Dual cytokine inducing properties of HMGB1

Abstract

High mobility group box protein 1 (HMGB1) is a nuclear protein that can be released either passively by necrotic cells or actively by stimulated cells. Extracellular HMGB1 is a potent inducer of inflammation and the importance of HMGB1 as a mediator in a number of inflammatory diseases including rheumatoid arthritis, sepsis and ischemia‐reperfusion injury, has been demonstrated by successfully targeting the protein in preclinical models. The aim of my thesis was to characterise the cytokine‐inducing properties of HMGB1 and to study which receptors were required for cytokine induction.

Several studies have indicated that HMGB1 can co‐operate with other pro‐ inflammatory molecules to induce inflammation. To further study this mechanism we formed complexes of HMGB1 together with the exogenous TLR ligands LPS, Pam3CSK4 and CpG‐ODN or the endogenous ligands IL‐1α and IL‐1β. Stimulation of macrophages or synovial fibroblasts with these different HMGB1 complexes resulted in significantly enhanced cytokine production as compared to stimulation with each ligand alone (papers I and II). Importantly, HMGB1 selectively enhanced the stimulatory activity of certain molecules as it did not display this activity with all tested ligands.

In papers II and III the receptor requirements of HMGB1 complexes were studied. HMGB1 in complex with LPS, Pam3CSK4 and IL‐ 1α/β stimulated cytokine release via the TLR4, TLR2 and IL‐1RI receptors, respectively, demonstrating that cytokine induction by HMGB1 complexes is dependent on the receptor for the respective partner molecule.

In paper IV we demonstrated that cytokine release stimulated by uncomplexed HMGB1 was dependent on TLR4 but not on RAGE or TLR2, and that a direct association of HMGB1 and TLR4 was detected both in vitro and in vivo. Using site‐ directed mutagenesis we furthermore determined that the cysteine in position 106 of HMGB1 was required for both binding to TLR4 and for cytokine induction.

In summary, in this thesis I have demonstrated that HMGB1 has the ability to induce cytokine production in two ways: through forming complexes with certain danger molecules and thereby increasing their stimulatory activities, and through direct interaction with the TLR4 receptor. The TLR4‐mediated endogenous cytokine‐inducing capacity of HMGB1 requires a cysteine in position 106, while the enhancing capacity of HMGB1‐partner ligand complexes is independent of HMGB1 ligation to TLR4 but dependent on signalling via the partner molecule receptor. Neither mechanism involved an interaction of HMGB1 with its suggested receptor RAGE. These results are of value for designing HMGB1‐targeting therapies that focus on blocking only certain HMGB1 functions or certain receptor interactions.