The association between regional transcriptome profiles and lung volumes in response to mechanical ventilation and lung injury

Abstract

Abstract Background Lung inhomogeneity plays a pivotal role in the development of ventilator-induced lung injury (VILI), particularly in the context of pre-existing lung injury. The mechanisms that underlie this interaction are poorly understood. We aimed to elucidate the regional transcriptomic response to mechanical ventilation (MV), with or without pre-existing lung injury, and link this to the regional lung volume response to MV. Methods Adult female BALB/c mice were randomly assigned into one of four groups: Saline, MV, lipopolysaccharide (LPS) or LPS/MV. Lung volumes (tidal volume, Vt; end-expiratory volume, EEV) were measured at baseline or after 2 h of ventilation using four-dimensional computed tomography (4DCT). Regional lung tissue samples corresponding to specific imaging regions were analysed for the transcriptome response by RNA-Seq. Bioinformatics analyses were conducted and the regional expression of dysregulated gene clusters was then correlated with the lung volume response. Results MV in the absence of pre-existing lung injury was associated with regional variations in tidal stretch. The addition of LPS also caused regional increases in EEV. We identified 345, 141 and 184 region-specific differentially expressed genes in response to MV, LPS and LPS/MV, respectively. Amongst these candidate genes, up-regulation of genes related to immune responses were positively correlated with increased regional tidal stretch in the MV group, while dysregulation of genes associated with endothelial barrier related pathways were associated with increased regional EEV and Vt when MV was combined with LPS. Further protein–protein interaction analysis led to the identification of two protein clusters representing the PI3K/Akt and MEK/ERK signalling hubs which may explain the interaction between MV and LPS exposure. Conclusion The biological pathways associated with lung volume inhomogeneity during MV, and MV in the presence of pre-existing inflammation, differed. MV related tidal stretch induced up-regulation of immune response genes, while LPS combined with MV disrupted PI3K/Akt and MEK/ERK signalling.

Keywords

• Mechanical ventilation
• Ventilator-induced lung injury
• Sepsis
• Lung inhomogeneities
• RNA sequencing
• Transcriptome