An in vitro methodology demonstrates the five steps of trained immunity in mice: implications on biomarker discovery, adaptive immune responses, mouse strains, sample cryopreservation and genetic ablation [ATAC-Seq]

Trained immunity is defined by increased inflammatory cytokine production by macrophages that are sequentially activated by two stimuli, separated by a resting period. The models used to study trained immunity should reflect the definition of this process, which includes five different steps. Here, we developed an in vitro methodology to study trained immunity using murine bone marrow-derived macrophages (BMDM), following sequential stimulation with β-glucan and lipopolysaccharide (LPS). Kinetic analysis of cytokine production demonstrated, for the first time, that trained BMDM secrete high levels of tumor necrosis factor (TNF) and interleukin 6 (IL6) following stimulation with β-glucan (stimulus 1), compared to unstimulated BMDM (step 1). After a resting period, trained BMDM do not secrete significant levels of these cytokines (step 2), but rapidly produce enhanced levels of TNF and IL6 after secondary stimulation with LPS (stimulus 2), compared to BMDM stimulated with either β-glucan (step 3) or LPS (step 4) alone. In addition, the sum of cytokine levels produced by both BMDM stimulated with either β-glucan (stimulus 1) or LPS (stimulus 2) individually, was significantly lower than the cytokine levels produced by trained BMDM stimulated with both β-glucan and LPS (stimulus 1+2) (step 5). Our findings also demonstrate that trained BMDM produce higher levels of serum amyloid A3 (SAA3) protein due to metabolic and epigenetic reprogramming. Functionally, trained BMDM upregulate the expression of stimulatory molecules such as MHC-II, CD80, CD86, OX40L and PDL1; downregulate the inhibitory molecules ICOSL and CD206; and induce CD4+ and CD8+ T cell proliferation in vitro through a cell contact-dependent mechanism. Our results also revealed strain-specific differences in the training response among mice, the negative impact of cell cryopreservation on trained immunity, and the potential use of genetically modified mice in elucidating mechanistic pathways through which different training stimuli may induce long-lived memory macrophages. These results also highlight the applicability of mouse in vitro methods as an additional tool to study trained immunity, as the use of inbred mice reduce the genetic variability present in humans and minimize variables that affect trained immunity such as infections, vaccinations and lifestyle-related stimuli. The standardization of trained immunity protocols using murine BMDM will facilitate the development of therapeutic strategies that target the innate immune memory response for the treatment of multiple immune-based diseases.