Computational design of highly signaling active membrane receptors through de novo solvent-mediated allosteric networks

Protein catalysis and allostery require the atomic-level orchestration and motion of residues, ligand, solvent and protein effector molecules, but the ability to design protein activity through precise protein-solvent cooperative interactions has not been demonstrated. Here, we report the design of a dozen novel membrane receptors catalyzing G-protein nucleotide exchange through diverse de novo engineered allosteric pathways mediated by cooperative networks of intra-protein, protein-ligand and solvent molecule interactions. Consistent with the predictions, designed protein activities strongly correlated with the level of solvent-mediated interaction network plasticity at flexible transmembrane helical interfaces. Several designs displayed considerably enhanced thermostability and activity compared to related natural receptors. The most stable and active variant crystallized in an unforeseen signaling active conformation, in excellent agreement with the design models. The allosteric network topologies of the best designs bear limited similarity to those of natural receptors and reveal a space of allosteric interactions larger than previously inferred from natural proteins. The approach should prove useful for engineering proteins with novel complex protein catalytic and signaling activities. Competing Interest Statement The authors have declared no competing interest.

A Biochemical and Biophysical Investigation of 5-HT3 Receptor Stability

Menno Berend Tol

The 5-hydroxytryptamine 3 receptor (5-HT3R) is a member of the pentameric ligand-gated ion channel (pLGIC) family, that plays an important role in fast signal transduction and cell-cell communication in synapses: They convert a chemical signal from a neurotransmitter into the electrical signal that represents the action potential. Activation of 5-HT3R with serotonin triggers a range of molecular reorganizations that lead to channel opening and subsequent membrane depolarization. Because of their role, pLGICs are involved in a wide range of diseases and disorders and are therefore important drug targets. A typical pLGIC is composed of a β -sheet rich extracellular domain (ECD) where ligand binding takes place, a transmembrane domain that is composed of four α-helices (TM1-TM4). In vertebrates pLGICs often contain a third intracellular domain (ICD), located between TM3 and TM4. The 5-HT3R is active as a homopentameric receptor of five identical A subunits, or as a heteropentamer containing one or more of several subunits (B to E) together with an A subunit. The 5-HT3 receptor is naturally located in the cell membrane in low amounts. For in vitro experiments such as biophysical investigation, the receptor has to be overproduced and purified in a detergent-solubilized state. Therefore, a robust and scalable expression system for homo- and heteropentameric 5-HT3 receptors was developed, building upon a tetracycline inducible cell line and Strep-tag affinity purification technology. Production of solubilized and purified 5-HT3R is well suited for use in the structural and functional characterization of the receptor. The stability of the 5-HT3 receptor is an important parameter during structural and functional experimentation. Using thermal unfolding the stability of 5-HT3R was investigated in plasma membranes as well as during detergent-extraction, purification and reconstitution into artificial lipid bilayers. A large loss in thermostability was found that correlates with the loss of the lipid bilayer during membrane solubilization and purification. Thermal unfolding of 5-HT3R occurred in consecutive steps at distinct protein locations: First a loss of ligand binding is detected, followed by formation of different transient low oligomeric states of receptor pentamers, followed by partial unfolding of helical parts of the protein, which finally leads to the formation large receptor aggregates. It was also found that structural destabilization of the receptor in detergents could be partially reversed by reconstituting the receptor into lipid bilayers. Thermal unfolding leads to the irreversible formation of aggregates which is known to prevent effective refolding. Therefore, an further investigation of 5-HT3R was made by unfolding the receptor in urea and sodium dodecyl sulfate (SDS). The results indicate iii iv ABSTRACT that, unlike thermal unfolding, unfolding in urea and SDS does not result in the formation of aggregates. Unfolding by urea shows a 50% loss of helical content while keeping the characteristic pentamer intact. Contrastingly, unfolding by SDS shows dissociation of the pentamer, but no noticeable change in secondary structure. Still, refolding from urea and SDS proved to be impossible. Two weak fluorophores, crystal violet (CrV) and ethidium bromide (EtBr), bind to the homologous nicotinic acetylcholine receptor (nAChR). Because the fluorescence of these compounds is also sensitive towards their environment they can be used as a molecular beacon for the functional state of a receptor. When applied to 5-HT3R, it was found that CrV and EtBr bind with micromolar affinity to the desensitized state of the receptor – similar to binding reported for the nAChR. It appeared that binding of CrV or EtBr to the 5-HT3R induces a certain functional state (resting, open or desensitized). Because no detectable change in fluorescence occurs, this prevents their use as molecular beacons. Still, non-competitive binding to the 5-HT3R was little reported on before and could be an important target in pharmacological research. The ICD of the 5-HT3R is predicted to contain a stretch of ∼60 disordered or flexible residues which might be involved in subunit assembly and membrane trafficking. These unstructured regions of a 5-HT3R were probed using limited proteolysis. Thereby two distinct fragments were formed that remain tightly associated with each other, while a small stretch of 35 residues between TM3 and TM4 was removed. Furthermore, it was shown that the thermal stability of the helical core is unchanged compared to untreated 5-HT3R, but that the extracellular ligand-binding domain is very sensitive to cleavage in the ICD. By probing the 5-HT3R with limited proteolysis a rigid core was identified, indicating the important role of the transmembrane domain for holding together the pLGIC scaffold. The results presented here contribute a significant body of knowledge to membrane proteins in general and to the 5-HT3 receptor in particular, upon which further research can be based leading to advances in the fields of biochemistry, biophysics and pharmacology.

EPFL2012

The Principles of Ligand Specificity on beta-2-adrenergic receptor

Shuguang Yuan

G protein-coupled receptors are recognized as one of the largest families of membrane proteins. Despite sharing a characteristic seven-transmembrane topology, G protein-coupled receptors regulate a wide range of cellular signaling pathways in response to various physical and chemical stimuli, and prevail as an important target for drug discovery. Notably, the recent progress in crystallographic methods led to a breakthrough in elucidating the structures of membrane proteins. The structures of beta(2)-adrenergic receptor bound with a variety of ligands provide atomic details of the binding modes of agonists, antagonists and inverse agonists. In this study, we selected four representative molecules from each functional class of ligands and investigated their impacts on beta(2)-adrenergic receptor through a total of 12 x 100 ns molecular dynamics simulations. From the obtained trajectories, we generated molecular fingerprints exemplifying propensities of protein-ligand interactions. For each functional class of compounds, we characterized and compared the fluctuation of the protein backbone, the volumes in the intracellular pockets, the water densities in the receptors, the domain interaction networks as well as the movements of transmembrane helices. We discovered that each class of ligands exhibits a distinct mode of interactions with mainly TM5 and TM6, altering the shape and eventually the state of the receptor. Our findings provide insightful prospective into GPCR targeted structure-based drug discoveries.

Nature Publishing Group2016

Reprogramming G Protein-Coupled Receptor Structure, Function And Signaling By Computational Design

Dániel Kéri

G protein-coupled receptors (GPCRs) are 7-transmembrane alpha-helical integral membrane proteins on which cells heavily rely to receive information regarding their external environment. These receptors are able to transfer information to intracellular downstream effectors upon stimulation from their cognate ligands, which can be considerably diverse. They can include light, small molecules, peptides, protons; Due to its evolutionary success, over millions of years of evolution, this protein family has expanded to include over 800 members. Despite the large size of the family and the diverse inputs they accept, the structure and the mechanism for relaying signals to their downstream partners, G proteins and B-arrestins, remains largely the same. As these receptors are involved in a great deal of biological processes such as sight (rhodopsin), cognition (dopamine, serotonin, opioid receptors), immunity (chemokine receptors), heart function (adrenergic, angiotensin receptors), etc.; a great deal of effort has gone into understanding their structure and function. Our efforts have gone into understanding their signaling properties and rationally modifying them. Most proteins are fairly flexible entities and their structures tend to oscillate and move around, sampling a number of conformations. Nature has taken advantage of these natural motions to regulate protein function from distant binding sites, in a phenomenon called allostery. GPCRs are unique in the sense that their primary function relies on allostery. By tweaking the allosteric signaling of GPCRs, we hope to have a better understand allostery, use the principles we learn to create designer GPCR biosensors with the desired sensitivity, and to eventually create de novo allosteric proteins. In this work, I present our progress in this effort, which has been significant. We have created a highly accurate in silico method to allosterically alter relative stabilities of the active and inactive states of the GPCR dopamine D2 receptor, but also to independently control its allosteric signaling strength. With our methodology we have been able to create receptors which have high basal activities by selectively stabilizing the active state of the receptor. This has already facilitated the structure determination of the active, G protein-bound dopamine D2 receptor; a first for this GPCR. Additionally, we have created a number of dopamine D2 variants with roughly 100 times higher sensitivity to dopamine than the WT receptor. These efforts continue to characterize the functional effects of these allosteric mutations. Furthermore, we are also working on a collaboration to rewire the signaling pathway of the adenosine A2A receptor. Adenosine is a common immunosuppresant in solid tumor micro-environments (TMEs). By rewiring the downstream signaling of the A2A receptor, we hope to reverse the response of T-cells in these TMEs to activate and attack the solid tumor.