Prior studies demonstrated that mechanical ventilation alone, using tidal volumes of 10 mL/kg, does not independently cause significant cytokine production, neutrophil recruitment, or lung permeability in mice. However, mechanical ventilation does augment both lung inflammation and injury in the presence of a variety of pro-inflammatory stimuli, including LPS, bacterial pneumonia, and viral pneumonia [9–12]. We hypothesized that mechanical ventilation with a tidal volume of 10 mL/kg generates endogenous ligands, either classical cytokines and/or damage-associated molecular patterns (DAMPs), which are recognized by MyD88-dependent transmembrane receptors, resulting in amplification of the inflammatory response from concurrently administered poly(I:C), a TLR3 ligand. Because all known TLRs except for TLR3, as well as many early response cytokine receptors, signal via the MyD88 adapter protein, we studied the effect of mechanical ventilation in the setting of TLR3 activation in normal (WT, MyD88+/+) mice and in MyD88-/- mice. The main findings of this study were: 1) mechanical ventilation augmented cytokine expression, PMN recruitment, and lung permeability, during TLR3 activation; and 2) maximal cytokine response, PMN recruitment, and vascular permeability induced by mechanical ventilation required the MyD88 adapter protein.
Because TLR4 signals via MyD88 and has been implicated as a primary receptor for DAMPs, we tested whether mechanical ventilation at normal tidal volumes resulted in TLR4-dependent signaling. The role of TLR4 in ventilator-associated lung injury is uncertain with conflicting data in the literature. Held and co-authors previously reported that high tidal volume ventilation causes NF-κB nuclear translocation and cytokine expression in lungs isolated from C3H/HeJ mice, which have a genetic polymorphism, resulting in a non-functional TLR4 . In contrast, Vaneker and co-authors reported that low-tidal volume ventilation caused inflammation in wildtype but not in TLR4-/- mice . Other studies have looked at the role of TLR4 in mechanical ventilation; however, these studies have also used concurrent LPS administration or bacterial infection, making it difficult to separate out the contribution of mechanical ventilation on TLR4-dependent signaling [39, 40]. We found that mechanically ventilated, TLR4-/- mice concurrently exposed to the TLR3 ligand, poly(I:C), did not have significantly different cytokine concentrations, PMN recruitment, or lung permeability as compared to similarly treated WT mice. Our data support the work of Held et al that mechanical ventilation with normal tidal volumes does not generate endogenous ligands recognized by TLR4. The explanation for the different finding by Vaneker and colleagues is unclear but may relate to inadvertent LPS exposure during animal preparation or possibly an unanticipated interaction effect between the mechanical ventilation and the elevated inspired oxygen fraction (0.4) used in this study. Indeed interaction effects between ventilation and hyperoxia on inflammatory responses have been previously reported in animal models [13, 14].
Selective attenuation of the mechanical ventilation effect on cytokine expression, PMN recruitment, and lung permeability in MyD88-/- mice suggests a role for signaling via the MyD88 adapter protein in ventilator-associated lung injury. The most likely mechanism involves stretch-induced generation of endogenous ligands, which signal via MyD88-dependent receptors. Known MyD88-dependent receptors include all of the TLRs with the exception of TLR3 , the IL-1 receptor , the IL-18 receptor , and Fas (CD95) . There are no published data examining the effect of moderate tidal volume ventilation on signaling by these receptors; however, there is one report of increased IL-1α and IL-1β release induced by low tidal volume ventilation . Identification of the MyD88-dependent pathways by which mechanical ventilation modulates innate immune responses could provide therapeutic avenues to reduce the risk of acute lung injury in mechanically ventilated patients; however, a thorough examination of these different pathways is beyond the scope of the current work.
Interestingly, mechanical ventilation is reported to increase expression of TLR4 and its associated co-adapter, CD14 [39, 40, 45, 46]. Thus, upregulation of TLRs is one potential mechanism by which mechanical ventilation could modify inflammatory responses. Although we did not specifically measure TLR3 expression in the current study, review of our previously published microarray study indicates that TLR4 mRNA but not TLR3 mRNA is increased in mice mechanically ventilated with a tidal volume of 10 mL/kg as compared with control mice . Therefore, upregulation of TLRs is unlikely to be the only mechanism through which mechanical ventilation can amplify inflammation during concurrent exposure to TLR ligands.
An important additional observation from these experiments was that ventilator-induced augmentation of PMN recruitment and lung permeability was only partially attenuated in MyD88-/- mice, indicating the presence of parallel MyD88-independent pathways. One possible explanation for this result is stretch-induced changes in endothelial cells. High tidal volume ventilation increases lung endothelial cell P-selectin and focal adhesion molecule expression . These changes may promote vascular PMN demargination independent of TLRs and their respective adapter proteins. Another possibility is that the combination of mechanical ventilation and TLR3 activation causes cellular injury sufficient to release mitochondrial-derived, formylated proteins, which have recently been reported as a mechanism for neutrophil recruitment in the setting of sterile tissue injury .