|Home | About | Journals | Submit | Contact Us | Français|
Alveolar capillaries are located in close proximity to the alveolar epithelium and beneath the surfactant film. We hypothesized that the shape of alveolar capillaries and accompanying oxygenation are influenced by surfactant surface tension in the alveolus. To prove our hypothesis, surfactant surface tension was regulated by conditional expression of surfactant protein (SP)-B in Sftpb−/− mice, thereby inhibiting surface tension–lowering properties of surfactant in vivo within 24 hours after depletion of Sftpb. Minimum surface tension of isolated surfactant was increased and oxygen saturation was significantly reduced after 2 days of SP-B deficiency in association with deformation of alveolar capillaries. Intravascularly injected 3.2-μm-diameter microbeads through jugular vein were retained within narrowed pulmonary capillaries after reduction of SP-B. Ultrastructure studies demonstrated that the capillary protrusion typical of the normal alveolar–capillary unit was reduced in size, consistent with altered pulmonary blood flow. Pulmonary hypertension and intrapulmonary shunting are commonly associated with surfactant deficiency and dysfunction in neonates and adults with respiratory distress syndromes. Increased surfactant surface tension caused by reduction in SP-B induced narrowing of alveolar capillaries and oxygen desaturation, demonstrating an important role of surface tension–lowering properties of surfactant in the regulation of pulmonary vascular perfusion.
Pulmonary surfactant is a complex mixture of lipids and associated proteins that are required for formation and stability of surfactant film in the alveolus. After secretion from type II epithelial cells, surfactant forms a lipid rich film that covers the entire alveolar surface, reducing surface tension at the air/liquid interface from 70 mN/m to near 0 mN/m (1). The reduction of alveolar surface tension is required for the maintenance of alveolar surface area, upon which respiration depends. Surfactant function requires the presence of surfactant protein (SP)-B, a small hydrophobic protein that is tightly associated with surfactant phospholipids in the alveolus (1, 2). In humans and mice, SP-B deficiency or mutations in SFTPB cause respiratory failure and death in adults and neonates (3–5). Conditional reduction of SP-B for 4 days in adult mice causes respiratory failure associated with abnormal high surface tension of the surfactant present in bronchoalveolar lavage fluid (BALF) (6–8). This conditional SP-B mouse provides a useful model for study of the influence of surface tension on lung structure, function, and inflammation in adult lung in vivo. The loss of surfactant surface activity influences the shape and function of cells beneath the surfactant film. For example, reduction in SP-B increases the surface tension of surfactant, influencing both cell shape and phagocytic activity of alveolar macrophages in vivo, providing support for the concept that surface tension influences cell shape and function in the alveolus (9).
Acute lung injury is a common cause of mortality and morbidity associated with high surface tension in the alveolus as a result of surfactant deficiency and dysfunction in both children and adults (10, 11). Concentrations of SP-B in BALF from patients with acute respiratory distress syndrome were decreased to 20 to 50% of normal (11). Pulmonary hypertension and intrapulmonary shunting is commonly associated with respiratory distress syndrome in neonates and adults, resulting in the impairment of ventilation perfusion or disrupting normal gas exchange.
The alveolar capillary unit consists of a region composed of both interstitial and epithelial components. In addition to alveolar pressure, other capillary compressive forces, including alveolar surface tension and tension in connective tissue fibers (12, 13), must be overcome for recruitment to occur. Because of the close apposition of the alveolar cell surfaces and epithelial component of vessels of the microcirculation, changes in surface tension may influence capillary blood flow via transmitted forces.
In the present study, we sought to test the hypothesis that decreased activity of pulmonary surfactant alters alveolar capillary shape that in turn influences alveolar capillary blood flow and gas exchange in vivo. For these studies, a mouse model in which the expression of SP-B in respiratory epithelial cells was conditionally controlled in vivo was used.
Protocols were approved by Animal Care and Use Committee at the Cincinnati Children's Research Foundation.
The conditional transgenic mice [CCSP-rtTA, (tetO)7 SFTPB/Sftpb−/−] were previously generated (7, 8) from two transgenic mouse lines (14, 15): (1) (CCSP-rtTA) mice expressing reverse tetracycline transactivator transcription factor (rtTA) under the control of a 2.3-kb element from the rat Clara cell secretory protein (CCSP) gene promoter that confers expression in both conducting epithelial cells and alveolar epithelial type II cells; and (2) [(tetO)7 SFTPB] mice, in which the human SP-B cDNA was expressed under the control of the tetracycline response elements. CCSP-rtTA/(tetO)7 SFTPB mice were bred with Sftpb+/− mice to generate CCSP-rtTA, (tetO)7, SFTPB/Sftpb−/− mice (7, 8). Because Sftpb−/− mice die perinatally unless the SP-B transgene is expressed, conditional transgenic mice were maintained on doxycycline-containing food from Embryonic Day 1 until 7 weeks of age. Doxycycline was removed for 4 days for the present studies. Control mice were maintained under continuous treatment with doxycycline and studied at 7 weeks of age. A separate group of 7-week-old animals, in which mice consumed no doxycycline (Off Dox) for 4 days and were then placed back on doxycycline (On Dox), were studied to assess the reversibility of SP-B–dependent changes. In these mice, SP-B was significantly decreased on Day 1 of withdrawal and was almost undetectable by Western blot analyses (8) 2 days after removal from doxycycline. Severe respiratory failure develops in these animals 4 days after removal from doxycycline, a time at which SP-C, and surfactant phospholipid content and composition, are unaltered (7, 8).
Surface activity of isolated large aggregate surfactant was measured with a captive bubble surfactometer (16, 17). The concentration of each sample was adjusted to 7 nmol phospholipid/μl, and 3 μl of the sample was applied to the air–water interface of a 25-μl volume bubble by microsyringe. Equilibrium surface tension was measured at 300 s and then bubble pulsation was started. The minimum surface tension was measured after 80% bubble volume reduction at a rate of 12 cycles/minute.
Sequential measurements of percent oxygen saturation of arterial hemoglobin (SpO2) were made in room air by using a small-animal pulse oximeter (MouseOx; Starr Life Sciences Corp., Allison Park, PA) according to the manufacturer's instructions.
To demonstrate the narrowed alveolar capillaries observed in the absence of SP-B associated with high surface tensions of surfactant in vivo, 3.2-μm polystyrene yellow-green fluorescent-labeled microbeads (Molecular Probes, Eugene, OR) were prepared. After intraperitoneal injection of heparin (250 U/mouse), mice for the groups of 0 d control, 2 d Off Dox, 3 d Off Dox, and 4 d Off Dox were anesthetized with ketamine/xylazine (0.16 mg/g, 8 μg/g body weight). Under additional local anesthesia with lidocaine, a 28-G venous catheter (Strategic Application, Inc., Libertyville, IL) was inserted into the jugular vein and 200 μl PBS with 0.5% NP-40 (Sigma, St. Louis, MO) containing 1.5 × 105 of 3.2-μm microbeads was infused by Auto Syringe Infusion Pump (Baxter, Deerfield, IL) at the rate of 20 μl/minute. Anesthetized mice were killed by exsanguination 5 minutes after microbead infusion, and lung homogenate was incubated for 3 days in 4 M KOH with 0.05% Tween 20 (Sigma) at 37°C on a shaker to completely dissolve the tissue, followed by 20 minutes of centrifugation at 2,000 × g. The pellets were rinsed with 2 ml PBS containing 0.5% NP-40 and centrifuged three times to isolate microbeads (18). The number of microbeads isolated from lung homogenate and in the syringe remaining after the microbead infusion was counted using a hemacytometer under light microscopy (19), and was calculated as % microbeads in lung tissue relative to infused microbeads.
To distinguish the influence of hypoxia on pulmonary vasoconstriction from that of SP-B deletion, control mice were kept in the box in which O2 was maintained at 16 ± 2% using mixture of air and N2 flow. On Day 3, the pulmonary deposition of infused 3.2-μm microbeads was studied as described above while placing the nose and mouth in 16% O2. SpO2 was monitored during hypoxia and during the infusion of microbeads.
In addition to analyzing the percentage of recovery of microbeads in lung tissue, the presence of microbeads in alveolar capillaries was studied morphologically by electron microscopy and by light microscopy. For electron microscopy, tissue inflation-fixed at 25 cm H2O with 4% formaldehyde (Electron Microscopy Sciences, Washington, PA) was post-fixed with Karnovsky fixative (20), stained, and embedded as previously described (21), and ultra-thin 90-nm sections were prepared. To prevent the microbeads from dissolving in xyline or being damaged by cutting, we used frozen fixed lung tissue and thicker sections for light microscopy. For the frozen section study, inflation-fixed tissues were embedded in a compound for cryo-tissues (Tissue-Tek O.C.T.; Sakura Finetek, Torrance, CA) after dipping into a series of sucrose-PBS solutions. Tissues were snap-frozen at −70°C, sectioned to 10 μm thickness, and stained by Diff-quick (Dade Behring, Newark, DE). Furthermore, the presence of fluorescently labeled microbeads in the entire lung was demonstrated by fluorescence microscopy (Ax 10 Plan 2; Zeiss, Göttingen, Germany).
Intravascular blood coagulation was inhibited by an intraperitoneal injection of heparin (250 U/mouse) 30 minutes before anesthetization with intraperitoneal injection of pentobarbital sodium. The chest cavity was opened and an incision made on the auricle of the left atrium. A 20-gauge blunt needle was tied into the trachea, and the lung was slowly inflated to 25 cm H2O as the total lung capacity. Inflation pressure was then reduced to 5 cm H2O and maintained during perfusion fixation with 2.5% glutaraldehyde (glutaraldehyde, EM grade; Electron Microscopy Sciences) in PBS (9, 22, 23). The volume of the fixed lung was determined by the displacement method (13, 16). Fixed tissues were divided into six lobes and cut into 1 to 2 mm3 pieces for post-fixation in 2% osmium tetroxide in 0.1 M sodium cacodylate buffer with 1.5% potassium ferrocyanide and 4% uranylacetate solution, and then dehydrated by a series of ethanol solutions and propylene oxide. Two pieces of fixed tissue were blindly selected from each lobe by technical personnel who were not involved in this study. Fixed tissues were embedded in resin (Embed 812; Electron Microscopy Sciences). Ultrathin (90-nm) sections were prepared by ultra microtome (Ultra cut E; Reichert-Jung, Vienna, Austria) and examined by using an electron microscope (H7600; Hitachi High-technologies America, Pleasanton, CA).
Mice were injected with sodium pentobarbital (100 mg/kg, intraperitoneally) and placed in a container filled with 100% oxygen to ensure complete collapse of the alveoli by oxygen absorption. The mice were killed by exsanguination, and the trachea cannulated and connected by a syringe to a pressure sensor. After opening the diaphragm, lungs were inflated in 100-μl increments every 10 seconds to a maximum pressure of 30 cm H2O and deflated to generate pressure–volume curves (7).
Data are expressed as mean ± SEM. Comparisons among groups were made by ANOVA with Tukey-Kramer multiple comparison test. Significance was accepted at the 5% level.
Deletion of the murine Sftpb causes respiratory failure at birth (3, 4) that can be rescued by the conditional expression of the human SP-B cDNA in alveolar epithelial cells (6–8). Removal of CCSP-rtTA, (tetO)7, SFTPB/Sftpb−/− mice (hereafter termed SP-B–deficient mice) from doxycycline (Off Dox) for 24 hours reduced in SP-B levels (8) and increased the minimal surface tension of surfactant isolated from their BALF (Table 1). While arterial oxygen saturation (SpO2) was similar in wild-type Sftpb+/+ and control mice during continuous treatment with doxycycline, removal of doxycycline significantly decreased SpO2 (Figure 1). To assess whether depletion of SP-B altered alveolar capillary diameter in vivo, microbeads of 3.2-μm diameter were infused through the jugular vein. These microbeads normally pass through alveolar capillaries and are not trapped in the lung. The fraction of microbeads trapped in lung tissue relative to that infused was calculated (Figure 2A). As expected, 3.2-μm microbeads were barely detectable in the lungs of control mice. In contrast, 25% (2–3 d Off Dox) to 60% (4 d Off Dox) of the infused microbeads were trapped in the lungs of SP-B–deficient mice. While these small diameter microbeads exit the lung in control mice (Figure 2C), they were readily observed within the alveolar capillaries of the SP-B–deficient mice (Figure 2D). The 3.2-μm microbeads trapped in the alveolar capillaries were detected by electron microscopy (Figure 2B). As shown in Figure 1, SpO2 was 87% in mice 3 days after SP-B deletion. To see whether hypoxia induced pulmonary vasoconstriction might contribute to pulmonary vascular protrusion, deposition of infused 3.2-μm microbeads was studied in the control mice during exposure to 16% O2 (Figure 2A, control 3 d hypoxia). SpO2 was kept at the targeted level and was 87.5 ± 2.6% on Day 3 in hypoxia. The recovery of microbeads was not significantly different from that of controls in room air and was significantly lower than that in SP-B–deficient mice at 2, 3, and 4 days Off Dox. Thus, retention of microbeads in SP-B–deleted mice was not caused by hypoxia-induced pulmonary vasoconstriction. The presence of fluorescently labeled 3.2-μm microbeads in the lungs of SP-B–deficient mice was demonstrated by fluorescence microscopy (Figure 3B), being readily detected throughout the alveolar walls. In contrast, only a few microbeads were detected in the control mice lung (Figure 3A). Thus, the reduction of SP-B resulted in trapping of 3.2-μm microbeads in the alveolar capillaries.
Electron microscopy of the lung demonstrated that reduction in SP-B was associated with changes in alveolar capillary shape (Figure 4). In control mice continuously treated with doxycycline (Figure 4A), or in mice rescued by resumption of doxycycline (Figures 4E and 4F), alveolar capillary walls protruded into the alveolar spaces. Flattening of alveolar capillaries was seen 2, 3, and 4 days after discontinuation of doxycycline (Figures 4B–4D), during which the surface tension of isolated surfactant was increased as assessed by captive bubble surfactometer. Because the apparent size of the alveolar capillaries varies under electron microscopy depending on where and with what angle the sections were made, the degree of alveolar capillary flattening (Figure 5A) was assessed by drawing a line between the two edge points of the capillary at the alveolar wall (line a). Line b was drawn at 90° from the midpoint, and the percentage of distance above the line a (line c), relative to the distance between capillary walls (line b), was determined (c ÷ b × 100) for the first five capillaries detected in blindly selected areas. Four mice were studied for each group, and twenty capillaries were analyzed for each group. The extent of protrusion of capillaries into the airspace was significantly decreased 2 days after removal from doxycycline and continued to decrease thereafter. The observed compression of alveolar capillaries was reversed after resumption of doxycycline (Figure 5B). Thus, decreased SP-B reduces surfactant function and is temporally associated with compression of alveolar capillaries. Re-expression of SP-B reversed the abnormalities in both surfactant function and alveolar capillary shape.
Two days after removal of the conditional SP-B mice from doxycycline, lung pressure–volume curve was not significantly different from that of control mice (Figure 6). Three days after removal from doxycycline, lung volumes at maximum pressure (total lung volume) were similar to those of controls; however, lung volumes were significantly decreased on the deflation limb of the pressure–volume curve at this time, indicating changes in alveolar stability. Four days after removal from doxycycline, total lung volumes were significantly decreased. These studies indicated that after reduction of SP-B there is a period during which surfactant surface activity was altered before significant changes in lung volumes or compliance were observed. During perfusion fixation for electron microscopic analyses, lungs were inflated and maintained at 5 cm H2O airway pressure. The total lung volumes, measured after perfusion fixation by the displacement method, were similar in all the groups (Table 2). In our previous studies, lung edema in lung histology was observed 4 days after removal from doxycycline in this model (8). In the present study, removal from doxycycline for 2 to 3 days did not influence lung weight. The lung weight was significantly increased 4 days after removal from doxycycline (0.19 ± 0.01 g) compared with control (0.15 ± 0.01 g, P < 0.05). Likewise, the ratio of lung weight to body weight was significantly increased on Day 4 after deletion of Sftpb (Figure 7), likely representing changes related to pulmonary edema.
In previous studies from this laboratory, inflammation and alveolar capillary protein leaks were not detected until 3 days after removal from doxycycline (8). To assess whether endothelial cell integrity was perturbed during the study period all the alveolar capillaries in randomly selected 200-mesh grid of 4 unit-square lung areas were examined by electron microscopy in a blinded fashion. No ultrastructural abnormalities were detected in the capillary endothelium 4 days after removal from doxycycline (data not shown).
The present study provides new insights into the pathogenesis of intrapulmonary shunting, commonly associated with both infantile and adult respiratory distress syndromes. Because of the remarkable structure of alveolar walls, pulmonary capillaries are located in close apposition to the surfactant film lining the respiratory epithelium. In the present study, decreased surfactant activity was caused by conditional removal of SP-B, resulting in flattening of the pulmonary microvasculature and arterial oxygen desaturation at the time before onset of lung inflammation, protein leaks, lung edema, and alveolar surface area alterations. The present study supports the concept that the reduction of alveolar surface tension related to pulmonary surfactant plays a critical role in the maintenance of alveolar capillary shape, which in turn influences pulmonary blood flow. Physical forces caused by the increased surface tension related to surfactant dysfunction altered alveolar capillary shape and pulmonary blood flow, causing hypoxemia.
Pulmonary arteries supply the alveolar capillaries that drain into pulmonary veins. The diameter of the alveolar capillary segment is just sufficient for the passage of red blood cells. These microvascular segments are relatively short, forming a dense network that provides an almost continuous sheet of red blood cells coursing through the alveolar walls, as required for efficient gas exchange. The air–blood barrier includes a basement membrane, fibroblasts, and elastin, providing little structural support for the microvessels. Capillary walls come into close apposition to the airspaces, creating a distance of approximately 0.2 to 0.4 μm across which gas exchange occurs, as indicated by the present ultrastructural studies. In the absence of surface-active surfactant, the high surface tension at the air–liquid interface in the alveoli creates collapsing forces. A surfactant film covers the entire surface of the alveolus, lowering surface tension to nearly 0 mN/m (2). The present studies provide physiologic and ultrastructural evidence that reduction of SP-B decreases surfactant activity which likely influences alveolar surface tension. Reduction of SP-B decreased surfactant activity, causing changes in alveolar capillary shape and size that, in turn, reduced oxygenation.
Excised perfused lung has been previously used to demonstrate the effects of surface tension on capillary resistance (24–26). In these studies, surface tension was altered by filling airways with saline or by nebulizing substances causing increased surface tension within the alveolus. Bachofen and Schurch demonstrated the influence of increased alveolar surface tension on alveolar capillary shape in vitro by lavaging airways with detergents (27). Hypoxemia occurs by multi-faced surface tension–related disease process, including decreased surface area (27) and lung edema, that might be difficult to segregate in other models. The conditional SP-B mice used in the present study have been previously well characterized, providing a model in which surfactant activity can be influenced in vivo (8). Genetic and biophysical studies support the important role of SP-B in surfactant function (1). Sftpb−/− mice die of respiratory failure at birth (4), while deletion of SP-A (28), SP-C (29), and SP-D (30) (Sftpa−/−, Sftpc−/−, and Sftpd−/− mice) does not alter perinatal survival in the mouse. Conditional replacement of SP-B in the Sftpb−/− mice provides an adult mouse model in which surfactant function can be readily manipulated without disrupting phospholipid synthesis or altering other surfactant component composition (6–8). SP-B expression and lung function are rapidly controlled by the addition of or removal from doxycycline. In the present study, SP-B levels in BALF decreased to 35% of that in controls within 24 hours after removal from doxycycline, and were further reduced thereafter. Reduction in SP-B was associated with significant increases in minimum surface tension of isolated surfactant within 24 hours after removal from doxycycline in this model. Pressure–volume curves were not altered and lung inflammation was not detected 2 days after removal of doxycycline, at which time surface tension of the isolated surfactant was significantly increased. Thus, the observed changes in alveolar capillary shape and oxygenation occurred before the onset of decreased lung volume, compliance, or inflammation. Consistent with this observation, endothelial cell injury was not detected 4 days after discontinuation of doxycycline in the present model.
Since capillaries undergo compression without change in wall thickness when alveolar pressures increased (27), the caliber of the alveolar capillaries depends on the level of lung inflation. In the present study, airway pressure was maintained at 5 cm H2O during perfusion fixation, a pressure known not to cause capillary flattening (31). The pressure–volume curves demonstrated that lung volume at 5 cm H2O were 68% of total volume (or lung volume at 30 cm H2O) in the control group, and were decreased to 51% after discontinuance of doxycycline. Likewise, the volume of fixed lung determined by the displacement method was unchanged after deletion of SP-B, suggesting that the flattened alveolar capillary shape in SP-B–deleted mouse lung was not caused by lung over-expansion during perfusion fixation.
Acute lung injury is a common cause of mortality and morbidity in both neonates and adults and is associated with severe surfactant dysfunction. Surfactant SP-B levels in BALF are decreased in premature neonates with respiratory distress syndrome and in adult patients with acute lung injury. Surfactant activity is decreased during lung injury, leading to alveolar collapse and ventilation perfusion abnormalities. Abnormalities in surfactant composition, the presence of surfactant inhibitory proteins, and decreased abundance of large aggregate surfactant have been observed in both animal models and patients with acute lung injury (32, 33). Hypoxemia accompanies these disorders caused by decreased ventilation, as well as ventilation perfusion mismatches in which nonventilated areas of the lung are perfused (34). Likewise, ventilated areas with abnormal surfactant may be underperfused with resultant arterial hypoxemia. Severe hypoxia induces pulmonary vasoconstriction mediated by pulmonary arterial vascular small muscle contraction (35–37). Hypoxia induced vasoconstriction when oxygen saturation is under 50% in all species so far examined (36, 38, 39). The decreased alveolar capillary size presently observed in SP-B–deleted mice is distinct from that caused by hypoxic vasoconstriction. The present study demonstrates that surfactant dysfunction generates physical forces within the alveolar walls that alters alveolar capillary size and decreases alveolar capillary blood flow with resultant arterial hypoxemia.
This work was supported by National Institutes of Health grants HL 061646 (to M.I., T.E.W., and J.A.W.) and HL56285 (to T.E.W.).
Originally Published in Press as DOI: 10.1165/rcmb.2008-0359OC on February 6, 2009
Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.