Rhodopsin: structural basis of molecular physiology.

The crystal structure of rod cell visual pigment rhodopsin was recently solved at 2.8-A resolution. A critical evaluation of a decade of structure-function studies is now possible. It is also possible to begin to explain the structural basis for several unique physiological properties of the vertebrate visual system, including extremely low dark noise levels as well as high gain and color detection. The ligand-binding pocket of rhodopsin is remarkably compact, and several apparent chromophore-protein interactions were not predicted from extensive mutagenesis or spectroscopic studies. The transmembrane helices are interrupted or kinked at multiple sites. An extensive network of interhelical interactions stabilizes the ground state of the receptor. The helix movement model of receptor activation, which might apply to all G protein-coupled receptors (GPCRs) of the rhodopsin family, is supported by several structural elements that suggest how light-induced conformational changes in the ligand-binding pocket are transmitted to the cytoplasmic surface. The cytoplasmic domain of the receptor is remarkable for a carboxy-terminal helical domain extending from the seventh transmembrane segment parallel to the bilayer surface. Thus the cytoplasmic surface appears to be approximately the right size to bind to the transducin heterotrimer in a one-to-one complex. Future high-resolution structural studies of rhodopsin and other GPCRs will form a basis to elucidate the detailed molecular mechanism of GPCR-mediated signal transduction.

Glucagon receptor activates extracellular signal-regulated protein kinase 1/2 via cAMP-dependent protein kinase.

We prepared a stable cell line expressing the glucagon receptor to characterize the effect of G(s)-coupled receptor stimulation on extracellular signal-regulated protein kinase 1/2 (ERK1/2) activity. Glucagon treatment of the cell line caused a dose-dependent increase in cAMP concentration, activation of cAMP-dependent protein kinase (PKA), and transient release of intracellular calcium. Glucagon treatment also caused rapid dose-dependent phosphorylation and activation of mitogen-activated protein kinase kinase/ERK kinase (MEK1/2) and ERK1/2. Inhibition of either PKA or MEK1/2 blocked ERK1/2 activation by glucagon. However, no significant activation of several upstream activators of MEK, including Ras, Rap1, and Raf, was observed in response to glucagon treatment. In addition, chelation of intracellular calcium reduced glucagon-mediated ERK1/2 activation. In transient transfection experiments, glucagon receptor mutants that bound glucagon but failed to increase intracellular cAMP and calcium concentrations showed no glucagon-stimulated ERK1/2 phosphorylation. We conclude that glucagon-induced MEK1/2 and ERK1/2 activation is mediated by PKA and that an increase in intracellular calcium concentration is required for maximal ERK activation.

Rapid activation of transducin by mutations distant from the nucleotide-binding site: evidence for a mechanistic model of receptor-catalyzed nucleotide exchange by G proteins.

G proteins act as molecular switches in which information flow depends on whether the bound nucleotide is GDP ("off") or GTP ("on"). We studied the basal and receptor-catalyzed nucleotide exchange rates of site-directed mutants of the alpha subunit of transducin. We identified three amino acid residues (Thr-325, Val-328, and Phe-332) in which mutation resulted in dramatic increases (up to 165-fold) in basal nucleotide exchange rates in addition to enhanced receptor-catalyzed nucleotide exchange rates. These three residues are located on the inward facing surface of the alpha5 helix, which lies between the carboxyl-terminal tail and a loop contacting the nucleotide-binding pocket. Mutation of amino acid residues on the outward facing surface of the same alpha5 helix caused a decrease in receptor-catalyzed nucleotide exchange. We propose that the alpha5 helix comprises a functional microdomain in G proteins that affects basal nucleotide release rates and mediates receptor-catalyzed nucleotide exchange at a distance from the nucleotide-binding pocket.

The function of interdomain interactions in controlling nucleotide exchange rates in transducin.

The intramolecular contacts in heterotrimeric G proteins that determine the rates of basal and receptor-stimulated nucleotide exchange are not fully understood. The alpha subunit of heterotrimeric G proteins consists of two domains: a Ras-like domain with structural homology to the monomeric G protein Ras and a helical domain comprised of six alpha-helices. The bound nucleotide lies in a deep cleft between the two domains. Exchange of the bound nucleotide may involve opening of this cleft. Thus interactions between the domains may affect the rate of nucleotide exchange in G proteins. We have tested this hypothesis in the alpha subunit of the rod cell G protein transducin (Galpha(t)). Site-directed mutations were prepared in a series of residues located at the interdomain interface. The proteins were expressed in vitro in a reticulocyte lysate system. The rates of basal and rhodopsin-catalyzed nucleotide exchange were determined using a trypsin digestion assay specifically adapted for kinetic measurements. Charge-altering substitutions of two residues at the interdomain interface, Lys(273) and Lys(276), increased basal nucleotide exchange rates modestly (5-10-fold). However, we found no evidence that interactions spanning the two domains in Galpha(t) significantly affected either basal or rhodopsin-catalyzed nucleotide exchange rates. These results suggest that opening of the interdomain cleft is not an energetic barrier to nucleotide exchange in Galpha(t). Experiments with Galpha(i1) suggest by comparison that the organization and function of the interdomain region differ among various G protein subtypes.

Rhodopsin activation affects the environment of specific neighboring phospholipids: an FTIR spectroscopic study.

Rhodopsin is a member of a superfamily of G-protein-coupled receptors that transduce signals across membranes. We used Fourier-transform infrared (FTIR) difference spectroscopy to study the interaction between rhodopsin and lipid bilayer upon receptor activation. A difference band at 1744 cm(-1) (+)/1727 cm(-1) (-) was identified in the FTIR-difference spectrum of rhodopsin mutant D83N/E122Q in which spectral difference bands arising from the carbonyl stretching frequencies of protonated carboxylic acid groups were removed by mutation. As the band was abolished by detergent delipidation, we suggested that it arose from carbonyl groups of phospholipid fatty acid esters. Rhodopsin and the D83N/E122Q mutant were reconstituted into various (13)C-labeled 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine vesicles and probed. The 1744-cm(-1) (+)/1727 cm(-1) (-) band could be unequivocally assigned to a change in the lipid ester carbonyl stretch upon receptor activation, with roughly equal contribution from both lipid esters. The band intensity scaled with the amount of rhodopsin but not with the amount of lipid, excluding the possibility that it was due to the bulk lipid phase. We also excluded the possibility that the lipid band represents a change in the number of boundary lipids or a general alteration in the boundary lipid environment upon formation of metarhodopsin II. Instead, the data suggest that the lipid band represents the change of a specific lipid-receptor interaction that is coupled to protein conformational changes.

Reconstitution of the vertebrate visual cascade using recombinant heterotrimeric transducin purified from Sf9 cells.

For reconstitution studies with rhodopsin and cGMP phosphodiesterase (PDE), all three subunits of heterotrimeric transducin (T alpha beta gamma) were simultaneously expressed in Sf9 cells at high levels using a baculovirus expression system and purified to homogeneity. Light-activated rhodopsin catalyzed the loading of purified recombinant T alpha with GTP gamma S. In vitro reconstitution of rhodopsin, recombinant transducin, and PDE in detergent solution resulted in cGMP hydrolysis upon illumination, demonstrating that recombinant transducin was able to activate PDE. The rate of cGMP hydrolysis by PDE as a function of GTP gamma S-loaded recombinant transducin (T(*)) concentration gave a Hill coefficient of approximately 2, suggesting that the activation of PDE by T(*) was cooperatively regulated. Furthermore, the kinetic rate constants for the activation of PDE by T(*) suggested that only the complex of PDE with two T(*) molecules, PDE. T(2)(*), was significantly catalytically active under the conditions of the assay. We conclude that the model of essential coactivation best describes the activation of PDE by T(*) in a reconstituted vertebrate visual cascade using recombinant heterotrimeric transducin.

Transducin-dependent protonation of glutamic acid 134 in rhodopsin.

A highly conserved carboxylic acid residue in rhodopsin, Glu(134), modulates transducin (G(t)) interaction. It has been postulated that Glu(134) becomes protonated upon receptor activation. We studied the interaction between rhodopsin and G(t) using Fourier transform infrared (FTIR) difference spectroscopy combined with attenuated total reflection (ATR). Formation of the complex between G(t) and photoactivated rhodopsin reconstituted into phosphatidylcholine vesicles caused prominent infrared absorption increases at 1641, 1550, and 1517 cm(-)(1). The rhodopsin mutant E134Q was also studied. When measured in the presence of G(t), replacement of Glu(134) by glutamine abolished the low-frequency part of a broad absorption band at 1735 cm(-)(1) that is normally superimposed on the light-induced absorption changes of Asp(83) and Glu(122) of rhodopsin. In addition, a negative absorption band at 1400 cm(-)(1) that is evoked by interaction of native metarhodopsin II (MII) with G(t) was not observed in the difference spectrum of the E134Q mutant. Thus, Glu(134) is ionized in the dark and exhibits a symmetrical COO(-) stretching vibration at 1400 cm(-)(1). Glu(134) becomes protonated in the G(t)-MII complex and displays a C=O stretching mode near 1730 cm(-)(1). The E134Q mutation also affects absorption changes attributable to lipids, suggesting that the protonation of Glu(134) is linked to transfer of the carboxylic acid side chain from a polar to a nonpolar environment by becoming exposed to the lipid phase when G(t) binds. These results show directly that Glu(134) becomes protonated in MII upon G(t) binding and suggest that changes in receptor conformation affect lipid-protein interactions.

Specific interaction of CCR5 amino-terminal domain peptides containing sulfotyrosines with HIV-1 envelope glycoprotein gp120.

The HIV-1 envelope glycoprotein gp120 interacts consecutively with CD4 and the CCR5 coreceptor to mediate the entry of certain HIV-1 strains into target cells. Acidic residues and sulfotyrosines in the amino-terminal domain (Nt) of CCR5 are crucial for viral fusion and entry. We tested the binding of a panel of CCR5 Nt peptides to different soluble gp120/CD4 complexes and anti-CCR5 mAbs. The tyrosine residues in the peptides were sulfated, phosphorylated, or unmodified. None of the gp120/CD4 complexes associated with peptides containing unmodified or phosphorylated tyrosines. The gp120/CD4 complexes containing envelope glycoproteins from isolates that use CCR5 as a coreceptor associated with Nt peptides containing sulfotyrosines but not with peptides containing sulfotyrosines in scrambled Nt sequences. Finally, only peptides containing sulfotyrosines inhibited the entry of an R5 isolate. Our data show that proper posttranslational modification of the CCR5 Nt is required for gp120 binding and viral entry. More importantly, the Nt domain determines the specificity of the interaction between CCR5 and gp120s from isolates that use this coreceptor.

Selective stabilization of the high affinity binding conformation of glucagon receptor by the long splice variant of Galpha(s).

To analyze functional differences in the interactions of the glucagon receptor (GR) with the two predominant splice variants of Galpha(s), GR was covalently linked to the short and the long forms Galpha(s)-S and Galpha(s)-L to produce the fusion proteins GR-Galpha(s)-S and GR-Galpha(s)-L. GR-Galpha(s)-S bound glucagon with an affinity similar to that of GR, while GR-Galpha(s)-L showed a 10-fold higher affinity for glucagon. In the presence of GTPgammaS, GR-Galpha(s)-L reverted to the low affinity glucagon binding conformation. Both GR-Galpha(s)-L and GR-Galpha(s)-S were constitutively active, causing elevated basal levels of cAMP even in the absence of glucagon. A mutant GR that failed to activate G(s) (G23D1R) was fused to Galpha(s)-L. G23D1R-Galpha(s)-L bound glucagon with high affinity, but failed to elevate cAMP levels, suggesting that the mechanisms of GR-mediated Galpha(s)-L activation and Galpha(s)-L-induced high affinity glucagon binding are independent. Both GR-Galpha(s)-S and GR-Galpha(s)-L bound the antagonist desHis(1)[Nle(9),Ala(11),Ala(16)]glucagon amide with affinities similar to GR. The antagonist displayed partial agonist activity with GR-Galpha(s)-L, but not with GR-Galpha(s)-S. Therefore, the partial agonist activity of the antagonist observed in intact cells appears to be due to GRs coupled to Galpha(s)-L. We conclude that Galpha(s)-S and Galpha(s)-L interact differently with GR and that specific coupling of GR to Galpha(s)-L may account for GTP-sensitive high affinity glucagon binding.

A binding pocket for a small molecule inhibitor of HIV-1 entry within the transmembrane helices of CCR5.

HIV-1 entry into CD4(+) cells requires the sequential interactions of the viral envelope glycoproteins with CD4 and a coreceptor such as the chemokine receptors CCR5 and CXCR4. A plausible approach to blocking this process is to use small molecule antagonists of coreceptor function. One such inhibitor has been described for CCR5: the TAK-779 molecule. To facilitate the further development of entry inhibitors as antiviral drugs, we have explored how TAK-779 acts to prevent HIV-1 infection, and we have mapped its site of interaction with CCR5. We find that TAK-779 inhibits HIV-1 replication at the membrane fusion stage by blocking the interaction of the viral surface glycoprotein gp120 with CCR5. We could identify no amino acid substitutions within the extracellular domain of CCR5 that affected the antiviral action of TAK-779. However, alanine scanning mutagenesis of the transmembrane domains revealed that the binding site for TAK-779 on CCR5 is located near the extracellular surface of the receptor, within a cavity formed between transmembrane helices 1, 2, 3, and 7.

Selective reconstitution of human D4 dopamine receptor variants with Gi alpha subtypes.

G protein-coupled receptors (GPCRs) are seven-transmembrane (TM) helical proteins that bind extracellular molecules and transduce signals by coupling to heterotrimeric G proteins in the cytoplasm. The human D4 dopamine receptor is a particularly interesting GPCR because the polypeptide loop linking TM helices 5 and 6 (loop i3) may contain from 2 to 10 similar direct hexadecapeptide repeats. The precise role of loop i3 in D4 receptor function is not known. To clarify the role of loop i3 in G protein coupling, we constructed synthetic genes for the three main D4 receptor variants. D4-2, D4-4, and D4-7 receptors contain 2, 4, and 7 imperfect hexadecapeptide repeats in loop i3, respectively. We expressed and characterized the synthetic genes and found no significant effect of the D4 receptor polymorphisms on antagonist or agonist binding. We developed a cell-based assay where activated D4 receptors coupled to a Pertussis toxin-sensitive pathway to increase intracellular calcium concentration. Studies using receptor mutants showed that the regions of loop i3 near TM helices 5 and 6 were required for G protein coupling. The hexadecapeptide repeats were not required for G protein-mediated calcium flux. Cell membranes containing expressed D4 receptors and receptor mutants were reconstituted with purified recombinant G protein alpha subunits. The results show that each D4 receptor variant is capable of coupling to several G(i)alpha subtypes. Furthermore, there is no evidence of any quantitative difference in G protein coupling related to the number of hexadecapeptide repeats in loop i3. Thus, loop i3 is required for D4 receptors to activate G proteins. However, the polymorphic region of the loop does not appear to affect the specificity or efficiency of G(i)alpha coupling.

The amino terminus of the fourth cytoplasmic loop of rhodopsin modulates rhodopsin-transducin interaction.

Rhodopsin is a seven-transmembrane helix receptor that binds and catalytically activates the heterotrimeric G protein transducin (G(t)). This interaction involves the cytoplasmic surface of rhodopsin, which comprises four putative loops and the carboxyl-terminal tail. The fourth loop connects the carboxyl end of transmembrane helix 7 with Cys(322) and Cys(323), which are both modified by membrane-inserted palmitoyl groups. Published data on the roles of the fourth loop in the binding and activation of G(t) are contradictory. Here, we attempt to reconcile these conflicts and define a role for the fourth loop in rhodopsin-G(t) interactions. Fluorescence experiments demonstrated that a synthetic peptide corresponding to the fourth loop of rhodopsin inhibited the activation of G(t) by rhodopsin and interacted directly with the alpha subunit of G(t). A series of rhodopsin mutants was prepared in which portions of the fourth loop were replaced with analogous sequences from the beta(2)-adrenergic receptor or the m1 muscarinic receptor. Chimeric receptors in which residues 310-312 were replaced could not efficiently activate G(t). The defect in G(t) interaction in the fourth loop mutants was not affected by preventing palmitoylation of Cys(322) and Cys(323). We suggest that the amino terminus of the fourth loop interacts directly with G(t), particularly with Galpha(t), and with other regions of the intracellular surface of rhodopsin to support G(t) binding.

Mutation of the fourth cytoplasmic loop of rhodopsin affects binding of transducin and peptides derived from the carboxyl-terminal sequences of transducin alpha and gamma subunits.

The role of the putative fourth cytoplasmic loop of rhodopsin in the binding and catalytic activation of the heterotrimeric G protein, transducin (G(t)), is not well defined. We developed a novel assay to measure the ability of G(t), or G(t)-derived peptides, to inhibit the photoregeneration of rhodopsin from its active metarhodopsin II state. We show that a peptide corresponding to residues 340-350 of the alpha subunit of G(t), or a cysteinyl-thioetherfarnesyl peptide corresponding to residues 50-71 of the gamma subunit of G(t), are able to interact with metarhodopsin II and inhibit its photoconversion to rhodopsin. Alteration of the amino acid sequence of either peptide, or removal of the farnesyl group from the gamma-derived peptide, prevents inhibition. Mutation of the amino-terminal region of the fourth cytoplasmic loop of rhodopsin affects interaction with G(t) (Marin, E. P., Krishna, A. G., Zvyaga T. A., Isele, J., Siebert, F., and Sakmar, T. P. (2000) J. Biol. Chem. 275, 1930-1936). Here, we provide evidence that this segment of rhodopsin interacts with the carboxyl-terminal peptide of the alpha subunit of G(t). We propose that the amino-terminal region of the fourth cytoplasmic loop of rhodopsin is part of the binding site for the carboxyl terminus of the alpha subunit of G(t) and plays a role in the regulation of betagamma subunit binding.

pH dependence of photolysis intermediates in the photoactivation of rhodopsin mutant E113Q.

Glutamic acid at position 113 in bovine rhodopsin ionizes to form the counterion to the protonated Schiff base (PSB), which links the 11-cis-retinylidene chromophore to opsin. Photoactivation of rhodopsin requires both Schiff base deprotonation and neutralization of Glu-113. To better understand the role of electrostatic interactions in receptor photoactivation, absorbance difference spectra were collected at time delays from 30 ns to 690 ms after photolysis of rhodopsin mutant E113Q solubilized in dodecyl maltoside at different pH values at 20 degrees C. The PSB form (pH 5. 5, lambda(max) = 496 nm) and the unprotonated Schiff base form (pH 8. 2, lambda(max) = 384 nm) of E113Q rhodopsin were excited using 477 nm or 355 nm light, respectively. Early photointermediates of both forms of E113Q were qualitatively similar to those of wild-type rhodopsin. In particular, early photoproducts with spectral shifts to longer wavelengths analogous to wild-type bathorhodopsin were seen. In the case of the basic form of E113Q, the absorption maximum of this intermediate was at 408 nm. These results suggest that steric interaction between the retinylidene chromophore and opsin, rather than charge separation, plays the dominant role in energy storage in bathorhodopsin. After lumirhodopsin, instead of deprotonating to form metarhodopsin I(380) on the submillisecond time scale as is the case for wild type, the acidic form of E113Q produced metarhodopsin I(480), which decayed very slowly (exponential lifetime = 12 ms). These results show that Glu-113 must be present for efficient deprotonation of the Schiff base and rapid visual transduction in vertebrate visual pigments.

Colour tuning mechanisms of visual pigments.

Spectral tuning by visual pigments involves modulation of physical properties of the 11-cis-retinylidene protonated Schiff base (PSB) chromophore by amino acid side chains in and around the chromophore-binding pocket. Specific molecular contacts between the chromophore and the amino acid side chains of the opsin chromophore-binding pocket have been determined recently using an interdisciplinary approach consisting of site-directed mutagenesis, optical and vibrational spectroscopy, and molecular graphics modelling. These studies provide insight into the mechanism of spectral tuning among visual pigments. In blue pigments a majority of the opsin shift is caused by polar amino acid side chains arrayed about the PSB to increase the energy gap between the ground (S0) and excited states (S1). In addition, a specific tyrosine near the chromophore ring causes a decrease in solvent polarizability. Other amino acid residues alter the binding pocket structure to strengthen electrostatic interaction between the PSB and its counterion and/or solvent dipoles. In the green and red pigments, the work of Kochendoerfer et al (1997; Biochemistry 26:6577-6587) demonstrates that local structural perturbations at the PSB or elsewhere are not responsible for spectral tuning. Instead, the green-to-red opsin shift is best explained by dipolar side chains near the chromophore ring that lower the transition energy that occurs upon electronic excitation by affecting the change in electric dipole moment. In summary, the absorption maximum of a visual pigment is primarily regulated by the interaction of the chromophore charge distribution with dipolar residues in its opsin chromophore-binding pocket. The work presented in this paper is reported in greater detail in Lin et al.

Dopamine D4/D2 receptor selectivity is determined by A divergent aromatic microdomain contained within the second, third, and seventh membrane-spanning segments.

Conserved features of the sequences of dopamine receptors and of homologous G-protein-coupled receptors point to regions, and amino acid residues within these regions, that contribute to their ligand binding sites. Differences in binding specificities among the catecholamine receptors, however, must stem from their nonconserved residues. Using the substituted-cysteine accessibility method, we have identified the residues that form the surface of the water-accessible binding-site crevice in the dopamine D2 receptor. Of approximately 80 membrane-spanning residues that differ between the D2 and D4 receptors, only 20 were found to be accessible, and 6 of these 20 are conservative aliphatic substitutions. In a D2 receptor background, we mutated the 14 accessible, nonconserved residues, individually or in combinations, to the aligned residues in the D4 receptor. We also made the reciprocal mutations in a D4 receptor background. The combined substitution of four to six of these residues was sufficient to switch the affinity of the receptors for several chemically distinct D4-selective antagonists by three orders of magnitude in both directions (D2- to D4-like and D4- to D2-like). The mutated residues are in the second, third, and seventh membrane-spanning segments (M2, M3, M7) and form a cluster in the binding-site crevice. Mutation of a single residue in this cluster in M2 was sufficient to increase the affinity for clozapine to D4-like levels. We can rationalize the data in terms of a set of chemical moieties in the ligands interacting with a divergent aromatic microdomain in M2-M3-M7 of the D2 and D4 receptors.