Role of the Fas/FasL system in a model of RSV infection in mechanically ventilated mice

Abstract

Infection with respiratory syncytial virus (RSV) in children can progress to respiratory distress and acute lung injury necessitating mechanical ventilation (MV). MV enhances apoptosis and inflammation in mice infected with pneumonia virus of mice (PVM), a mouse pneumovirus that has been used as a model for severe RSV infection in mice. We hypothesized that the Fas/Fas ligand (FasL) system, a dual proapoptotic/proinflammatory system involved in other forms of lung injury, is required for enhanced lung injury in mechanically ventilated mice infected with PVM. C57BL/6 mice and Fas-deficient ("lpr") mice were inoculated intratracheally with PVM. Seven or eight days after PVM inoculation, the mice were subjected to 4 h of MV (tidal volume 10 ml/kg, fraction of inspired O(2) = 0.21, and positive end-expiratory pressure = 3 cm H(2)O). Seven days after PVM inoculation, exposure to MV resulted in less severe injury in lpr mice than in C57BL/6 mice, as evidenced by decreased numbers of polymorphonuclear neutrophils in the bronchoalveolar lavage (BAL), and lower concentrations of the proinflammatory chemokines KC, macrophage inflammatory protein (MIP)-1α, and MIP-2 in the lungs. However, when PVM infection was allowed to progress one additional day, all of the lpr mice (7/7) died unexpectedly between 0.5 and 3.5 h after the onset of ventilation compared with three of the seven ventilated C57BL/6 mice. Parameters of lung injury were similar in nonventilated mice, as was the viral content in the lungs and other organs. Thus, the Fas/FasL system was partly required for the lung inflammatory response in ventilated mice infected with PVM, but attenuation of lung inflammation did not prevent subsequent mortality.

Fig. 1.

C57BL/6 mice and Fas-deficient lpr mice had similar clinical scores and weight loss after infection with pneumonia virus of mice (PVM). A: mean clinical score of C57BL/6 (closed circles) and lpr (open circles) mice that were intratracheally infected with 3.03 × 10⁴ copies of PVM and examined at daily intervals according to a standardized scoring system (Table 1). B: change in body weight expressed as the percent of the original weight of C57BL/6 (black bars, n = 26) and lpr (gray bars, n = 29) mice that were treated intratracheally with 3.03 × 10⁴ copies of PVM and evaluated 7 or 8 days later. There was no difference in clinical scores or body weight in the C57BL/6 and lpr mice. Means ± SE.

Fig. 2.

The enhanced inflammatory response to mechanical ventilation (MV) in the setting of PVM infection is Fas dependent. C57BL/6 (black bars) and lpr (gray bars) mice were treated intratracheally with 3.04 × 10⁴ copies of PVM. Seven days after PVM instillation, the mice were allowed to breath spontaneously (PVM + SB) or were exposed to 4 h of MV [PVM + 4 h MV; tidal volume (Tv) 10 ml/kg, fraction of inspired O₂ (Fi_O2) = 0.21, and positive end-expiratory pressure (PEEP) = 3 cm H₂O]. Total polymorphonuclear neutrophils (PMN, A), KC (B), macrophage inflammatory protein (MIP)-2 (C), MIP-1α (D), interferon (IFN)-γ (E), and IL-10 (F) were measured in the bronchoalveolar lavage fluid (BALF). Note that MV resulted in an increase in neutrophils, KC, and MIP-2 in the C57BL/6 mice, and this was attenuated in the lpr mice. Means ± SE of 6–9 mice/group. *P < 0.05.

Fig. 3.

The lpr mice show attenuated permeability responses to PVM infection independent of MV. C57BL/6 (black bars) and lpr (gray bars) mice were treated intratracheally with 3.04 × 10⁴ copies of PVM. Seven days after PVM instillation, the mice were allowed to breath spontaneously (PVM + SB) or were exposed to MV (PVM + 4 h MV; Tv 10 ml/kg, Fi_O2 = 0.21, and PEEP = 3 cm H₂O). Four hours after the onset of MV, all mice were killed, and the concentrations of the permeability markers total protein (A) and α₂-macroglobulin (B) and the epithelial cell injury marker receptor for advance glycation end-products (RAGE, C) were measured in the BALF. None of these parameters was enhanced by MV, although the values were lower in the lpr mice. Means ± SE of 6–9 mice/group. *P < 0.05.

Fig. 4.

Apoptotic response. C57BL/6 (black bars) and lpr (gray bars) mice were treated intratracheally with 3.04 × 10⁴ copies of PVM. Seven days after PVM instillation, the mice were allowed to breath spontaneously (PVM + SB) or were exposed to 4 h of MV (PVM + 4 h MV; Tv 10 ml/kg, Fi_O2 = 0.21, and PEEP = 3 cm H₂O). There was a trend toward lower caspase-3 in the lpr mice, but this did not reach statistical significance (A). A similar nonsignificant pattern was seen with poly(ADP-ribose) polymerase (PARP, B). Note that neither caspase-3 or PARP activities were enhanced by MV. Means ± SE of 6–9 mice/group. Immunohistochemistry studies (C) suggested that the cells staining for caspase-3 (brown reaction product, arrows) were located in the alveolar spaces.

Fig. 5.

The lpr mice infected with PVM had less ventilator-induced lung injury compared with C57BL/6 mice. Representative hematoxylin and eosin-stained lung tissue sections from C57BL/6 and lpr mice that were infected with PVM and allowed to breath spontaneously (PVM + SB, row on top) or subjected to 4 h of MV (PVM + MV, row on bottom) 7 days after PVM infection. MV led to an increase in inflammatory cells in the alveolar spaces, thickening of the alveolar septa, and peribronchial wall thickening in C57BL/6 mice. These changes in response to MV were less prominent in the lpr mice. Magnification ×400.

Fig. 6.

Physiological response to MV in the setting of PVM infection. A: Kaplan-Meier curve and analysis of survival of C57BL/6 (closed circles) and lpr (open circles) mice during MV on day 8 after infection with PVM. There was a significant decrease in survival in the lpr mice, all of which died between 1 and 3.5 h after the onset of ventilation. This decrease in survival, however, was not associated with changes in the heart rate (B), end-tidal CO₂ (EtCO₂, C), or peak airway pressures (D). Means ± SE of 7 mice/group. To determine if the difference in survival was because of differences in cardiac inflammation, we measured leukocyte infiltration in the heart by staining the heart tissue with the leukocyte antigen CD45 and then calculating a “leukocyte ratio” as the total no. of CD45-positive pixels divided by the total no. of pixels of heart tissue (E). There was no difference in the leukocyte ratio of C57BL/6 and lpr mice. Means ± SE of 3 mice/group.