Unraveling the molecular details of the innate immune response

Abstract

There is increasing evidence of the important role of innate immune system in tumor surveillance (Vesely et al., 2011). The spontaneous immune response of T cells against tumors has a powerful prognostic value and can contribute to tumor control in patients. Indeed, T cell-based immunotherapies have shown to be effective in a variety of human cancers (Schumacher and Schreiber, 2015). A major player in this innate immune response is the stimulator of interferon genes (STING) protein, which is activated when a cell is infected by pathogens (Tanaka and Chen, 2012) or in the presence of tumor-derived cytosolic DNA (Woo et al., 2014). This pathway is critical for spontaneous T cell priming against tumors. Thus, understanding this mechanism at molecular level can be of paramount importance for improving prognosis and tumor-growth control in patients, as well as for developing new biomarkers and therapeutic vaccines. The activation of STING induces IFN-β production, and occurs after binding cyclic GMP-AMP (cGAMP), derived from pathogen or host-damaged cells. After that, the C-terminal tail (CTT) of STING is phosphorylated by TANK-binding kinase 1 (TBK1). Small-molecule ligands of STING are thus potential candidates for anti-cancer drugs or vaccine adjuvants. There are several available structures of cGAMP bound to STING, but none of them include the functionally relevant CTT segment, probably because of its flexibility. As a consequence, the structural changes that ligands such as cGAMP induce in STING CTT to make it phosphorylatable by TBK1 are not fully understood. The limitations of structure determination and the rather static picture that most of the currently available structural techniques provide are hampering the design of new STING ligands. In this context, computational methods such as molecular dynamics (MD) simulations can provide a detailed description of the thermodynamic properties and time-dependent phenomena of proteins (McCammon et al., 1977). Thanks to the exponential growth in computational power, MD simulations are currently applied to study dynamic events in proteins and other biomolecules at biologically relevant timescales (Shaw et al., 2010). This is changing the way we study the functional mechanisms of life systems, and the integration of structural data, molecular biology experiments and computational simulations will be essential for a full understanding of the complex biological and pathological processes that occur in cells. As 2013 Novel Laureate in Chemistry Michael Levitt said in 1998, “Computing has changed biology forever; most biologists just don't know it yet”. In EBioMedicine, Tsuchiya et al. (2016) identify the mechanism of STING activation at molecular level by using a combination of computational simulations and experimental approaches. They have studied the conformational effect induced by the ligand cGAMP on the flexible C-terminal tail (CTT) of the innate immune protein STING, by molecular dynamics (MD) simulations. The MD trajectories used in their study were long enough (1 μs) to observe the effects of cGAMP ligand binding on the flexibility of the CTT region. The findings from the molecular dynamics analyses were confirmed by molecular biology experiments, which allowed them to propose a new mechanism for STING activation. They found that in cGAMP-bound STING: i) the closed-Lid conformation was kept throughout the simulation, which was previously shown as essential for IFN-β activity; (Gao et al., 2014) ii) the end of CTT moved onto the Lid; and iii) the CTT region acquired a transient but defined β-sheet structure, which resembles the structure of the SER-rich region of IRF3 also phosphorylated by TBK1 (Takahasi et al., 2010). The importance of the residues forming the β-sheet in CTT for the activation of IFN-β was proved by site-directed mutagenesis experiments. The findings in Tsuchiya and colleagues' study open the door for future in silico screening studies in search of candidate molecules for therapeutic intervention. As they propose in their work, attention must be paid not only to the ligand-binding domain of STING but also to the effect of ligand binding in the flexible CTT region. The proposed model based on MD simulations can help to plan new virtual and experimental screening schemes in search of potent STING ligands that can promote IFN-β activity. This could lead to new anti-tumor therapeutic strategies through the enhancement of spontaneous immune response in patients. Tsuchiya and colleagues' study is a nice example of how the combination of computational simulations and molecular biology experiments can contribute to improve our understanding of relevant problems in biomedicine, such as the natural immune response against tumors.