Pulmonary ketogenesis promotes tolerance to bacterial infection

Tolerance is a major defense strategy against infection. During the host response to pathogens, tolerance restricts inflammatory damage to tissues, maintaining the long-term integrity of organs. The mechanisms that establish tolerance are poorly understood. We analyzed pulmonary isolates of Pseudomonas aeruginosa that evolved to coexist with tolerant hosts and found that these opportunistic pathogens facilitate tolerance by stimulating ketogenesis. Ketone body production in the airway limits the accumulation of detrimental factors that injure the lung, such as inflammatory cytokines and effector phagocytes. Although ketones are typically synthesized in the liver, in situ metabolo-transcriptomic studies revealed that P. aeruginosa drives their generation in the lung by co-opting the metabolism of alveolar macrophages, which are pulmonary cells of hepatic origin. This phenomenon was restricted to clinical isolates and not observed with laboratory strains of P. aeruginosa , confirming the unique metabolo-tolerogenic properties of ESKAPE pathogens adapted to the human lung. Overall design: WT or Irg1-/- C57bl/6 mice (8 week old) were treated with PBS or infected either with WT PAO1 or a mix of the 17 host-adapted P. aeruginosa clinical isolates (106 total CFU per mouse in 50uL). 16 hours after infection, mice were euthanized, their lungs were harvested, and single cell suspensions of the lungs were prepared as described. Briefly, the lungs were placed in an Eppendorf tube containing an enzymatic digestion solution of collagenase I (2 mg/mL), dispase (20 mg/mL), elastase (1 mg/mL), and DNAse (1 uL/mL) in PBS. The lungs were minced within the tube and then incubated with shaking at 37C for 30 minutes. 4 volumes of PBS supplemented with 10% FBS was added to quench the digestion, and the digestion solution was strained over a 70-micron filter. The cell suspension was spun down at 4C and 1400 rpm for 7 min. Red blood cell lysis was performed using the Invitrogen RBC lysis buffer. The resulting cell pellet was resuspended in PBS supplemented with 0.04% BSA before being loaded onto the 10X Genomics Chromium Single Cell Controller. Cell viability analysis was performed before loading the samples, and was above 95% per sample. A total of ~8000 cells were analyzed per sample. FASTQ file generation, alignment, filtering, barcode counting, and UMI counting were done using the 10X Cell Ranger software.