Main

In recent years, the role of pulmonary surfactant in airway diseases other than respiratory distress syndrome has become a focus of investigation(1–4). Aside from two recent reports(3, 4), little information has been published regarding the possible role of surfactant dysfunction in respiratory tract infections. Pulmonary surfactant is known to stabilize alveoli, but more recently it has become clear that surfactant also serves to maintain patency of conducting airways(5, 6). Any inflammatory airway disease will cause leakage of plasma proteins and inflammatory mediators into the airways. This can disturb surfactant function(7–12) and might contribute to obstruction of small airways(2, 9, 13–15). We report the analysis of surfactant function in BAL fluids from RSV-infected mice. It was shown that RSV infection leads to a dysfunction of pulmonary surfactant, which coincides with an increase in the percentage of inflammatory cells of BAL fluids from infected mice. Surfactant function was also found to correlate negatively with the protein concentration of BAL. We conclude that RSV infection in mice results in an inhibition of surfactant function, most likely owing to leakage of plasma proteins into the airways. Such a surfactant dysfunction will probably increase airways resistance.

METHODS

Virus growth and titration. RSV stock solutions were obtained by inoculation of monolayers of HEp-2 cells with the virus. Cells and virus were harvested when an 80-90% cytopathic effect was observed. Cells were disrupted by sonication, and clarified virus-containing solutions were stored at -70 °C. Virus titers were determined by plaque assay in HEp-2 cells under methyl cellulose(16), and stock solutions of 106-108 pfu/mL were used for the inoculation of mice. Because RSV titers are known to decrease during storage, a small aliquot of the virus was tested again at the time of inoculation to confirm the titer.

Infection of mice. BALB/c mice, 12-15 wk old (Harlan Sprague Dawley, Indianapolis, IN), of either sex, weighing 19-28 g were infected by nasal inoculation under light anesthesia with methoxyflurane (Pitman-Moore, Mundelein, IL). Mice were inoculated with 2-4 × 50 μL of RSV stock solution, resulting in a total dose of 5 × 104 to 1 × 105 pfu (low dose) or 1 × 106 to 2 × 107 pfu(high dose) as calculated retrospectively from the virus titer determined at the time of inoculation. Mice inoculated with a similar volume of uninfected tissue culture medium served as controls. This study was approved by the Institutional Animal Care and Use Committee of the State University of New York at Buffalo.

BAL. On d 6 after inoculation, mice received a lethal injection of pentobarbital (Abbott Laboratories, Chicago, IL) intraperitoneally (100 mg/kg), and a cannula was inserted into the trachea. Normal saline solution, 2% (vol/wt) of the body weight, was instilled through the cannula. The instillation was done under a positive pressure of 30 cm H2O, ensuring adequate but gentle filling of the lungs with lavage fluid. By applying a negative pressure of 20 cm H2O, lavage fluid was retrieved from the lungs. The same fluid was instilled six times to concentrate the BAL fluid. With this method, 80-90% of the instilled volume was recovered. A 40-μL aliquot of BAL fluid was cytocentrifuged on a microscope slide to allow differentiation of white blood cells (Cytospin3, Shandon, Pittsburgh, PA). Slides were fixed with spray-fixative (Shandon) and stained with Wright-Giemsa stain (CMS, Houston, TX). Cells were differentiated under light microscopy based on morphologic and staining properties. After aliquots of BAL fluid required for analysis of cells, proteins, and phospholipids had been removed, the remainder of the BAL fluid was centrifuged at 40 000 × g for 1 h at 4 °C to concentrate pulmonary surfactant. A volume of supernatant, corresponding to 75% of the centrifuged fluid, was removed and saved for studies assessing the presence of surfactant-inhibiting agents (see below). The pellet, containing pulmonary surfactant, was resuspended in the remaining supernatant and assayed for surfactant function as described below.

Protein and phospholipid assays. Previous studies(7–12) have demonstrated that plasma proteins may seriously interfere with surfactant function. Therefore, protein concentrations of the BAL fluid were assayed using the method described by Lowry et al.(17). As a measure of surfactant content, the phospholipid concentrations of BAL samples were determined. Lipid extraction was performed according to the method of Bligh and Dyer(18), and the lipid phosphorus content was assessed using the method described by Chen et al.(19).

Surfactant function. The capillary surfactometer(12, 15) was used to evaluate the ability of pulmonary surfactant in BAL fluid to maintain patency of a narrow tube. Briefly, 0.5 μL of the concentrated BAL fluid was inserted into the narrowed middle section of a specially prepared glass capillary. Pressure, continuously recorded, was slowly raised at one end of the capillary, eventually resulting in extrusion of the liquid, whereby pressure promptly dropped to zero. The BAL fluid would not return to the narrow section if it contained normally functioning pulmonary surfactant. Consequently, the capillary remained patent and pressure stayed at zero. If, however, surfactant function had been disrupted, the liquid would return to the narrow section, and pressure would again increase. Pressure was recorded for 120 s after the initial extrusion of sample fluid, and the total time that pressure was zero, demonstrating capillary patency, was expressed as a percentage of the 120-s period.

Surfactant-inhibiting agents. Disrupted surfactant function, detected with the capillary surfactometer, may be due to a decreased concentration and/or an abnormal composition of the surfactant. Inhibition by proteins or other agents may also interfere with the surfactant function. Therefore, we examined whether the supernatant removed after centrifugation of BAL fluids contained surfactant-inhibiting agents. The supernatant was added to a well characterized surfactant preparation, CLSE (kindly donated by ONY Inc., Amherst, NY), in a volume that diluted the original CLSE concentration of 33 mg/mL to 1 mg/mL, and this mixture was tested for its ability to maintain capillary patency. When CLSE is diluted with saline solution to 1 mg/mL, patency is maintained at 100% of the 120-s study period.

Data analysis. Statistical analyses were performed using the Statview 4.5 (Abacus Concepts, Berkeley, CA) program for the Macintosh. Values were expressed as means ± SEM. Differences between three groups were analyzed using the nonparametric Kruskall-Wallis analysis of variance. For comparisons between two groups the Mann-Whitney U test was used. To evaluate a correlation between parameters, linear regression analysis was performed. A p value <0.05 was considered to indicate significance.

RESULTS

Mice were studied on d 6 after inoculation, because inflammatory response reaches a peak around this day (our unpublished data). BAL fluids were obtained from 66 mice (30 controls, 12 infected with the low dose, and 24 with the high dose of RSV). Mean volumes of BAL fluids retrieved were similar for all three groups (p = 0.22): 0.422 ± 0.01 mL (controls), 0.394 ± 0.009 (low dose), and 0.404 ± 0.01 (high dose).

Cellular content of BAL fluids. Differentiation of white cells in BAL fluid was performed on 55 samples (Table 1). Total cell counts could not be determined because of the limited volume of BAL samples, therefore relative percentages of cell types were given rather than absolute numbers. In 11 animals, the volume of BAL samples was not sufficient to allow for all analyses, therefore evaluation of cell distribution was not performed. In BAL fluids of mock-infected control mice, mainly mononuclear phagocytes were seen. RSV infection resulted in an increase in the percentage of both lymphocytes and neutrophilic granulocytes. This effect was more pronounced in the high dose group than in the low dose group, and the observed differences were all statistically significant.

Table 1 Cellular content of BAL fluids
Full size table

Protein and phospholipid concentrations.Figure 1 shows the mean protein and phospholipid concentrations of BAL fluids obtained on d 6 after inoculation. A dose-dependent increase in protein concentration was found; differences between groups were highly statistically significant (p < 0.0001). Differences in phospholipid concentrations among the three groups were not statistically significant.

Figure 1
figure 1

Protein and phospholipid concentrations of BAL fluids on d 6 after inoculation. Concentrations of proteins (A) and phospholipids (B) were determined. ***p < 0.0001vs control and high dose group. Differences in phospholipid concentrations were not significant.

Full size image

Surfactant function. Concentrated BAL fluids from RSV-infected mice had a reduced ability to maintain capillary patency. The degree of dysfunction increased with the titer of virus inoculated, with all differences between groups being statistically significant (Fig. 2). Whereas CLSE diluted with saline solution to 1 mg/mL kept the ability to maintain capillary patency 100% of the study period, diluting CLSE with the supernatant of the centrifuged BAL fluids from RSV-infected mice resulted in a significant dose-dependent inhibition of this ability (Fig. 3).

Figure 2
figure 2

Surfactant function in mice infected with a high or a low dose of RSV, compared with control animals. BAL fluids were centrifuged, and the surfactant-containing pellet was resuspended and examined for its ability to maintain patency of a narrow capillary (“open in%”).*p < 0.01 vs control group and high dose group,***p < 0.0001 vs control group.

Full size image
Figure 3
figure 3

Presence of surfactant-inhibiting agents in BAL fluid. Supernatant of centrifuged BAL fluids was mixed with CLSE, and the ability of this mixture to maintain capillary patency was studied. **p = 0.002 vs control group, ***p < 0.0001vs control group or low dose group.

Full size image

Surfactant function correlated with percentage of inflammatory cells and protein concentration. Surfactant function correlated negatively with the combined percentage of lymphocytes and neutrophilic granulocytes in BAL fluid (r = -0.67, p = 0.0004; Fig. 4A). Surfactant function also showed a negative correlation with protein concentrations of BAL fluids of RSV-infected mice (r = -0.44,p = 0.007; Fig. 4B). Thus, surfactant function was inhibited progressively as protein concentrations in BAL increased. The percentage of inflammatory cells showed a positive relation to the protein concentration (r = 0.51, p = 0.004; Fig. 4C).

Figure 4
figure 4

Correlation of surfactant function with percentage of inflammatory cells and protein concentration in BAL of RSV-infected mice. Using linear regression, a negative correlation was found between surfactant function and the percentage of inflammatory cells in BAL fluid (A). Surfactant function was also inversely related to protein content of BAL fluid(B), and a positive relationship was found between protein content and the percentage of inflammatory cells in BAL fluid (C).

Full size image

DISCUSSION

Results of our studies clearly show that surfactant function is adversely affected by RSV infection. It can be argued that the BAL will yield material originating more from alveoli than from terminal conducting airways. However, alveolar cells type II constitute the only source of pulmonary surfactant. The material from those cells will form a film at the alveolar air-liquid interface that will exert high surface pressure, particularly when the film is compressed during expiration, and that will cause the film to be extruded into the cylindrical airway. Thus, if the alveolar film is dysfunctioning, the film lining the cylindrical airway will also be of abnormal quality. Even though the surfactant available for examination may not have originated from the cylindrical surface, it will still reveal the physical status of the film lining the conducting airway.

The surfactant dysfunction we observed in BAL fluids could have been due to a decreased quantity or a poor quality of the pulmonary surfactant, or both. However, a decreased quantity of surfactant is not a likely explanation for the observed dysfunction, because the phospholipid concentration in the BAL fluid was not reduced by the infection. Instead, we postulate that the surfactant quality was impaired by inhibiting agents. This assumption is based on the observation that the BAL supernatant clearly had an inhibiting effect on the surfactant preparation CLSE, particularly when the supernatant originated from mice infected with the high dose of RSV. Because the BAL was carried out under standardized conditions, the concentration of surfactant and inhibitors in the lavage fluid is likely to reflect the situation in the alveoli and conducting airways. However, for an evaluation with the capillary surfactometer, the surfactant in BAL fluid had to be concentrated. The potential inhibitors were not totally removed in the concentration procedure, and therefore they could still affect the surfactant, as they would do in the airway. Protein and phospholipid concentrations of BAL fluids reported in this study were determined in aliquots removed before concentration; however, it seems reasonable to assume that these results are reflective of the concentrations in the centrifuged surfactant suspension.

The percentage of inflammatory cells (lymphocytes and neutrophils) in BAL fluids showed a negative correlation with surfactant function. Protein levels in BAL fluid also correlated negatively with surfactant function, and a positive correlation was shown between the percentage of inflammatory cells and the protein concentration. Both the percentage of inflammatory cells and the concentration of proteins reflect the severity of the inflammation. Although it is quite possible that the inflammatory cells released mediators that catalyzed hydrolysis of surfactant components, the potency of plasma proteins to inhibit surfactant function has been demonstrated before in several studies(7–12). Therefore, a likely explanation for our results is that the proteins invading the inflamed airways caused surfactant inhibition. Whether products of activated lymphocytes and neutrophils can also interfere with surfactant function merits further study.

In experiments with excised rat lungs, the need for pulmonary surfactant to maintain airway patency was demonstrated(5, 6). Removal of the endogenous surfactant by a lavage procedure increased airway resistance in this model, and when surfactant was returned, airway patency was restored. Thus, surfactant function maintains airway patency, and when a dysfunction transpires, columns of liquid are likely to block narrow conducting airways. If the dysfunction is serious, blocking of conducting airways may ultimately lead to increased airway resistance.

If surfactant dysfunction is indeed a serious consequence of RSV infection, as the present studies in mice indicate, surfactant supplementation may be beneficial to infants suffering from severe bronchiolitis. It was recently reported(20) that surfactant replacement was very beneficial for infants suffering from meconium aspiration, a condition considered to be caused, at least in part, by inactivation of surfactant(21–24). The adult respiratory distress syndrome is also felt to be partly due to surfactant dysfunction(1), and surfactant treatment has been reported to be beneficial for this condition as well(25–28). Finally, in a mouse model of influenza A pneumonia, surfactant therapy was found to improve the condition(29). Thus, surfactant supplementation appears to be beneficial in several conditions other than the neonatal respiratory distress syndrome. Our study indicates that RSV bronchiolitis might be yet another condition in which surfactant replacement might be beneficial.

In conclusion, RSV infection of BALB/c mice was shown to result in surfactant dysfunction, which correlated with the degree of inflammation in the airways. These findings support the hypothesis that surfactant dysfunction is important for the pathophysiology of RSV infection. Surfactant replacement may prove to be efficacious in alleviating respiratory symptoms in infants with bronchiolitis.