Reply to Stewart and Sundar

From the Authors:

We thank Drs. Stewart and Sundar for their interest in our work (1). Mouse models have made an invaluable contribution not only to elucidating the molecular basis of the circadian clock but, more importantly, to understanding numerous diseases across medicine. Chronobiologists have repeatedly pointed out that the translation of these findings to humans must take into account, among other things, that mice, unlike humans, are nocturnal, as we also did in the discussion of our findings (1). Our study was specifically designed to address the following questions: 1) is ventilator-induced lung injury (VILI) time of day–dependent and, if so, 2) is this time-of-day dependence driven by light/dark cycles or governed by the internal clock; and, in the latter case, 3) which molecular clocks contribute to this dependence? To this end, we used approaches that have been established in chronobiology for decades, and we refer the authors to the relevant reviews. Specifically, constant darkness (i.e., DD condition) is used to test whether a time of day–dependent phenomenon is regulated by the internal clock rather than, as suggested by Stewart and Sundar, to test for the difference between diurnality and nocturnality. Bmal1-knockout mice were used because Bmal1 is the only gene in mammals in which deletion leads to a complete loss of circadian clock function, and we accordingly refer to Bmal1 as the core clock gene in our paper. With similar rationale, a vast body of literature has previously used Bmal1-deficient mice to elucidate effects of the circadian clock on physiological functions and pathological mechanisms.

Stewart and Sundar further speculate that the applied high-tidal-volume ventilation to induce lung injury may impair cardiac function, which, in turn, may impact the release of inflammatory markers. We refer the authors to our previous work with this established ventilation strategy in which we ruled out potential confounders such as dynamic hyperinflation or hemodynamic deterioration due to injurious ventilator settings (2). For all ventilator settings, appropriate controls were performed in parallel, as detailed in our paper, so differences between groups could be unambiguously attributed to specific interventions.

Finally, Stewart and Sundar correctly mention that metabolism and pharmacodynamics, among others, are regulated by circadian rhythms and, conversely, may influence circadian rhythms. We have addressed this aspect in the Discussion section of our paper. As such, the reported regulation of ventilator-induced lung injury by the circadian clock opens up novel avenues for exploration and research, which were, however, beyond the scope of our recent work.