effects on calcium was examined by incubating samples in EDTA before evaluation. Adsorption of LES alone is drastically impaired by EDTA incubation, suggesting that interactions between the surfactant lipids and SP-B and/or SP-C may require calcium for optimization of adsorption efficiency (17,36). Supplementation with SP-A does lead to an enhancement of adsorption even in the presence of EDTA, however, 3 min as opposed to 1 s are required to reach equilibrium. In these experiments the surfactant was less active after incubation at 37°C. This loss of activity might be related to lipid peroxidation, as previously suggested (40), and the possibility of lipid degradation cannot be eliminated. Nevertheless, the effect of EDTA was readily apparent. It appears that some of the effects of SP-A are not absolutely dependent on the presence of calcium, but calcium substantially enhances the SP-A effects. However, it has been demonstrated that SP-A binds calcium with high affinity (12), so the possibility exists that a small amount of calcium may remain associated with the SP-A even after incubation with EDTA. The results of dynamic cycling in the presence of EDTA also imply that the enhanced adsorption due to SP-A is impaired, since a large area compression is required and the purification plateau and adsorption shoulder are still observed (data not shown). Calcium is required for the formation of tubular myelin in vitro (3 1) and is required for the enhancement of surface activity and reversal of blood protein inhibition observed with the

AND

SURFACE

SORTING

L217

pulsating bubble surfactometer (7, 34). If the formation of tubular myelin is considered as the effector through which the actions of SP-A are transmitted, then a mechanism for tubular myelin enhancement of adsorption and monolayer purification is required. Results presented here and elsewhere (17, 39) indicate that organic extracts of surfactant, which do not contain SP-A but do contain SP-B and SP-C, have surface activities essentially indistinguishable from that of surfactants containing SP-A when assayed at high concentrations. Thus it would appear that many of the functions of tubular myelin may be replaced by high bulk phase surfactant concentrations. As the concentration of surfactant is increased, adsorption probably increases as a result of the increased frequency of collisions of vesicles with the interface that result in conversion to a surface film. Purification of a monolayer arising in this fashion may proceed via mechanical squeeze-out during the compression process mediated by the low-molecular-weight hydrophobic proteins. Tubular myelin, on the other hand, is presumed to act even when the bulk lipid concentration is low. Natural surfactant, particularly the large aggregate fraction, is more effective in reducing surface tension to near zero in the pulsating bubble surfactometer at low phospholipid concentrations, e.g., Ref. 36. Benson et al. (3) observed that the ability of natural surfactant to adsorb readily correlates with the presence of tubular myelin. These observations do not prove that it is the tubular myelin which adsorbs more quickly but the results are consistent with this suggestion. Tubular myelin may increase the efficiency of the conversion event which moves bilayer lipids near the air-liquid interface to the monolayer. The latticelike structure typical of tubular myelin could be important for orienting surfactant lipids and proteins such that adsorption is facilitated. When the organization of this structure is considered it is not hard to imagine how DPPC and non-DPPC lipids may be segregated into different regions and become adsorbed differentially. The experiments presented here cannot address these questions, however, it has been demonstrated that SP-A can bind specifically to DPPC (18) and SP-B may have a selective association with phosphatidylglycerol (1, 40). In the present investigation we have demonstrated that SP-A has the capacity, when combined with SP-B, SP-C, and surfactant lipids of accelerating the adsorption process and likely enriching this monolayer in DPPC, thereby reducing compression requirements and stabilizing the monolayer. When the various functions of SP-A are considered it becomes clear that this protein is performing an essential function when surfactant concentrations are limiting, such as in neonatal respiratory distress syndrome or when the surfactant is compromised by the presence of inhibitory agents (7,34). As well, SP-A maintains appropriate alveolar levels of surfactant by controlling secretion from and uptake into type II cells (37). SP-A also counteracts infection of the lung by pathogens by activating alveolar macrophages (32, 33). All of these roles of SP-A depict it as a multifunctional protein, which maintains the homeostasis of the alveolus.

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.

L218

SP-A ENHANCES

ADSORPTION

The authors thank Drs. Shou-Hwa Yu and Michael Schoel for helpful discussions and Mary Ann Ormseth and Karl Babl for their excellent technical assistance. We also thank Cornell’s Abattoir for their continuing support. This work was supported by the Medical Research Council of Canada and the Alberta Heritage Foundation for Medical Research. A. M. Cockshutt is the recipient of a Medical Research Council Studentship and S. Schurch is a Scholar of the Alberta Heritage Foundation for Medical Research. Address for reprint requests: S. Schurch, Dept. of Medical Physiology, The University of Calgary Health Sciences Centre, 3330 Hospital Drive N.W., Calgary, Alberta T2N 4N1, Canada. Received 3 September 1991; accepted in final form 20 March 1992. REFERENCES 1. Baatz, J. E., B. Elledge, and J. A. Witsett. Surfactant protein SP-B induces ordering at the surface of model membrane bilayers. Biochemisty 2.

29: 6714-6720,

1990.

Bangham, A. D., C. J. Morley, and M. C. Phillips. physical properties of an effective lung surfactant. Biochim. phys.

Acta

573: 552-556,

The Bio-

1979.

Benson, B. J., M. C. Williams, K. Sueishi, J. Goerke, and T. Sargeant. Role of calcium ions in the structure and function of pulmonary surfactant. Biochim. Biophys. Acta 793: 18-27,1984. 4. Boonman, A., F. H. J. Machiels, A. F. M. Snik, and 9. Egberts. Squeeze out from mixed monolayers of dipalmitoyl phosphatidylcholine and egg phosphatidylglycerol. J. Colloid In3.

terface 5.

Sci.

120: 456-468,

6.

Clements,

J. A. Functions

Respir.

115: 67-71,

7.

Cockshutt, A. M., J. I. Weitz, and F. Possmayer. Pulmonary surfactant-associated protein A enhances the surface activity of lipid extract surfactant and reverses inhibition by blood proteins in vitro. Biochemistry 29: 8424-8429, 1990. Egberts, J., H. Sloot, and A. Mazure. Minimal surface tension, squeeze-out and transition temperatures of binary mixtures of dipalmitoylphosphatidylcholine and unsaturated phospholipids. Biochim. Biophys. Acta 1002: 109-113, 1989. Enhorning, G. A pulsating bubble technique for evaluating pulmonary surfactant. J. Appl. Physiol. 43: 198-201, 1977. Goerke, J., and J. A. Clements. Alveolar surface tension and lung surfactant. In: Handbook of Physiology. The Respiratory System. Mechanics of Breathing. Bethesda, MD: Am. Physiol. Sot., 1986, sect. 3, vol. III, part 1, chapt. 16, p. 247-261. Goerke, J., and J. Gonzales. Temperature dependence of dipalmitoyl phosphatidylcholine monolayer stability. J. Appl. Phys-

8.

9. 10.

11.

iol. 51: 1108-1114,

of the alveolar lining.

Am.

J. Biol.

Eng.

58: 151-153,

1980.

273, 1990.

Possmayer, F. A proposed nomenclature for pulmonary surfactant-associated proteins. Am. Reu. Respir. Dis. 138: 990-998,1988. 25. Robertson, B., and B. Lachmann. Experimental evaluation of surfactants for replacement therapy. Exp. Lung Res. 14: 279-310, 1988. 26.

Rotenberg, Y. L., L. Boruvka, and A. W. Neumann. Determination of surface tension and contact angle from the shapes of axisymmetric fluid interfaces. J. Co&id Interface Sci. 93: 169-183,

27.

Rouser, G., S. Fleischer, and A. Yamamoto. Two dimensional thin layer chromatographic separation of polar lipids and determination of phospholipids by phosphorus analysis. Lipids 5:

28.

Schurch, S., H. Bachofen, properties of rat pulmonary bubble method: adsorption,

1983.

494-496,

phys.

Acta

1970.

1103:

127-136,

J. Goerke, and F. Green. Surface surfactant studied with the captive hysteresis, stability. Biochim. Bio1992.

Schurch, S., H. Bachofen, J. Goerke, and F. Possmayer. A captive bubble method reproduces the in situ behaviour of lung surfactant monolayers. J. Appl. Physiol. 67: 2389-2396, 1989. 30. Shelley, A., J. E. Paciga, and J. U. Balis. Lung surfactant phospholipids in different animal species. Lipids 19: 857-862, 29.

1984. 31.

32.

33.

1981. 34.

35.

Suzuki, Y., Y. Fujita, and K. Kogishi. Reconstitution of tubular myelin from synthetic lipids and proteins associated with pig pulmonary surfactant. Am. Rev. Respir. Dis. 140: 75-81, 1989. Tenner, A. J., S. L. Robinson, J. Borchelt, and J. R. Wright. Human pulmonary surfactant protein (SP-A), a protein structurally homologous to Clq, can enhance FcR- and CRl-mediated phagocytosis. J. Biol. Chem. 264: 13923-13928, 1989. Van Iwaarden, F., B. Welmers, J. Verhoef, H, P. Haagsman, and L. M. G. van Golde. Pulmonary surfactant protein A enhances the host-defense mechanism of rat alveolar macrophages. Am. J. Respir. Cell Mol. Biol. 2: 91-98, 1990. Venkitaraman, A. R., S. B. Hall, J. A. Whitsett, and R. H. Notter. Enhancement of biophysical activity of lung surfactant extracts and phospholipid-apoprotein mixtures by surfactant protein A. Chem. Phys. Lipids 56: 185-194, 1990. Weaver, T. E., and J. A. Whitsett. Function and regulation of expression of pulmonary surfactant-associated proteins. Biochem. J. 273: 249-264,

36. 37. 38.

611, 1979.

King, R. J. Lipid-apolipoprotein interactions in surfactant studied by reassembly. Exp. Lung Res. 6: 237-253, 1984. 17. Kobayashi, T., and B. Robertson. Surface adsorption of pulmonary surfactant in relation to bulk-phase concentration and presence of calcium. Respiration 44: 63-70, 1983. Y., and T. Akino. Pulmonary surfactant protein A 18. Kuroki, (SP-A) specifically binds dipalmitoylphosphatidylcholine. J. Biol.

1951.

24.

375, 1986.

16.

193: 265-275,

l

1977.

Hildebran, J. N., J. Goerke, and J. A. Clements. Pulmonary surface film stability and composition. J. Appl. Physiol. 47: 604-

Chem.

J. Chem.

12.

15.

1991.

22 Notter, R. H. Surface chemistry of pulmonary surfactant: the role of individual components. In: Pulmonary Surfactant, edited by B. Robertson, L. M. G. van Golde, and J. J. Batenburg. New York: Elsevier, 1984, p. 17-65. 23. Pison, U., K. Shiffer, S. Hawgood, and J. Goerke. Effects of the surfactant-associated proteins, SP-A, SP-B, and SP-C, on phospholipid surface film formation. Prog. Respir. Res. 25: 271-

Reu.

Haagsman, H. P., T. Sargeant, P. V. Hauschka, B. J. Benson, and S. Hawgood. Binding of calcium to SP-A, a surfactant-associated protein. Biochemistry 29: 8894-8900, 1990. D. Dann, 13. Hawgood, S., B. J. Benson, J. Schilling, J. A. Clements, and R. T. White. Nucleotide and amino acid sequences of pulmonary surfactant protein SP 18, and evidence for cooperation between SP 18 and SP 28-36 in surfactant lipid adsorption. Proc. Natl. Acad. Sci. USA 84: 66-70, 1987. 14. Keough K. M. W., and H. W. Taeusch. Surface balance and differential scanning calorimetric studies on aqueous dispersion of mistures of dipalmitoyl phosphatidylcholine and short drain, saturated phospatidylcholoines. J. Colloid Interface Sci. 109: 364-

266: 3068-3073,

21. Malcolm, J. D., and C. D. Elliott. Interfacial tension from height and diameter of a single profile drop or captive bubble. Can.

and E. I. Franses. Surface interfaces. Langmuir 6: 1647-

1990. Dis.

Chem.

19. Laemmli, U. Cleavage of structural proteins during the assembly of the head of bacteriophage T.+ Nature Lond. 227: 680-685,197O. 20. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall. Protein measurement with the Folin phenol reagent.

1987.

Chung, J. B., R. E. Hannemann, analysis of lipid layers at air/water 1655,

AND SURFACE SORTING

39.

40.

1991.

Weber, M. J., and F. Possmayer. Calcium interactions in pulmonary surfactant. Biochim. Biophys. Acta 796: 198-203, 1984. Wright, J. R., and J. A. elements. Metabolism and turnover of lung surfactant. Am. Rev. Respir. Dis. 135: 426-444, 1987. Yu, S.-H., P. G . R. Harding, N. Smith, and F. Possmayer. Bovine pulmonary surfactant: chemical composition and physical properties. Lipids 18: 522-529, 1983. Yu, S.-H., and F. Possmayer. Comparative studies on the biophysical activities the low-molecular-weight hydrophobic proteins purified from bovine pulmonary surfactant. Biochim. Biophys. Acta 961: 337-350, 1988. Yu, S.-H., and F. Possmayer. Role of bovine pulmonary surfactant-associated proteins in the surface active property of phospholipid mixtures. Biochim. Biophys. Acta 1046: 233-241, 1990.

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (163.015.154.053) on August 8, 2018. Copyright © 1992 American Physiological Society. All rights reserved.