Apoptotic cells actively release metabolite messanger with anti-inflammatory properties

Apoptosis accounts for ~ 90% of homeostatic cell turnover in the body1. While caspase-dependent apoptosis is known to influence immune tolerance, induction of cell proliferation, and tissue regeneration2-4, how these apoptotic cells contribute to such diverse effects is less understood. Soluble factors released from apoptotic cells could be a means of 'talking' to healthy cells in the tissue neighborhood. While a few phagocyte-attracting 'find-me' signals released from apoptotic cells are known5, the larger 'metabolite secretome' from apoptotic cells is not yet defined. Here, via unbiased profiling of the apoptotic cell secretome, we identified 123 metabolites that are specifically released during apoptosis, while cells still retain membrane integrity. Release of ~25 of these metabolites is dependent on Pannexin 1 (Panx1) channels, which are opened by caspase-dependent cleavage during apoptosis. Interestingly, one of the Panx1-dependent metabolites from apoptotic cells is spermidine, a polyamine whose release could be, in part, due to the continued metabolic activity of dying cells. RNAseq analysis of healthy cells exposed to the apoptotic secretome revealed Panx1-dependent gene programs, including genes linked to suppression of inflammation, cell proliferation, and wound healing. Analysis of the apoptotic secretome across cell types and different apoptotic stimuli, identified a core “set” of 8 Panx1-dependent metabolites. Strikingly, a selected cocktail of these metabolites demonstrated significant immunosuppressive effects by potently reducing disease severity in an inflammatory arthritis model and a lung graft rejection model. Collectively, these data advance the concept that apoptotic cells are not 'inert' corpses waiting for removal by phagocytes, but rather they actively communicate with the tissue environment through their metabolite secretome, with implications in tissue inflammation. Overall design: The experiment consist of three conditions: Phagocytes (LR73 cells) treated for four hours with supernatants taken from either live Jurkats , apoptotic Jurkats, or apoptotic Jurkats expressing a dominant-negative form of the Panx1 channel (caspase site mutation that does not allow cleavage by executioner caspase). All conditions consisted of four biological replicates.