Surfactant displaces particles toward the epithelium in airways and alveoli

Respiration Physiology, 80 (1990) 17-32

17

Elsevier RESP 01642

Surfactant displaces particles toward the epithelium in airways and alveoli Samuel Schflrch 1, Peter Gehr 2, Vinzenz Im H o f 3, Marianne Geiser 2 and Francis Green 1 IRespiratory Research Group, The University of Calgary, Calgary, Alberta, Canada, 21nstitute of Anatomy, University of Berne, Berne, Switzerland and 3Respiratory Unit, Department of Medicine, lnselspital, University of Berne, Berne, Switzerland (Accepted 26 January 1990) Abstract. This study was designed to investigate the early stages of particle deposition on airway and

alveolar surfaces. To do this we used morphometric studies of aerosol deposition, in situ measurements of surface tension, and in vitro assays of particle displacement and mathematical modelling. We observed that latex particles, equal or less than 6 #m in diameter deposited in hamster lungs were submerged in the subphase of the alveolar lining layer and became completely coated with an osmiophilic film. Similar results were obtained for particles deposited in the conductive airways which were also covered with a surface active film, having a surface tension of 32 _+2 dyn' cm - 1. In vitro experiments showed that pulmonary surfactant promotes the displacement of particles from air to the aqueous phase and that the extent of particle immersion depends on the surface tension of the surface active film. The lower the surface tension the greater is the immersion of the particles into the aqueous subphase. Mathematical analysis of the forces acting on a particle deposited on an air-fluid interface show that for small particles ( < 100 #m) the surface tension force is several orders of magnitude greater than forces related to gravity. Thus, even at the relatively high surface tension obtained in the airways (32 + 2 dyn. c m - a) particles will still be displaced into the aqueous subphase. Particles in peripheral airways and alveoli likely are below the surfactant film and submerged in the subphase. This may promote clearance by macrophages. In addition, particle displacement into the subphase is likely to increase the contact between the epithelial cell and particle. Toxic or allergenic particles would be available to interact with epithelial cells and this may be important in the pathophysiology of airway disease.

Airway; Alveoli; Surface tension; Particle immersion; Subphase; Osmiophilic layer

Inhaled particles may be deposited in the conducting airways or they may enter the gas exchange region of the lungs. By definition, deposition of particles is terminated as soon as they touch the walls of the airspaces. Following deposition, retention and clearance begin. Clearance of insoluble particles from the airspace surfaces in the lungs has two Correspondence to." S. Schllrch, Respiratory Research Group, Departments of Medical Physiology and Medicine, Health Sciences Centre, The University of Calgary, Calgary, Alberta, Canada T2N 4N1. 0034-5687/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)

18

s. SCHURCHet al.

predominant phases. The first phase, which is rapid, applies to particles deposited on the surfaces of conducting airways. This phase, which is mediated by mucociliary activity, is usually complete in 24-36 h but may talte longer (Gore and Patrick, 1978; Stahlhofen etaL, 1986; Smaldone etaL, 1988). The second phase which involves particles deposited in the gas exchange regions of the lung, is slow in comparison, and may last for months or years. Clearance for such particles involves phagoeytosis by pulmonary macrophages, macrophage migration and mucociliary activity. For both deposition and clearance the structure of the airway coating and the surface properties of the aqueous phase adjacent to air are important (e.g. Rensch et al., 1983; Widdicombe, 1985). The mucous layer coating the conducting airways is thought to be based on a two-phase system, consisting of a less viscous sol phase that includes the beating cilia and above a more viscous gel phase, the mucus blanket, which contains inhaled particles, cells and cell debris. This mucus blanket is thought to be moved towards the larynx by ciliary action (e.g. Kilburn, 1968). Although this concept of a two phase system is widely accepted as occurring in the larger airways, there is still debate concerning the presence of a gel layer in the smaller peripheral airways (Brain et al., 1984; Gil and Weibel, 1971). Recent structural studies indic ate that the sol and gel layers of the larger airways are separated by a partially stretched and partially reticulated osmiophilic membrane (Widdicombe, 1985). These phospholipid membranes may act as lubricants to facilitate the sliding of one phase on the other (Morgenroth, 1985; AUegra et al., 1985). Several investigators (Green, 1973; Rensch et al., 1983) have pointed out the possible role of surfactant in particle clearance from non-ciliated peripheral airspaces and alveoli. Alveolar surface tension reaches values close to zero on expiration (e.g., Schtlrch et aL,1985) while the surface tension in the trachea and in the first and second generation bronchi is much higher (approximately 30 dyn. cm- 1, this work). A surface pressure gradient thus exists with a high surface pressure (low tension) in the lung periphery and low surface pressure (high tension) in the conducting airways. It has been suggested that particles deposited onto the surfactant f'dm in alveoli would be swept along this surface pressure gradient toward the conduction airways (Rensch et aL, 1983). This postulated process is based on the assumption that deposited particles are floating on the surface film. However, morphological investigations in our laboratories have shown that latex (polystyrene) particles, equal or less than 6 #m in diameter were completely submerged in the subphase of the airways (Geiser et aL, 1988). In the present study we have extended these observations to show that latex particles deposited in alveolar spaces are also found below the surface active layer. The mechanism of particle displacement depends critically on the existence of a surfactant film at the airway or alveolar surfaces. With in vitro methods, using particles on surfactant films in a Wilhelmy-Langmuir balance, we will demonstrate that pulmonary surfactant promotes the displacement of particles from air to the aqueous phase and that the extent of particle immersion depends on the surface tension of the surface active film. The lower the surface tension, the greater is the immersion of particles into the airspace subphase adjacent to the epithelium.

SURFACTANT DISPLACES PARTICLESIN AIR SPACES

19

Additionally we show the upper airways to be covered by a surface active film, having a surface tension of 32 + 2 dyn. cm- ~. This follows from direct measurements of the airway surface tension using a bronchoscope and a drop spreading technique. Compared to the alveolar region, particles in the upper airways are similarly affected by surface tension forces; however, the situation is more complex than in the alveoli in view of the complex nature of the mucus layer.

Methods

Animal exposures and electron microscopy Syrian golden hamsters inhaled a monodispersed aerosol containing 1, 3 and 6 #m polystyrene (latex) beads by spontaneous breathing through an intratracheal cannula (intubation or tracheotomy). A new inhalation apparatus has been used which applies pneumotachography and laser light scattering photometry for continuous monitoring of flow rate and particle number concentration. The apparatus also includes an analog computer to calculate the number of particles deposited per breath. The number of particles deposited depends on the breathing pattern which in turn depends on the intensity of the anesthesia. The breathing frequency was about 45 min- ~, and the average tidal volume was estimated to be 0.65 ml. The particle concentrations were a p p r o x i m a t e l y 10 4 to 10 6 × cm-3 of inhaled air and the total number of particles deposited in the lungs of these animals during an inhalation period of 20 to 30 min was 4 to 60 × 10 6 particles, depending on the particle size (Im Hof et aL, 1989). Fifteen minutes after the termination of the inhalation period the trachea was cannulated and after three quasistatic inflation-deflation cycles, the lungs were inflated to total lung capacity (TLC) at a pressure of 25 cm H20. The lungs were kept at 60Y/o TLC following deflation from TLC to 5 cm H20. The lungs were then f'Lxedby intravascular perfusion of 2.5~ glutaraldehyde, 1Y/o osmium tetroxide and 0.5~o uranyl acetate, in that order. After the fixation, the lungs were carefully excised, separated from other organs and their volumes estimated by water displacement. The lungs were further processed for light and electron microscopic observation and stereological analysis (Geiser etal., 1989; Im H o f e t a L , 1989). Surface tension measurements in airways We determined the surface tension in airways in situ with two approaches. Sheep were used in these experiments in view of our need for an animal with an airway large enough to bronchoscope. In the first experiment we cut a freshly excised sheep trachea lengthwise, spread the unfolded trachea fiat and secured it with pins on a board. We determined the air-substrate surface tension of the mucosal surface of the trachea by using the drop spreading method as previously described (Sch0rch et al., 1985). Briefly, we placed a series of small droplets (0.1-0.5 mm in diameter) of fluorocarbon fluid (FC43, 3M Co., Saint Paul, MN) by means of a micropipet connected to a microsyringe (Hamilton) onto the mucosal surface of the trachea. The surface tension of this surface

20

s. SCHURCH et al.

was estimated from the relative diameter of the droplets. The relative diameter is the diameter of the droplet on the surface divided by the diameter of that droplet in its spherical shape, measured prior to deposition while the drop is still attached to the pipet. The surface tension can be determined from a calibration curve which gives the relationship between the relative diameter of the drop placed onto a surfactant film and the surface tension of that film (Schtlrch et al., 1985). For the second experiment we used a bronchoscope (Olympus) whose suction tunnel contained a thin Teflon tubing. The bronchoscope was introduced into a second generation bronchus of an anesthetized sheep that was intubated and ventilated by a respirator. A small droplet (0.2 mm in diameter) of dimethylphthalate doped with 1 mg. ml- 1of c~stal violet was squeezed out of the Teflon tubing and the drop diameter was measured while the drop was still hanging on the tip of the tubing. The respirator was then stopped for about 3 rain, and the test fluid droplet was placed onto the wall of the bronchus by moving the tip of the light guide slightly. Upon touching the wall, the droplet spread out to a lens with a diameter that is characteristic for the surface tension of the wall-air interface. Panicle displacement from air to the aqueous phase

An aqueous suspension of monodispersed polystyrene particles, average diameter 11.9 #m (Sigma), were cleaned by dialysis for 24 h to remove the surface active agents that are added to stabilize latex suspensions. After freeze drying, some of the particles were blown onto the interior surface of a freshly excised sheep trachea. The trachea had been cut lengthwise, unfolded and secured with pins on a fiat board. Photographs were taken with a Nikon Optiphot-M (Metallurgical) microscope equipped with a photographic attachment (UFX) and a Universal Epi Illuminator (100 W halogen). The Epi illuminator system allows the use of bright field, dark field and differential interference (DIC Nomarski) examination techniques for reflected light with a single nosepiece and one set of optics (10 x, 20 ×, 40 x, bright field, dark field, plan, extra long working distance). The photographs were taken with an Ilford XPI film exposed to 400 ASA. Latex beads on an airway surface.

We studied the displacement of spherical particles on monolayers of 1,2-dipalmitoyl-sn-3-glycerophosphorylcholine (DPPC) in a Langmuir-Wilhelmy type surface balance. Monolayers of DPPC (Sigma) were spread from a solution of 2 mg" m l - ~ in ethanol onto aqueous substrates of solutions of 0.9~o NaC1 and various amounts of sucrose, which were added to adjust the density of the monolayer supporting aqueous medium. The monolayers were spread within an area bound by a Teflon ribbon supported by a rhombic frame. This version of a Langrnuir-Wilhelmy balance has been described previously by Schoedel et al. (1969). In order to reduce the surface tension of an insoluble monolayer such as formed from DPPC, the rhombic frame can be pulled to an elongated shape. This frame was made to fit into a glass petri dish, 15 cm in diameter, which served as the Langmuir trough.

Particles on air-water surfactant monolayers.

SURFACTANT DISPLACES PARTICLES IN AIR SPACES

21

The surface tension was measured by a platinum Wilhelmy plate that was connected to an electrobalance (Cahn, Ventron Instruments, Paramount, CA). After adjusting the monolayer surface tension to a preset value, the glass dish Langmuir trough was placed onto the microscope stage. Particles of polystyrene (11.9/~m in diameter, Sigma, fig. 3). and of polymethylmethacrylate (PMMA, 10-100#m in diameter, Polysciences; figs. 4, 5) were blown onto the monolayer. The particles were then observed by DIC Nomarski microscopy for epi illumination (as described above). DIC Nomarski microscopy is especially suitable to detect level differences (Z-direction). Photographs of the particles were taken after focussing the objectives on the particle-air-fluid 3-phase line. The maximum diameter of the particles immersed into the aqueous phase was then slightly out of focus on these photographs (figs. 3-5).

Results

Electron microscopy. Depending on the size, the inhaled particles are found in different regions of the lungs, but with overlapping deposition patterns. Figure 1 shows 1 and

Fig. 1. Transmissionelectronmicrograph(TEM)of epithelialsurfaceofconductingairway 15 rain following inhalation of mixture of I and 6/~m polystyrenebeads. Both sizes of particles have been displaced into the sol phase and are seen in close contact with cilia. The particles have become coated with a densely osmiophilic layer (arrows) which may be derived from the interface of the gel and sol phases (Marker = 5 #m).

22

s. SCHURCH et al.

6/am particles deposited on a large conducting airway that have been retained in the sol phase, caught between cilia. The particles are covered with osmiophilic (phospholipid) material. It has not been possible to determine whether this coating consists of a mono-, bi- or multiple layer. Figure 2A demonstrates particles displaced toward the epithelium in an alveolar space. There is an osmiophilic layer at the air-aqueous phase interface. Surface tension forces had displaced the particles into the subphase causing deformation of the capillary walls which partially surround the particles. Figure 2B, at a higher magnification, demonstrates an osmiophilic coating around a latex particle in an alveolus. This coating is related to phospholipids of pulmonary surfactant. Surface tension in airways

We placed 10 fluorocarbon (FC43) droplets at 10 different places on the excised sheep trachea as described above. The surface tension was 32.5 + 2.5 dyn. c m - 1 (mean + 1 SEM, n = 10).

In a sheep trachea.

By using a bronchoscope and observing the spreading behavior of test fluid droplets we determined the surface tension in a second generation sheep bronchus. The spreading indicated a surface tension of 32 + 2 d y n . c m -~ (mean + 1 SEM, 15 droplets at various places in two bronchi of the same second

In a sheep bronchus.

A

ta

Fig. 2. (A) TEM of alveolarwall showing 1/~mdiameter polystyrenebeads in an alveolar space (AS). The particles have been displaced toward the epithelium and are completelycovered by an osmiophiliclayer (surfactant). Surface tension forces have caused deformation of the underlying capillary (CAP) by the particles. (B) Higher magnificationof same general area showing osmiophilic (lipid) coating of a 3 #m particle (arrows).

SURFACTANT DISPLACES PARTICLES IN AIR SPACES

23

generation). These experiments indicated that there is a film of surface active material at the wall-air interface of these bronchi. Mucus and other biopolymers, even in high concentrations all have surface tensions between 40 and 50 dyn. c m - 1 (unpublished observation).

Latex (polystyrene) particles deposited onto a sheep trachea The majority of the polystyrene particles (average diameter 11.9 #m) on the trachea appeared to be submerged in the aqueous phase and only a very small segment of the spheres appeared at the air-substrate interphase (fig. 3A). Particles on sutfactant air-water monolayers in a surface balance Figure 3B demonstrates the appearance of latex (polystyrene)particles on a surfactant monolayer in a Langmuir-Wilhelmy surface balance. The photographs were taken with an Ilford X P I film exposed to 1600 ASA to improve sharpness of the photograph since the particles were moving slowly on the aqueous substrate. The aqueous substrate for the monolayer was 0.9~o NaC1 with 55~o sucrose. The density of the substrate was 1.26 g. ml - 1, considerably higher than that of the particles, which was approximately 1.05 g. c m - 3. At a monolayer surface tension of 30 dyn. c m - ~, the majority of the polystyrene particles on the monolayer appeared to be submerged in the aqueous phase and only a small segment of the particles appeared at the air-substrate surface (fig. 3B). Figure 4 demonstrates P M M A particles at various surface tensions. At the lowest surface tension, approximately 20 dyn. c m - 1, the particles appear almost totally submerged in the aqueous phase (C) while at a higher surface tension of 45 dyn. cm - 1 (A) almost 50~o of the particle surface appears exposed to air. At a lower density (44~o) of the sucrose subphase (1.20 g . m l - ~ , which is close to the density of P M M A (1.19 g. m l - 1), the particles appear totally submerged at film tensions of 20 dyn . c m - 1 or below (fig. 5B).

Fig. 3. Polystyrene beads (average diameter 11.9/~m submerged in the aqueous phase. A: on a sheep trachea; B: on a layer of surfactant (dipalmitoylphosphatidylcholine,DPPC) on a subphase of 55% sucrose and 0.9% NaCI in water. The density of the aqueous phase was 1.26g'cm -3, that of polystyrene 1.05 g" cm - 3, the surface tension was 30 dyn' cm - ~. Note: In A and B only very small segments appear exposed to air (arrows), the arrowheads indicate the total diameters of the beads. The photographs were taken with a Nikon metallurgicalmicroscope, using differential interference contrast for epi-illumination.

24

S. SCHURCH etal.

V

A

6

I

4 0

0

B

~ °

•

•

o

•

OOo .o.% o

)O

•

0

0

go

.c

dlfk~

C

Fig. 4. Polymethylmethacrylate (PMMA) beads on a DPPC monolayer supported by an aqueous subphase (see fig. 3, density 1.26 g . c m - 3 ) at decreasing surface tensions, A: 50 dyn. cm-1, B: 40 dyn. cm-~, C: 30 dyn" cm - 1. Note: The appearance ofthe labelled particle at decreasing film surface tensions, D (80 #m) indicates the total diameter, d is the diameter of the segment exposed to air, this diameter is decreasing with the surface tensions, indicating increasing particle immersion.

SURFACTANT DISPLACES PARTICLES IN AIR SPACES

25

A

O O O

Fig. 5. PMMAbeads on a DPPC monolayersupportedby an aqueous subphase of a density(1.20g. cm - 3) slightly above that of PMMA (1.19g. cm-3) at decreasing film surface tensions, A: 50 dyn-cm-1, B: 35 dyn .cm-t; C: 20 dyn •cm-t. Note: Label D indicates the total particle diameter, 80#m, d is the diameter of the segment exposed to air. In Fig. 5C, at a surface tension of 20 dyn'cm-~, the beads are completelysubmerged.

Force analysis of small particles at fluid-air surfaces Introduction. We followed the analysis as presented by Princen (1969) and Rapacchietta etal. (1977a,b) who studied the position of cylindrical and spherical particles as a function of the interfacial tension or interfacial free energies. Especially the critical size for floatability of particles on a fluid-air or fluid-fluid interface was considered. Thus, these authors were interested in particles floating on a surface whose subphase density was less that of the particle. For example, Princen (1969) pointed out the phenomenon of a waxed needle floating on a clean water surface. To our knowledge nobody has analyzed the situation where the interface is modified by an insoluble surfactant monolayer of variable surface tension and where the substrate density might be higher than that of the particles. In the following we have applied the analysis of Rapacchietta et al. (1977a,b) to discuss the particle displacement by surface tension forces. We have adapted this analysis to include a monolayer of changing surface tension since surface tension varies in different regions of the lung.We have considered only ideal, that is homogenous and smooth, solid, spherical surfaces. Figure 6A,B represents three mechanically enforced stationary states of a spherical particle having a relatively large contact angle (60 °) with the fluid phase, A, and a relatively low contact angle (20 °) with the fluid phase, B. The mechanically enforced stationary state may be achieved by using some agent which exactly compensates for any existent net vertical force, e.g., the particle may be suspended by a string in a particular position. Assuming particular numerical values for the surface tension of the monolayer at the air-fluid interface, for a particle size equal or below about 100 ~tm and subphase densities equal or above that of water, we see from eq. (1)*, (2)* and (3)* that :the surface

26

S. SCHURCH et al.

B

....................

1

I

1

I

2

!, ...-'-

..........................

Fig. 6. Three (I, II, III) mechanically enforced stationary states (see text) of a spherical polystyrene particle, having a contact angle of 0 = 60 ° (A) and 0 = 20 ° (B). The lower contact angle (B) is characteristic for greater particle wettability by a fluid phase having a monolayer surface tension of approximately 30 dyn' cm- ~,while the higher contact angle, A, indicates less wettability. In this case the monolayer surface tension is about 50 dyn' cm- '. In both figures the equilibrium position is achieved when the meniscus of the fluid phase 3 is almost horizontal up to the 3-phase line at the particle. In this instance the vertical component of the force generated by the surface tension 713is vanishing. This occurs for almost total particle immersion in B (III), where the surface tension 7~3is approximately 30 dyn" cm - t, while at a higher surface tension of about 50 dyn' cm- 1, much less of the particle is immersed into the fluid phase (A,III). (See text for the force analysis).

t e n s i o n force, F1, is several o r d e r s o f m a g n i t u d e greater t h a n the forces r e l a t e d to gravity, weight a n d b u o y a n c y . T h e equilibrium p o s i t i o n is only p o s s i b l e if sin (0 + q~) a p p r o a c h e s zero or in other w o r d s if (0 + 4,) a p p r o a c h e s 180 ° . In fig. 6A, wettability o f the particle by the a i r - f l u i d m o n o l a y e r is relatively p o o r , e.g. for a m o n o l a y e r surface t e n s i o n o f 45 d y n . c m - 1 the c o n t a c t angle, 0, on P M M A is a p p r o x i m a t e l y 60 °. T h e equilibrium p o s i t i o n c h a r a c t e r i z e d by a v a n i s h i n g vertical c o m p o n e n t o f the r e s u l t a n t vertical force is a c h i e v e d at a p o s i t i o n w h e r e a relatively large p a r t o f the s p h e r e is e x p o s e d to air.

SURFACTANT DISPLACES PARTICLES IN AIR SPACES

27

On the other hand, in fig. 6B, for a greater wettability or for a relatively low contact angle, e.g. 0 = 20 ° for a monolayer surface tension of about 25 dyn. cm - l, the spherical particle is almost totally immersed into the fluid phase. In other words, the surface tension force which is proportional to sin (0 + ~) approaches zero only for almost total immersion of the particle. As described below, the contact angles which aqueous droplets make on PMMA are estimated from independent measurements. The surface tension of these drops was modified by DPPC monolayers of differing surface tension, and the drops were placed onto a PMMA sheet formed by solvent casting, to determine the advancing contact angle,

Force analysis.

For a spherical particle, the length of the 3-phase line is

1 = 2rrR. sin~

(see fig. 6)

The vertical component of the surface tension force is Fl = 21rR" Y13"sin~" sin(0 + q~)

(1)

where ~13 is the surface tension at the phase 1,3 interface, ~bindicates the position of the 3-phase line; 0 is the contact angle, Z o is the height of the meniscus. The weight of the particle generates the second force term F2 = 4/37rR3 "P2g

(2)

where P2 is the density of phase 2, g is the acceleration due to gravity. Integrating the vertical component of the hydrostatic pressure distribution around the entire particle results in the following expression: F3 = nR2" (P3 - Pl)g" Zo(sin~) 2 + nR3" (P3 - Pl)g" (sin~) 2 cos0 + 2/3nR3" (P3 - Pl)g" ((cos~) 3 - 1} - 4/3nRs "Pig

(3)

The net vertical force is obtained by adding eq. (1-3). The sign convention can be seen from eq. (1): sinking forces are positive, lifting forces are negative. The equilibrium position is achieved if the net vertical force is zero. The Laplace equation of capillarity governs the formation of a meniscus to the spherical surface. In order to obtain Zo, the capillary rise or height of the meniscus, the Laplace equation has to be solved for the given interfacial tensions and densities. This involves numerical approximations since an analytical solution for the present case does not exist. We refer to the article by Rapacchietta etal. (1977a,b) for a complete discussion and solution of the problem.

28

s. SCHURCH et al.

In the following we have approached the problem from the experimental point of view, discussing the magnitude of the various force terms for small particles of approximately 5/am radius up to larger particles of about 1 mm in radius. Further, we estimated the magnitude of the capillary rise Z o by confocal light microscopy (Tracor Northern Tandem Scanning Reflected Light Microscope). Using a 100 x objective, the resolution in the Zo direction is better than 0.5/am, which allowed us to estimate the capillary rise for particles having a radius as small as 5/am. To estimate the force terms, eq. (1-3) for latex (polystyrene) particles, we assume the following values (see fig. 6):

Force and particle size.

Pl = 0 (density of the air phase) P2 = 1.05 g. cm - 3 (density of polystyrene) P3 = 1.26 g. c m - 3 (density of the aqueous phase, consisting of 54~o sucrose in 0.9~'o N a e l ) 722 = 33 dyn. c m - l (surface tension or free energy of polystyrene). We are considering three particle sizes, (R = 10/am, R --- 100 #m and R = 10- 1 cm), and a monolayer surface tension of 7a3 = 30 dyn. c m - 1. An estimate for the surface tension (free energy of polystyrene is taken from Gerson (1982) who determined Zisman's critical surface tension for polystyrene to be 33.0 dyn. c m - 1 and for P M M A to be 42.8 dyn" c m - 1 Using equations (1), (2),(3) for small particles of radius R = 10- 3cm, and a surface tension (free energy of 723 = 30 dyn. c m - 1, we obtained: F 1 = 1.9- 10- 2. sintp, sin(0 + q~) dyn [ > 0 ] F 2 = 4.4.10 -6 dyn [ > 0 ] F 3 = 1.9.10- 7. (sinq~)2 dyn [ > 0]; [Z o = 10- 4 cm] + 3 . 8 . 1 0 - 6. (sin~b)2 cost~ dyn [ < 0] + 2.6.10 - 6 . {(cos~b)3-1} dyn [ < 0]

(1)* (2)*

(3)*

Note: Sinking forces are positive, tiffing forces are negative. As can be seen, the surface tension term F 1 is several orders of magnitude greater than all other terms for small particles. Thus, the sinking force F 1 dominates all the other force terms for small particles, except for the case when sin(0 + q)) approaches zero. That is to say, the particles can only achieve an equilibrium position if the fluid level is almost horizontal up to the 3-phase line at the sphere. Even for larger particles, e.g. with a radius of 10- 2 cm, the surface tension force term F1 is approximately 2-3 orders of magnitude greater than the weight and buoyancy terms F z and F 3, given the same conditions as for the smaller particles of radius 10 -3 cm. For very large particles of radius 10-1 cm, for a monolayer surface tension of 30 dyn. c m - ~ and the same densities as above, the terms related to gravity approach the same order of magnitude as the surface force term:

SURFACTANT DISPLACES PARTICLES IN AIR SPACES

29

F~ ~ 19 sintp, sin(0 + ~p)dyn[ > 0 ] (1)** F 2 ,~ 4 dyn [ > 0] (2)** F 3 ~ 2 sin2~p dyn; Zo = 10 -3 cm [ > 0 ] (3)** + 4 sin2tp • cos~p dyn [ < 0] + 3 (cos3tp - 1) dyn [ < 0 ] Note: Sinking forces are positive, lifting forces are negative. Such relatively large particles may assume an equilibrium position characterized by a capillary height that approaches the magnitude of the particle radius (direct observation by confocal light microscopy).

Forces at differing monolayer surface tensions.

Figures 4,5 demonstrate that the particle positions, as indicated by the differing sizes of the spherical segments, depend on the monolayer surface tension. At a lower surface tension (fig. 4B) a larger surface area of the spherical particles is wetted by the aqueous phase. At lower surface tensions the particles appear displaced into the aqueous medium to a greater extent than at higher surface tensions. This phenomenon can be discussed in terms of the contact angle at the 3-phase line (fig. 6). Although we were not able to measure the contact angles directly on single particles, it is clear from the application of Young's equation (Young, 1855), that the contact angle critically depends on the surface tension of the monolayer at the water-air interface. In order to estimate the contact angles we placed on a P M M A surface sessile drops of an aqueous medium having DPPC monolayers of differing surface tensions. Various surface tensions could easily be achieved by spreading small amounts, less than 0.1 #1 of the spreading solution of DPPC, 2 mg. ml-~ in ethanol, on these aqueous drops, whose volumes were approximately 0.5 ml. At a monolayer surface tension of approximately 40 dyn. c m - 1, the contact angle was about 45 ° while at a surface tension of 30 dyn. c m - ~ the contact angle was about 20 ° . Entering these contact angles into eq. (1) we see that near equilibrium, where sin(0 + q~) must approach zero, 0 + tp must approach 180 °. Therefore, at a surface tension of 40 dyn. c m - l, 0 = 45 °, ~pis approximately 135 °, while at the lower surface tension of 30 dyn. c m - ~, 0 = 20 °, tp is approximately 160 ° .

Discussion

Particle displacement by surface forces.

The force analysis demonstrates that for latex particles (polystyrene) or P M M A particles whose radius is less than about i00/am, the surface forces, in general, are the dominating factor in particle displacement from one mechanically enforced stationary,state to the next stationary state. Figure 6 illustrates such mechanically enforced stationary states, I, II, III where the net vertical force is exactly compensated by an external force: for example, the particle is held in a certain position by a string. Force contributions related to gravity, weight and buoyancy, only

30

s. SCHURCHet al.

become important for particles approaching an equilibrium position, characterized by a surface of the fluid phase that is almost horizontal up to the 3-phase line at the spherical particle. The position of the 3-phase line, critically depends on the contact angle the surfactant covered fluid phase makes with the particle. For a given particle material, the lower the surface tension of the surfactant film, the lower will be the contact angle between the particle and the fluid phase and the greater will be the extent of particle wetting. In fig. 6B the contact angle is approximately 20 °, while in 6A, the angle is 60 °. Our investigations with PMMA particles and films with differing surface tensions demonstrated that the extent of particle immersion in an aqueous phase is dependent on the surface tension of the monolayer that covers the aqueous phase. The lower the surface tension of the monolayer, the greater will be the particle immersion (fig. 4). If the density of the aqueous subphase is considerably greater than that of the particle a very small part, less than 1~ of the particle surface, will remain exposed to air upon reducing the monolayer surface tension to values of 20 dyn. cm-1 or below. This follows from the application of the force analysis to latex (polystyrene) or PMMA particles. In this case the surface tension force approaches zero and the buoyancy force compensates the sinking forces. However, if the density of the aqueous subphase is about equal or less than that of the particle, the particle will be totally submerged below a certain surface tension of the monolayer (fig. 5B). At a monolayer surface tension of 30 dyn. cm- ~ PMMA particles appeared to be immersed slightly more than polystyrene particles. This is in line with the more hydrophilic nature of PMMA compared to polystyrene. PMMA has a surface free energy of approximately 40 erg. cm-2 compared to polystyrene, 33 erg. cm-2 (Gerson, 1982). Thus, for a given monolayer surface tension we expect a lower advancing contact angle for PMMA than for polystyrene. This can be tested by placing aqueous drops that are coated with a DPPC monolayer onto sheets of PMMA or polystyrene. Conducting airways

Latex (polystyrene) has a surface free energy of 33 erg. cm- 2 (Gerson, 1982), close to that measured in the upper airways (32 erg. cm- 2). Polystyrene particles, 12/~m in diameter if placed onto the surface of a sheep trachea or onto a surfactant film that has a surface tension of approximately 30 dyn. cm- 1, appear almost totally pulled into the aqueous (mucus) phase (fig. 3A) or almost totally immersed into the surfactant covered sucrose solution (fig. 3B) whose density (1.26 g. cm-3) is much higher than that of polystyrene (1.05 g. cm- 3). The present study has shown the following important findings: (1) Inhaled particles of polystyrene which are deposited on the lining material of the conducting airways will be pulled towards the epithelium by surface forces exerted on them by the surfactant t'tim at the air-water interface (figs. 3-5). Therefore, the particles are not situated on top of the mucus, but rather inside the mucus. Theoretical analyses indicate that most particles ofrespirable size (< 10/~m) would behave similarly. (2) As these particles were found in close association with the airway epithelium (fig. 1). (Im Hof et al., 1989; Geiser et aL, 1989), they must have been displaced through the gel phase into the sol

SURFACTANT DISPLACES PARTICLES IN AIR SPACES

31

phase. During this process they become coated with osmiophilic material. This displacement might enhance clearance by bringing the particles into contact with macrophages and cilia. Direct observation of particles deposited in peripheral airways and alveoli showed these particles to be submerged in the aqueous phase. At the alveolar level surface forces appear to pull the particles into the subphase with sufficient force to depress the underlying capillaries. It would appear that as the particles enter the subphase they become coated with an osmiophilic film related to the phospholipids of surfactant. Since the particles appear to be totally coated by this osmiophilic material, the surfactant film must have completely surrounded the particles after deposition. Since the alveolar surface tension falls toward zero on lung deflation, even hydrophobic particles, such as Teflon particles, will be wetted by the aqueous phase with its surfactant film, and the particles will be located below the film. Most particles inhaled, for example, dust particles, pollen, spores, etc. are relatively hydrophilic, hence extremely low surface tensions would not be needed to displace such particles into the aqueous subphase. If, as we have shown particles do not remain floating on the surface active film, the condition for a possible transport mechanism by a surface tension gradient cannot be met (Davis et aL, 1974). It is more likely that particles deposited in alveolar spaces will be moved by capillary action or gradients in subphase pressure (Guyton et al., 1984) towards places with substantial amounts of subphase material such as alveolar corners, clefts and crevices where the interface curvature is relatively high. This will enhance alveolar clearance by alveolar macrophages in two ways: first, particles coated with surfactant are more readily phagocytosed than are uncoated particles (Jarstrand et al., 1984) and, secondly, this process will tend to bring the particle into contact with macrophages which are known to preferentially inhabit the recesses and crevices formed by highly curved surfaces in the lung. In conclusion we have demonstrated that surface active material lining the airways and alveoli promotes the displacement of deposited particles into the aqueous subphase. In addition, we have shown that the lower the surface tension, the greater is the extent of particle immersion. These findings are likely to have important implications with regard to particle clearance from the airways and aiveoli. In addition, particle displacement into the subphase is likely to increase the possibility of contact between the epithelial cell and particle. For inert particles this would have few consequences. Toxic or allergenic particles, on the other hand, would be available to interact with epithelial cells at the mucosal surface and this may be important in the pathogenesis of certain forms of airway disease.

Peripheral airways and alveoli.

Acknowledgements.The authors thank Dr H. Bachofenfor his valuablesuggestions.Theyalso thank Mrs E. Siegfried,Mrs Jo Anna Swainger, Mr K. Babl and Mr Ch. Lehmann for their outstandingtechnical assistance.This workwas supportedby the MRC,MT-6435,The AlbertaHeritageFoundationfor Medical Research, and the Swiss NSF, Grant No. 3.909-0.85.

32

S. SCHURCH et aL

References Allegra, L., R. Bossi and P. Braga (1985). Influence of surfaetant on mucociliary transport. Eur. J~. Respir~ Dis. (Suppl. 142) 67: 71-76. Brain, J. D., S. B. Bloom, P. A. Valberg and P. Gehr (1984). Correlation between the behaviour of magnetic iron oxide particles in the lung of rabbits and phagocytosis. Exp. Lung Res. 6: 115-131. Davis, S.H., A.K. Liu and G.R. Scaly (1974). Motion driven by surface tension gradients in a tube lining. J. Fluid Mech. 62: 737-751. Geiser, M., V. Im Hof, P. Gehr and L.M. Cruz-Orive (1988). Histological and stereological analysis of particle deposition in the conducting airways of hamster lungs. J. Aerosol Med. 1: 197. Geiser, M., L. M. Cruz-Orive, V. Im Hof and P. Gehr (1989). Counting particles retained in the conducting airways of hamster lungs with the fractionator. Acta Stereol. (in press). Gerson, D. F. (1982). An empirical equation-of-state for solid-fluid interracial free energies. Colloid Polymer Sci., 260: 539-544. Gil, J. and E.R. Weibel (1971). ExtraceUular lining of bronchioles after perfusion-fixation of rat lungs for electron microscopy. Anat. Rec. 169:185-200. Gore, D.J. and G. Patrick (1978). The distribution and clearance of inhaled UO2 particles on the first bifurcation and trachea of rats. Phys. Med. Biol. 23: 730-737. Green, G.M. (1973). Alveolar bronchiolar transport mechanisms. Arch. Intern. Med., 131: 109-114. Guyton, A.C., D.S. Moffatt and T.H. Adair (1984). Role of alveolar surface tension in transepithelial movement of fluid. In: Pulmonary Surfactant, edited by B. Robertson, L. M. G. Van Golde and J.J. Battenburg. New York, Elsevier, pp. 171-185. Im Hof, V., G. Scheuch, M. Geiser, J. Beghart, P. Gehr and J. Heyder (1989). A new technique for quantitative deposition of particles in lungs of small rodents. J. Aerosol Med. (in press). Jarstrand, C. (1984). Role of surfactant in the pulmonary defenee system. In: Pulmonary Surfactant, edited by B. Robertson, L.M.G Van Golde and J.J. Batenburg, New York, Elsevier, pp. 187-202. Kilburn, K.H. (1968). A hypothesis for pulmonary clearance and its implications. Am. Rev. Respir. Dis. 98: 449-463. Morgenroth, K. (1985). Morphology of the bronchial lining layer and its alteration in IRDS, ARDS and COLD. Eur. J. Respir. Dis. (Suppl. 142) 67: 7-18. Princen, H.M. (1969). The equilibrium shape of interfaces, drops and bubbles. Rigid and deformable particles at interfaces. In: Surface and Colloid Science Vol. 2, edited by E. Matijevic. New York, Wiley Interscience, pp. 1-84. Rapacchietta, A.V., A.W. Neumann and A.N. Omenyi (1977a). Force and free-energy analysis of small particles at fluid interfaces. I. Cylinders. J. Colloid Interf. Sci. 59: 541-554. Rapacchietta, A.V., A.W. Neumann and S.N. Omenyi (1977b). Force and free-energy analysis of small particles at fluid interfaces. II. Spheres. J. Colloid lnterf. Sci. 59: 555-567. Rensch, H., H. Von Seefeld, K. F. Gebhardt, D. Renzow and P.J. Sell (1983). Stop and go particle transport in the peripheral airways? A model study. Respiration 44: 346-350. Schoedel, W., H. Slama and E. Hansen, E. (1969). Zeitabh~ngige Ver~mderungen des Filmdruckes alveolarer Oberflachenfilme im Langrnuir-Trog. Plfagers Arch. 306: 20-32. Schttrch, S., H. B achofen and E.R. Weibel (1985). Alveolar surface tension in excised rabbit lungs: effect of temperature. Respir. Physiol. 62: 31-45. Smaldone, G.C., R.J. Perry, W.D. Bennett, M. S. Messina, J. Zwang, and J. Ilosite (1988). Interpretion of '24 hour lung retention' in studies of mucociliary clearance. J. Aerosol Med. 1:11-20. Stahlhofen, W., J. Gebhart, G. Rudolf and G. Scheuch, (1986). Measurement of lung clearance with pulses of radioactively labelled particles. J. Aerosol Sci. 17: 333-338. Widdicombe, J.G. (1985). Introduction to: Endobronchial surface active phospholipids: Morphology, biochemistry, function and therapeutic aspects edited by L. M. G. Van Golde, J.G. Widdicombe and P. Vermeire. Eur. J. Respir. Dis. (Suppl. 142) 67: 1-5. Young, T. (1855); Miscellaneous works. Vol. 1, edited by G. Peacock. London Murray, p. 418.