Defining the interface between the C-terminal fragment of alpha-transducin and photoactivated rhodopsin

A novel combination of experimental data and extensive computational modeling was used to explore probable protein-protein interactions between photoactivated rhodopsin (R*) and experimentally determined R*-bound structures of the C-terminal fragment of alpha-transducin (Gt(alpha)(340-350)) and its analogs. Rather than using one set of loop structures derived from the dark-adapted rhodopsin state, R* was modeled in this study using various energetically feasible sets of intracellular loop (IC loop) conformations proposed previously in another study. The R*-bound conformation of Gt(alpha)(340-350) and several analogs were modeled using experimental transferred nuclear Overhauser effect data derived upon binding R*. Gt(alpha)(340-350) and its analogs were docked to various conformations of the intracellular loops, followed by optimization of side-chain spatial positions in both R* and Gt(alpha)(340-350) to obtain low-energy complexes. Finally, the structures of each complex were subjected to energy minimization using the OPLS/GBSA force field. The resulting residue-residue contacts at the interface between R* and Gt(alpha)(340-350) were validated by comparison with available experimental data, primarily from mutational studies. Computational modeling performed for Gt(alpha)(340-350) and its analogs when bound to R* revealed a consensus of general residue-residue interactions, necessary for efficient complex formation between R* and its Gt(alpha) recognition motif.     

Binding of transducin and transducin-derived peptides to rhodopsin studies by attenuated total reflection-Fourier transform infrared difference spectroscopy.

Fourier transform infrared difference spectroscopy combined with the attenuated total reflection technique allows the monitoring of the association of transducin with bovine photoreceptor membranes in the dark. Illumination causes infrared absorption changes linked to formation of the light-activated rhodopsin-transducin complex. In addition to the spectral changes normally associated with meta II formation, prominent absorption increases occur at 1735 cm-1, 1640 cm-1, 1550 cm-1, and 1517 cm-1. The D2O sensitivity of the broad carbonyl stretching band around 1735 cm-1 indicates that a carboxylic acid group becomes protonated upon formation of the activated complex. Reconstitution of rhodopsin into phosphatidylcholine vesicles has little influence on the spectral properties of the rhodopsin-transducin complex, whereas pH affects the intensity of the carbonyl stretching band. AC-terminal peptide comprising amino acids 340-350 of the transducin alpha-subunit reproduces the frequencies and isotope sensitivities of several of the transducin-induced bands between 1500 and 1800 cm-1, whereas an N-terminal peptide (aa 8-23) does not. Therefore, the transducin-induced absorption changes can be ascribed mainly to an interaction between the transducin-alpha C-terminus and rhodopsin. The 1735 cm-1 vibration is also seen in the complex with C-terminal peptides devoid of free carboxylic acid groups, indicating that the corresponding carbonyl group is located on rhodopsin.     

Reaction rate and collisional efficiency of the rhodopsin-transducin system in intact retinal rods.

A model of transducin activation is constructed from its partial reactions (formation of metarhodopsin II, association, and dissociation of the rhodopsin-transducin complex). The kinetic equations of the model are solved both numerically and, for small photoactivation, analytically. From data on the partial reactions in vitro, rate and activation energy profile of amplified transducin turnover are modeled and compared with measured light-scattering signals of transducin activation in intact retinal rods. The data leave one free parameter, the rate of association between transducin and rhodopsin. Best fit is achieved for an activation energy of 35 kJ/mol, indicating lateral membrane diffusion of the proteins as its main determinant. The absolute value of the association rate is discussed in terms of the success of collisions to form the catalytic complex. It is greater than 30% for the intact retina and 10 times lower after permeabilization with staphylococcus aureus alpha-toxin. Dissociation rates for micromolar guanosinetriphosphale (GTP) (Kohl, B., and K. P. Hofmann, 1987. Biophys. J. 52:271-277) must be extrapolated linearly up to the millimolar range to explain the rapid transducin turnover in situ. This is interpreted by an unstable rhodopsin-transducin-GTP transient state. At the time of maximal turnover after a flash, the rate of activation is determined as 30, 120, 800, 2,500, and 4,000 activated transducins per photoactivated rhodopsin and second at 5, 10, 20, 30, 37 degrees C, respectively.     

Bound conformations for ligands for G-protein coupled receptors

Summary The conformation of the C-terminus of the α-subunit of transducin, the G-protein of vision, has been determined by transfer NOE when bound to activated (MII) rhodopsin. One hundred three new NOE constraints are apparent when light is shown on a mixture of rhodopsin bilayers and the undecapeptide. Analogs of the α-peptide with covalent constraints were designed restricting the bound conformation; they stabilize MII thus supporting the deduced structure. The NMR structure of a complex of the intracellular loops of rhodopsin facilitates docking of the α-peptide and also shows proximity of residues known by mutational analysis to interact to generate the activated rhodopsin-transducin interface. This constrains the location of transmembrane helices in the structure of activated rhodopsin. Methods for the prediction of affinity have been used to estimate the relative binding constants of peptide analogs with the loop complex and show strong correlation with experimental data. Various models of the rhodopsin-transmembrane helical segments have been computationally fused with distance geometry to determine the overall model which best fits the experimental data on the rhodopsin-transducin interface.     

Roles of the transducin alpha-subunit alpha4-helix/alpha4-beta6 loop in the receptor and effector interactions

The visual GTP-binding protein, transducin, couples light-activated rhodopsin (R*) with the effector enzyme, cGMP phosphodiesterase in vertebrate photoreceptor cells. The region corresponding to the alpha4-helix and alpha4-beta6 loop of the transducin alpha-subunit (Gtalpha) has been implicated in interactions with the receptor and the effector. Ala-scanning mutagenesis of the alpha4-beta6 region has been carried out to elucidate residues critical for the functions of transducin. The mutational analysis supports the role of the alpha4-beta6 loop in the R*-Gtalpha interface and suggests that the Gtalpha residues Arg310 and Asp311 are involved in the interaction with R*. These residues are likely to contribute to the specificity of the R* recognition. Contrary to the evidence previously obtained with synthetic peptides of Gtalpha, our data indicate that none of the alpha4-beta6 residues directly or significantly participate in the interaction with and activation of phosphodiesterase. However, Ile299, Phe303, and Leu306 form a network of interactions with the alpha3-helix of Gtalpha, which is critical for the ability of Gtalpha to undergo an activational conformational change. Thereby, Ile299, Phe303, and Leu306 play only an indirect role in the effector function of Gtalpha.     

FTIR spectroscopy of complexes formed between metarhodopsin II and C-terminal peptides from the G-protein alpha- and gamma-subunits

Metarhodopsin II (MII) provides the active conformation of rhodopsin for interaction with the G-protein, Gt. Fourier transform infrared spectra from samples prepared by centrifugation reflect the pH dependent equilibrium between MII and inactive metarhodopsin I. C-terminal synthetic peptides (Gtalpha(340-350) and Gtgamma(60-71)farnesyl) stabilize MII. We find that both peptides cause similar spectral changes not seen with control peptides (Gtalpha (K341R, L349A) and non-farnesylated Gtgamma). The spectra reflect all the protonation dependent bands normally observed when MII is formed at acidic pH. Beside the protonation dependent bands, additional features, similar with both peptides, appear in the amide I and II regions.     

Rhodopsin-interacting surface of the transducin gamma subunit

The visual signaling pathway is initiated by photoactivation of the GPCR rhodopsin, which activates nucleotide exchange on the heterotrimeric G-protein transducin (Gt). Domains on both Gtalpha and Gtbetagamma subunits participate in coupling to rhodopsin. Previously, we have shown by high-resolution NMR that the farnesylated C-terminal peptide of Gtgamma(60-71), DKNPFKELKGGC, assumes an amphipathic helical conformation during interaction with metarhodopsin II [Kisselev, O. G., and Downs, M. A. (2003) Structure 11, 367-373]. This conformation was docked to the structure of holo-Gt to create a model of rhodopsin-Gt interaction. Here we test this model by mutational analysis of Gt. To evaluate the contribution of specific amino acids of the Gtgamma C-terminal region involved in binding and GTP-dependent release of transducin from native rhodopsin membranes, we have systematically substituted each of the amino acids in the C-terminal region of Gtgamma for alanine. The mutants were co-expressed with six-histidine-tagged Gtbeta subunits in Sf9 insect cells. The Gtbeta-6-His-gamma mutant proteins were purified and assayed in the presence of Gtalpha for the GTP-dependent interactions with light-activated rhodopsin. Several of the alanine mutants, N62A, P63A, and F64A, exhibited significant functional defects at the level of R*-Gt complex formation. These data show that the conserved N-terminal end of the helical domain in the Gtgamma(60-71) region has the most significant effect on rhodopsin-Gt interactions, which places important constraints on the model of the rhodopsin-Gt complex.     

Different properties of the native and reconstituted heterotrimeric G protein transducin

Visual signal transduction serves as one of the best understood G protein-coupled receptor signaling systems. Signaling is initiated when a photon strikes rhodopsin (Rho) causing a conformational change leading to productive interaction of this G protein-coupled receptor with the heterotrimeric G protein, transducin (Gt). Here we describe a new method for Gt purification from native bovine rod photoreceptor membranes without subunit dissociation caused by exposure to photoactivated rhodopsin (Rho*). Native electrophoresis followed by immunoblotting revealed that Gt purified by this method formed more stable heterotrimers and interacted more efficiently with membranes containing Rho* or its target, phosphodiesterase 6, than did Gt purified by a traditional method involving subunit dissociation and reconstitution in solution without membranes. Because these differences could result from selective extraction, we characterized the type and amount of posttranslational modifications on both purified native and reconstituted Gt preparations. Similar N-terminal acylation of the Gtalpha subunit was observed for both proteins as was farnesylation and methylation of the terminal Gtgamma subunit Cys residue. However, hydrogen/deuterium exchange experiments revealed less incorporation of deuterium into the Gtalpha and Gtbeta subunits of native Gt as compared to reconstituted Gt. These findings may indicate differences in conformation and heterotrimer complex formation between the two preparations or altered stability of the reconstituted Gt that assembles differently than the native protein. Therefore, Gt extracted and purified without subunit dissociation appears to be more appropriate for future studies.     

Bound conformations for ligands for G-protein coupled receptors

The conformation of the C-terminus of the -subunit of transducin, the G-protein of vision, has been determined by transfer NOE when bound to activated (MII) rhodopsin. One hundred three new NOE constraints are apparent when light is shown on a mixture of rhodopsin bilayers and the undecapeptide. Analogs of the -peptide with covalent constraints were designed restricting the bound conformation; they stabilize MII thus supporting the deduced structure. The NMR structure of a complex of the intracellular loops of rhodopsin facilitates docking of the -peptide and also shows proximity of residues known by mutational analysis to interact to generate the activated rhodopsin-transducin interface. This constrains the location of transmembrane helices in the structure of activated rhodopsin. Methods for the prediction of affinity have been used to estimate the relative binding constants of peptide analogs with the loop complex and show strong correlation with experimental data. Various models of the rhodopsin-transmembrane helical segments have been computationally fused with distance geometry to determine the overall model which best fits the experimental data on the rhodopsin-transducin interface.     

Interaction between the main components of the vertebrate retinal rod phototransduction system in solutions of detergent n-nonyl-β-D-glucoside

cGMP-Specific phosphodiesterase (PDE6) is the key enzyme of the phototransduction system of vertebrate retinal rod outer segments (ROS). The properties of PDE in extracts prepared by solubilization of bovine ROS in a high concentration (0.5% w/v) of detergent n-nonyl-β-D-glucoside (NG) and following centrifugation (ROS-NG) have been studied. Basal PDE activity of the preparations was low, but it greatly (>50-fold) increased (up to ∼20 μmol cGMP hydrolyzed/min per mg rhodopsin (R)) in the presence of trypsin. In bleached GTPγS-containing preparations the specific PDE activity was dependent on ROS-NG concentration and was half-maximal at about 0.8 μM of ROS G protein transducin (Gt). In dark-adapted GTPγS-containing ROS-NG preparations bleaching of 0.2% of the rhodopsin resulted in half-maximal PDE activation. The same result was obtained when PDE in dark-adapted ROS-NG preparations was activated by addition of a highly purified bleached rhodopsin solubilized by 0.5% solution of NG. The results demonstrate that the presence of NG has no significant influence either on the properties of the main ROS phototrans-duction system elements (R, Gt and PDE) or on the interaction between photoactivated R and Gt and suggest that the detergent NG can be used for crystallization of the rhodopsin-transducin complex.     

Structural requirements for the stabilization of metarhodopsin II by the C terminus of the alpha subunit of transducin

The retinal receptor rhodopsin undergoes a conformational change upon light excitation to form metarhodopsin II (Meta II), which allows interaction and activation of its cognate G protein, transducin (G(t)). A C-terminal 11-amino acid peptide from transducin, G(talpha)-(340-350), has been shown to both bind and stabilize the Meta II conformation, mimicking heterotrimeric G(t). Using a combinatorial library we identified analogs of G(talpha)-(340-350) that bound light-activated rhodopsin with high affinity (Martin, E. L., Rens-Domiano, S., Schatz, P. J., and Hamm, H. E. (1996) J. Biol. Chem. 271, 361-366). We have made peptides with key substitutions either on the background of the native G(talpha)-(340-350) sequence or on the high affinity sequences and used the stabilization of Meta II as a tool to determine which amino acids are critical in G protein-rhodopsin interaction. Removal of the positive charge at the N termini by acylation or delocalization of the charge by K to R substitution enhances the affinity of the G(talpha)-(340-350) peptides for Meta II, whereas a decrease was observed following C-terminal amidation. Cys-347, a residue conserved in pertussis toxin-sensitive G proteins, was shown to interact with a hydrophobic site in Meta II. These studies provide further insight into the mechanism of interaction between the G(talpha) C terminus and light-activated rhodopsin.     

Rhodopsin-transducin interface: studies with conformationally constrained peptides.

To probe the interaction between transducin (G(t)) and photoactivated rhodopsin (R*), 14 analog peptides were designed and synthesized restricting the backbone of the R*-bound structure of the C-terminal 11 residues of G(t)alpha derived by transferred nuclear Overhauser effect (TrNOE) NMR. Most of the analogs were able to bind R*, supporting the TrNOE structure. Improved affinities of constrained peptides indicated that preorganization of the bound conformation is beneficial. Cys347 was found to be a recognition site; particularly, the free sulfhydryl of the side chain seems to be critical for R* binding. Leu349 was another invariable residue. Both Ile and tert-leucine (Tle) mutations for Leu349 significantly reduced the activity, indicating that the Leu side chain is in intimate contact with R*. The structure of R* was computer generated by moving helix 6 from its position in the crystal structure of ground-state rhodopsin (R) based on various experimental data. Seven feasible complexes were found when docking the TrNOE structure with R* and none with R. The analog peptides were modeled into the complexes, and their binding affinities were calculated. The predicted affinities were compared with the measured affinities to evaluate the modeled structures. Three models of the R*/G(t)alpha complex showed strong correlation to the experimental data.     

Three-dimensional model for meta-II rhodopsin,an activated G-protein-coupled receptor

A novel approach that iteratively combined the results of energy calculations and experimental data was used to generate a three-dimensional (3D) model of the photoactivated state (R*) of bovine rhodopsin (Rh). The approach started with simplified energy calculations in an effort to find a set of sterically and energetically reasonable options for transmembrane (TM) helix arrangements with all-trans-retinal. Various 3D models of TM helix packing found by computations were then compared to limited site-directed spin-label experimental data regarding the transition of the TM helices of Rh in the inactive state (R) to those in the R* state to identify the most plausible model of the TM helical bundle. At the next step, all non-TM structural elements, such as the non-TM helix 8, the N- and C-terminal fragments, and the loops connecting TM helices, were reconstructed, and after the entire R* structure had been relaxed, all other currently available additional experimental data, both mutational and spectroscopic, on the structure of the meta-II state of rhodopsin were used to validate the resulting 3D model.     

Calcium regulates the rate of rhodopsin disactivation and the primary amplification step in visual transduction

The kinetics of the light-induced activation of transducin as well as the subsequent disactivation process can be monitored by means of a specific light scattering transient PA. In this communication it is demonstrated that the rate of transducin disactivation is calcium dependent, increasing when the calcium concentration is decreased. As a consequence of the accelerated recovery in low calcium, the time to the peak of the transducin activation process is shortened and the gain of the primary amplification step, i.e. the number of transducin molecules activated per bleached rhodopsin, is reduced. Experiments using hydroxylamine as an artificial quencher of rhodopsin activity suggest that calcium acts upon rhodopsin kinase and not upon the rate of the GTPase. This would indicate that calcium may control visual adaptation not only by regulating guanine cyclase activity, but also by affecting the primary step in the transduction cascade, the rhodopsin-transducin coupling.     

Probing rhodopsin-transducin interactions by surface modification and mass spectrometry

The interactions of rhodopsin and the alpha-subunit of transducin (G(t)) have been mapped using a surface modification "footprinting" approach in conjunction with mass spectrometric analysis employing a synthetic peptide corresponding to C-terminal residues 340-350 of the alpha-subunit of G(t), G(t)alpha(340-350). Membrane preparations of unactivated (Rh) and light-activated rhodopsin (Rh*), each in the presence or absence of G(t)alpha(340-350), were acetylated with the water-soluble reagent sulfosuccinimidyl acetate, and the extent of the acetylation was determined by mass spectrometry. By comparing the differences in acetylation among Rh, Rh*, and the Rh-G(t)alpha(340-350) and Rh*-G(t)alpha(340-350) complexes, we demonstrate that the surface exposure of the acetylation sites was reduced by the conformational change associated with light activation, and that binding of G(t)alpha(340-350) blocks acetylation sites on cytoplasmic loops 1, 2, and 4 of Rh*. In addition, we show evidence of interaction between the end of the C-terminal tail of rhodopsin and G(t)alpha in the unactivated state of rhodopsin.     

Dark and photoactivated rhodopsin share common binding modes to transducin

The structure of the photoactivated deprotonated rhodopsin intermediate was compared with two different structures of dark rhodopsin. Structure comparisons relied on the computation of molecular indices and on docking simulations with heterotrimeric transducin (Gt). The results of this study provide the first evidence that dark and photoactivated rhodopsins share a common recognition mode to Gt, characterized by the docking of the Gt alpha C-tail in the proximity to the E/DRY motif of rhodopsin.     

Chemical modification of transducin with iodoacetic acid: transducin-alpha carboxymethylated at Cys(347) allows transducin binding to Light-activated rhodopsin but prevents its release in the presence of GTP.

Modification of transducin (T) with iodoacetic acid (IAA) inhibited its light-dependent guanine nucleotide-binding activity. Approximately 1 mol of [(3)H]IAA was incorporated per mole of T. Cys(347), located on the alpha-subunit of T (T(alpha)), was identified as the major labeled residue in the [(3)H]IAA-modified holoenzyme. In contrast, Cys(135) and Cys(347) were modified with [(3)H]IAA in the isolated T(alpha). IAA-modified T was able to bind tightly to photoexcited rhodopsin (R*), but GTP did not promote the dissociation of the complex between alkylated T and R*. In addition, R* protected against the inhibition of T by IAA. A comparable inactivation of T and analogous interactions between T and R* were observed when 2-nitro 5-thiocyanobenzoic acid (NTCBA) was used as the modifying reagent (J. O. Ortiz and J. Bubis, 2001, Effects of differential sulfhydryl group-specific labeling on the rhodopsin and guanine nucleotide binding activities of transducin, Arch. Biochem. Biophys. 387, 233-242). However, while carboxymethylated T was capable of liberating GDP in the presence of R*, NTCBA-modified T was unable to release the guanine nucleotide diphosphate upon incubation with the photoactivated receptor. Thus, IAA-labeling stabilized a T:R* complex intermediate carrying the empty nucleotide pocket conformation of T. On the other hand, NTCBA-modified T seemed to be "locked" in the GDP-bound state of T, even in the presence of R*.     

Preparation of an activated rhodopsin/transducin complex using a constitutively active mutant of rhodopsin

The interaction of rhodopsin and transducin has been the focus of study for more than 30 years, but only recently have efforts to purify an activated complex in detergent solution materialized. These efforts have used native rhodopsin isolated from bovine retina and employed either sucrose density gradient centrifugation or size exclusion chromatography to purify the complex. While there is general agreement on most properties of the activated complex, subunit stoichiometry is not yet settled, with rhodopsin/transducin molar ratios of both 2/1 and 1/1 reported. In this report, we introduce methods for preparation of the complex that include use of recombinant rhodopsin, so as to take advantage of mutations that confer constitutive activity and enhanced thermal stability on the protein, and immunoaffinity chromatography for purification of the complex. We show that chromatography on ConA-Sepharose can substitute for the immunoaffinity column and that bicelles can be used instead of detergent solution. We demonstrate the following: that rhodopsin has a covalently bound all-trans-retinal chromophore and therefore corresponds to the active metarhodopin II state; that transducin has an empty nucleotide-binding pocket; that the isolated complex is active and dissociates upon addition of guanine nucleotide; and finally that the stoichiometry corresponds reproducibly to a 1/1 molar ratio of rhodopsin to transducin.     

Rhodopsin and phototransduction: a model system for G protein-linked receptors.

Rhodopsin is the photoreceptor protein in rod cells of the vertebrate retina and the first member of the class of G protein-coupled receptors for which the amino acid sequence was determined. Rhodopsin is available in greater quantities than any other receptor of its class and therefore has been studied biochemically and biophysically by methods difficult or impossible to apply to its fellow receptors. Such studies support a model in which rhodopsin consists of seven transmembrane helices that form a binding pocket for its ligand, 11-cis retinal. Insights into the structure and function of rhodopsin serve as a model for understanding the structure and function of other members of the receptor class. Rhodopsin undergoes a change in conformation upon photoexcitation and activates a G protein, transducin, and is phosphorylated by a receptor-specific kinase, rhodopsin kinase. The phosphorylated photoactivated rhodopsin is bound by arrestin, thereby terminating activity of the receptor in the signal transduction process. These auxiliary proteins that function with rhodopsin on rod cells serve as models for understanding how other members of the receptor family may function in conjunction with other G proteins, kinases, and arrestin-like proteins.     

Chemical modification of bovine transducin: effect of fluorescein 5'-isothiocyanate labeling on activities of the transducin alpha subunit

Fluorescein 5'-isothiocyanate (FITC) was used to modify the lysine residues of bovine transducin (T), a GTP-binding protein involved in phototransduction of rod photoreceptor cells. The incorporation of FITC showed a stoichiometry of approximately 1 mol of FITC/mol of transducin. The labeling was specific for the T alpha subunit. There was no significant incorporation on the T beta gamma subunit. The modification had no effect on the transducin-rhodopsin interaction or on the binding of guanosine 5'-(beta, gamma-imidotriphosphate) [Gpp(NH)p] to transducin in the presence of photolyzed rhodopsin. The dissociation of the FITC-transducin-Gpp(NH)p complex from rhodopsin membrane remained unchanged. However, the intrinsic GTPase activity of T alpha and its ability to activate the cGMP phosphodiesterase were diminished by FITC modification. The rate of FITC labeling of the transducin-Gpp(NH)p complex was about 3-fold slower than that of transducin. Limited tryptic digestion and peptide mapping were used to localize the FITC labeling site. The majority of the FITC label was on the 23-kilodalton fragment, and a minor amount was on the 9-kilodalton fragment of the T alpha subunit. These results indicate that FITC labeling does not alter the activation of transducin by photolyzed rhodopsin but does affect the GTP hydrolytic activity as well as the GTP-induced conformational change of T alpha, which ultimately leads to the activation of cGMP phosphodiesterase.