The Mechanical Problemthree Different Solutions

The seven transmembrane helices (TM1-7) represent the conserved hydrophobic core of G protein-coupled receptors (GPCRs). This is a versatile scaffold, which allows for the many modes of ligand-dependent activation. There are many variations on the theme [5] , because the ligand may be prebound (as in rhodopsin), may bind primarily to an extracellular domain (e.g., the fly trap domain of metabotropic glutamate receptors), to the extracellular face of the hydrophobic core (as in peptide receptors), or within the hydrophobic core (as in receptors for biogenic amines, e.g., adrenergic receptors). Regardless of the details, the fundamental problem is to relay the agonist-i nduced confor-mational change via the helical arrangement to the intracellular side, which— from the viewpoint of the G protein—is the business side of the receptor.

GDP is deeply buried within the G protein a subunit. Heterotrimeric G protein a subunits are larger in size than the small monomeric RAS -like G proteins because they have an "extra" helical domain. In addition, the G protein Py dimer covers a large surface of the G protein a subunit and physically blocks the exit of GDP. This also explains why Py dimers increase the affinity for GDP of those G protein a subunits that have measurable spontaneous GPD release rates [6, 7]. Finally, as can be seen from Fig. 4.1a, GPCRs cannot contact the GDP binding pocket directly: the nucleotide binding pocket embedded in the GTPase domain is about 30 A away from the interface between receptor and G protein. This distance precludes a direct contact site between intracellular loops of the receptor and GDP binding pocket. This is also true for squid rhodopsin [8], which is shown in Fig. 4.1a. This most recent addition to the available structures of GPCRs (mammalian rhodopsin in several conformations and the human p2-adrenergic receptor) is remarkable for several reasons, not the least of which is the extended protrusion of transmembrane helices 2 and 3 (TM2 and TM3) into the cytoplasm. But it is evident from Fig. 4.1a that—-n spite of this long intracellular loop—there is not any conceivable way how to arrange the G protein heterotrimer and the receptor to allow for a direct contact between receptor and GDP binding pocket. Several GPCRs have long C-termini; in an extended conformation, these may readily reach into the vicinity of the GDP binding pocket. However, it can be convincingly argued that the long C-termini are irrelevant to the basic mechanism of G protein activation because GPCRs with short C-termini also efficiently activate G proteins. In addition, in those instances, where this issue has been examined, truncation of a long terminus does not affect G protein activation (see, e.g., Reference 9) provided that it does not affect the structure of helix 8 (the proximal portion of the C-terminus adjacent to the seventh transmembrane helix (TM7) [10] .

Therefore, it is generally accepted that receptors act "at a distance" : they must somehow relay the conformational signals arising within the hydropho-bic core via their G protein interaction surface to the guanine nucleotide binding pocket to trigger GDP release. Three solutions have been proposed:

Figure 4.1 Mechanism of G protein activation. (a) Interface of a G protein-coupled receptor, such as squid rhodopsin (shown as grey Ribbon model, Protein Data Bank (PDB) ID 2ZIY; Reference 8) and a G protein heterotrimer, such as transducin (PDB ID 1GOT; Reference 13). The transducin heterotrimer is composed of the nucleotide-binding Gt a-subunit (shown as green, transparent space fill/ribbon model) and the Gt P (blue) and y (red) subunits. Despite the long C-terminal cytoplasmic domain of squid rhodopsin, the nucleotide GDP (shown as sticks representation) is still too far away to be in direct contact with the receptor. This raises the question of how receptors can activate G proteins to release GDP. Three models have been proposed, which provide different solutions to the mechanical problem (b-d): (b) The "lever-arm" model: (1) the receptor uses N-terminal Na helix of Ga (red) as a lever to pull GPy away from Ga (indicated by arrows). (2) This motion invokes movements in parts of the helical region of Ga (aA and aB, red) and the switch regions 1 and 2 (red) that causes the release of GDP. The right panels show the zoomed region near GDP. (c) "Gear-shift" model: (1) the receptor pushes Na toward the GPy subunits. (2) This motions brings GP Asp186 (shown as stick representation; motion represented by the arrow in the blowup) in close proximity to GDP, which (3) pushes the coiled-coil helices NP and Ny (red) toward the helical domain of Ga to invoke GDP release. (d) "C-terminal latch" model: the interaction of the receptor with the C-terminus of Ga causes a rigid movement of the adjacent (C-terminal) helix a5 (Ca; red, indicated by an arrow), which is translated to the TCAT motif (stick representation) and triggers the release of GDP.

two of these models (termed "lever-arm" and "gear-shift" models) assume that GDP release is contingent on an active involvement of GPy Both models postulate a conceptual analogy between GPy and GEFs of small G proteins and ribosomal elongation effectors (EF-Tu): GEFs engage a large surface of the small G protein (or of EF-Tu) and are in close vicinity to residues stabilizing GDP binding; the same is true for GPy In the third model referred to as "C-terminal latch" model, GPy provides a receptor docking site, but is not actively involved in the release process per se.

0 0

Post a comment