Introduction Systemic hypoxaemia and recurrent bacterial infections frequently co-exist in patients with acute and chronic lung disease and correlate with poor clinical outcomes. Inappropriate neutrophilic inflammation is regularly seen in these circumstances and the HIF/PHD pathway is implicated in the response of the innate immune system to both hypoxia and bacteria. Here we aimed to dissect and modify the interactions between hypoxia and innate host-pathogen response in the lung.

Methods C57BL/6 mice were either housed in room air or ‘preconditioned’ by being housed in 10% oxygen for 7 days. They then received intratracheal 1x10⁷ type 2 S. pneumoniae under recovery anaesthesia with subsequent exposure to hypoxia (10% O₂) or room air (21% O₂). Mice were assessed clinically, rectal temperatures recorded and culled for broncho-alveolar lavage (BAL) and tissue sampling (blood and lung) at various time points. Peripheral blood glucose was measured from tail vein venepuncture using a handheld blood glucose monitor. RNA from peripheral blood leucocytes was isolated and analysed using RNAseq. 18FDG-PET was performed on animals 14 h following infection to observe glucose utilisation. Histology was performed on formalin fixed sections for glycogen storage.

Results Exposure to acute hypoxia resulted in significant morbidity (sickness (5.7 vs 2.1, p < 0.02) and hypothermia (31.8 vs 36.0⁰C, p < 0.05)) and rapid 100% mortality by 48 h post infection. This response was independent of bacterial burden, and leukocyte recruitment. In keeping with a negative energy state, hypoxic mice displayed loss of liver glycogen, with increased serum ketone production and lower circulating glucose levels. Preconditioned mice showed marked protection from both the acute hypoxia-associated systemic phenotype and the negative energy state. Transfer of preconditioned bone marrow to naïve mice also rescued the pathophysiological response. RNAseq analysis of the circulating leukocyte population identified signal-induced suppression of HIF-1a pathway genes, which were linked to reduced leukocyte glucose utilisation in vivo by 18FDG-PET.

Conclusions Hypoxic preconditioning reverses the morbidity and mortality associated with acute hypoxia following intrapulmonary bacterial challenge. This response is dependent on the preconditioning of the innate immune system by suppressing HIF1 alpha and altering circulating leukocyte metabolism.

Support RSD is funded by the MRC. SRW is funded by the Wellcome Trust.