Multidrug resistance proteins (MRP) transport the cytokine/nuclear protein High Mobility Group Box 1 (HMGB1) across membranes

High mobility group box 1 (HMGB1) is a mobile chromatin protein that can relocate into the external environment and act as a cytokine. HMGB1 cellular release can be passive, by cell necrosis, or active, by living cells after a specific stimulus is received. HMGB1 does not possess a leader sequence, a sequence usually present at the Nterminus of secreted proteins that is sufficient to address them through the ER/Golgi exocytic pathway. This work focused on the molecular mechanism by which HMGB1, as a leaderless protein, can be actively exocytosed by living cells. HMGB1 undergoes post-translational modifications that allow it to move to various cellular SUb-compartments: 1) Translocation from the nucleus to the cytosol 2) Loading inside exocytic vesicles (secretory Iysosomes) 3) Release into the extracellular milieu by vesicle degranulation. The post-translational modification involved in the first translocation step was already known to be lysine acetylation. Here I demonstrate the necessity of cysteine glutathionylation for the translocation inside secretory Iysosomes, and the direct involvement of Multidrug Resistance Proteins (MRP/ABCC) in the transport process. I started with the observation that HMGB1 and IL-1 beta, another leaderless cytokine, do colocalize inside secretory Iysosomes in macrophages, and that a pharmacological inhibitor of IL-1beta exocytosis, glybenclamide, inhibits HMGB1 release too. Glybenclamide is a general inhibitor of the ABC transporter Superfamily, a large family of transmembrane proteins that transport ions, drugs and small peptides. Based on the sensitivity or insensitivity to different drugs, the HMGB1 transporter was likely to belong to ABC C group. Overexpression of MRP1/ABCC1, MRP2/ABCC2 and MRP3/ABCC3 endows activated cells with the ability to secrete HMGB 1. Moreover, macrophages from Mrp1-1- mice have impaired HMGB1 exocytosis. I also suggest that HMGB1 is unfolded during its translocation. MRP transporters are known to recognize glutathionylated substrates. Indeed, HMGB1 can be glutathionylated, and depletion of the glutathione intracellular pool impairs HMGB1 secretion. MRP transporters can be overexpressed in drug resistant tumour cells, where they pump chemotherapeutic agents out of the cell. I suggest that cells that are drug-resistant because of MRP1 overexpression might also secrete HMGB1