Emblem  Laboratory of Biomolecular Modeling  

home / work / projects / Modeling of melatonin receptors

Molecular modeling of melatonin receptors

Melatonin receptors belong to the superfamily of G-protein coupled receptors (GPCRs), which represent a very important class of biological macromolecules. These objects are the integral membrane proteins (MPs) accommodating the wide range of influences: from light and metal cations to peptides and proteins. GPCRs mediate a great amount of processes in living organisms. Many diseases are caused by disfunctions of the systems that transduce signals via GPCRs. That’s why studying of spatial structure and, therefore, molecular mechanisms of functioning of these receptors is an important task in modern biology and medicine. To date, more than 50% of drugs are targeted on GPCRs, and this number grows rapidly.

Fig. 1 Schematic representation of G-protein coupled receptor folding.


However, experimental methods (e.g., X-ray crystallography and NMR spectroscopy) often fail to determine three-dimensional (3D) structure of GPCRs. The only exception is the bovine visual rhodopsin: its 3D X-ray structure has been determined with atomic resolution. Fig. 1 represents the TM folding of rhodopsin. It is commonly accepted now that all GPCRs share similar fold of TM domain. This reveals an opportunity to build their 3D models by means of molecular modeling.

Fig. 2 Melatonin (N-acetyl-5-methoxytryptamin) structure.


One of the research projects in our Laboratory is targeting on building of 3D models of the human melatonin receptors MT1 and MT2. This work is being performed in collaboration with the Laboratory of Therapeutic Chemistry at the University of Lille, France. Melatonin (a derivative of tryptophan) is an important biological agent, critical for regulation of circadian rhythms; it possesses strong immunomodulator, antioxidant, and some other activities (Fig. 2).

Main objectives of the project:

  • Homology-based building of molecular models of melatonin receptors using visual rhodopsin as a structural template; refinement and optimization of the obtained structures taking into account available experimental data;
  • Detailed characterization of the receptors’ active site, rationalization of experimental data on selectivity of binding for a number of ligands;
  • In silico calculation of free energy of binding of some ligands, comparison of these results with the experimental findings;
  • In perspective, goal-oriented computer-aided design of new high-affinitive and selective ligands, their testing in vitro.

Figure 3 shows homology models of MT1 and MT2 receptors with template (visual rhodopsin) structure superimposed, as well as structures of active sites binding melatonin molecule. These models were optimized using several criteria:

  1. Optimal melatonin conformation in the binding site (minimal value of the „penalty“, imposed on the geometry of three h-bonds that bind melatonin to the active site);
  2. Conformity of hydrophobic and variable organization of TM domains of receptors models and those of rhodopsin;
  3. Protein „packing quality“ in the TM domain, according to the Eisenberg’s 3D-1D score and
  4. Melatonin docking results to receptors’ models.

Fig. 3 Melatonin receptors’ structures. Upper panel: Structures of MT1 (blue) and MT2 (green) melatonin receptors models and rhodopsin structure (red). Only backbones are shown, except active site residues (His5.46, Ser3.35 and Ser3.39), for which side chains are plotted. Lower panel: Structure of MT1 (left) and MT2 (right) active sites in the complex with melatonin molecule. Backbones of proximal to the melatonin binding site TM-helices are drawn as tubes; side chains are shown for active site residues.

Optimization consisted of rotation of TM3 α-helix around its axis as a rigid body, energy minimization of receptor model complex with melatonin molecule and simultaneous assessment of model correctness according to mentioned criteria.

„Optimized“ structures of receptor-melatonin complexes were used to delineate differences between MT1 and MT2 binding sites that draw either receptor subtype selective to some melatonin analogs. Based on the complementarity of molecular hydrophobic potential (MHP) concept, a hypothesis for selectivity of indole-type melatonin analogs was proposed. Figure 4 demonstrates application of the MHP-complementarity approach for explanation of selectivity of two MT receptors’ ligands: towards MT1 (fig. 4b) or MT2 (fig. 4a) subtypes.

Fig. 4 Hydrophobic/hydrophilic accommodation of two melatonin analogs in the binding site of the receptors MT1 and MT2. (A) 2-benzylmelatonin (selective to MT2 receptor); (B) benzoxazole analog (selective to MT1 receptor). Upper panels of the figure show skeletal model (1) and surface hydrophobicity (2) of isolated compounds. The surface is colored according to MHP values created by the ligand’s atoms (MHPligand). Lower panels show hydrophobicity of protein environment for the ligands in complex with MT1 (3) and MT2 (4). Molecular surface of the ligands is mapped according to MHP created by surrounding protein atoms (MHPsite).

At present, a new investigation is being held in our group: a novel method to assess the „packing quality“ of TM domains of MPs. This approach, based on the analysis of high-resolution spatial structures of MPs, will hopefully be useful for optimization of theoretical models of MPs structures, in the first hand — GPCR ones.

(Authors: Chugunov Anton, Efremov Roman.)

Address: 117997 Russia, Moscow, ul. Miklukho-Maklaya 16/10.
Tel.: +7 (495) 336-20-00.
Email: efremov@nmr.ru

© 2003–2007
batch2k.