Rational Our group has recently shown that inhalation of hydrogen sulfide (H2S) during mechanical ventilation protects against VILI [1] . Here we explored the gene expression profile of this model in order to identify underlying molecular mechanisms. Methods C57/BL6 mice were randomized into four groups (n = 5/group). Two groups of control animals were breathing spontaneously air with or without H2S (80 ppm). Two groups were ventilated with synthetic air with or without H2S as described previously [1] . Lung injury was assessed by an acute lung injury scoring system in hematoxylin and eosin (H& E) stained tissue sections. Bronchoalveolar lavage fluid (BALF) was analysed for inflammatory infiltration. Micro array analysis was performed with RNA from lung tissue samples. Genes with different expression (ANOVA p ⩽ 0.05; fold induction or suppression ⩾2.2) were considered for further analysis. Gene expression of Atf3 was validated by sqPCR and Western Blot. The functional relevance of Atf3 up-regulation was validated in Atf3-Morpfolino knockout mice ventilated in presence of H2S. Results Mechanical ventilation led to lung injury, i.e., H& E-stained samples revealed clear thickening of alveolar walls and increased amounts of inflammatory cell infiltrates in lung sections and BALF from animals ventilated with synthetic air. In contrast, the presence of H2S prevented lung injury as well as neutrophil and macrophage infiltration. Spontaneous breathing control animals regardless of the presence or absence of H2S showed no significant effects on lung injury or inflammation. H2S application led to differential expression of 137 genes. Among these, Atf3 was chosen for further analysis. In contrast to the protective effects of H2S in ventilated mice, Atf3-Morpholino mice developed despite the presence of H2S sings of lung injury, i.e., enlarged alveolar walls, increased inflammatory cell infiltrates, and occurrence of sporadic blood vessel leakages. Further analysis revealed significant increase of the VILI score in the Atf3-Morpholino group as compared to the air control and control-Morpholino groups. However, histology demonstrated that down-regulated Atf3 protein synthesis failed to completely abolish H2S mediated protection in VILI. Thus, Atf3 significantly contributes to the protective effect of H2S, but seems not to be the only player. Likewise and with respect to the inflammatory response, inhibition Atf3 protein synthesis in H2S ventilated mice resulted in significant increase of neutrophils in BALF as compared to H2S ventilation alone, but neutrophil counts were still lower as compared to injurious ventilation without any treatment. Conclusions (1) VILI can be prevented by application of H2S. (2) 137 genes were differentially expressed in the presence of H2S during mechanical ventilation. (3) Our data suggest that the Atf3 signaling pathway significantly contributes to the lung protective and anti-inflammatory effects of H2S.