|Home | About | Journals | Submit | Contact Us | Français|
G protein-Coupled Receptors (GPCRs) use a complex series of intramolecular conformational changes to couple agonist binding to the binding and activation of cognate heterotrimeric G protein (Gαβγ). The mechanisms underlying this long-range activation have been identified using a variety of biochemical and structural approaches and have primarily used visual signal transduction via the GPCR rhodopsin and cognate heterotrimeric G protein transducin (Gt) as a model system. In this chapter, we will review the methods that have revealed allosteric signaling through rhodopsin and transducin. These methods can be applied to a variety of GPCR-mediated signaling pathways.
GPCRs regulate physical processes ranging from vision to olfaction to cardiac contractility to neurotransmission, yet these receptors likely act via a conserved signaling cycle (Figure 1). In this cycle, the activation states of GPCRs and cognate G proteins are encoded into protein conformations that influence the binding of partner signaling molecules. For example, agonist binding to a GPCR induces a conformational change that forms a binding site for heterotrimeric G proteins 25 Å away (Figure 2) (1, 2). The high affinity GPCR-G protein complex promotes exchange of GDP for GTP in the guanine nucleotide-binding pocket of the of the G protein α subunit (Gα) another 30 Å away from the proposed receptor-binding site (3, 4). The identity of the bound nucleotide then influences the formation of an effector-binding site, yet another 10 Å from bound GTP (Figure 3) (5–7). The activation steps in this complex series have been investigated by a number of complementary methods.
Each GPCR has incredible fidelity to a small set of cognate ligands that act as agonists, antagonists, or inverse agonists to promote different activation states (8). Despite this diversity, GPCRs all share a common architecture comprising seven transmembrane helices (9, 10), and several conserved sequence motifs believed to be critical for regulation of receptor activation (11–15). As a result, many of the general mechanisms of activation are assumed to be shared amongst all GPCRs.
The best-characterized GPCR, rhodopsin, is responsible for perception of light under low light conditions. Numerous properties of this photoreceptor have made it amenable to characterization by both in vivo and in vitro methods. Rhodopsin is naturally abundant (16), easily purified, stable in a variety of detergents, and has spectroscopic properties that provide a convenient method to monitor activation. In addition, rhodopsin covalently ligates an inverse agonist, 11-cis-retinal, that locks it into an inactive conformation, which decreases the conformational heterogeneity. Accordingly, rhodopsin was the first GPCR to be characterized by x-ray crystallography (17), and eight years elapsed before the structure of other GPCRs (18–22) verified the conserved arrangement of transmembrane helices in the family.
Cryoelectron microscopy (10, 23), site-directed spin labeling, and electron paramagnetic resonance spectroscopy (SDSL-EPR) (24, 25) suggested that large conformational changes accompany rhodopsin activation, the most dramatic of which is the rigid body movement of helix VI away from helix III (Figure 3). This helix movement has long been hypothesized to be required for the formation of a G protein-binding site, and the recent crystal structures of opsin (2) and opsin in complex with a high affinity peptide mimicking the C-αt subunit (1) confirmed this hypothesis.
Heterotrimeric G proteins comprise Gα, Gβ, and Gγ (26). Upon activation by GPCRs, heterotrimeric G proteins disassociate into GTP-bound Gα, and Gβγ, both of which can interact with effector proteins to regulate downstream signaling (Figure 1) (27). In humans, there are currently 18 known G α subtypes belonging to 4 subfamilies, 5 Gβ isoforms, and 12 Gγ isoforms. Each has differing binding specificities for both effectors and GPCRs (28).
Gα and Gβ γ can act on a diverse variety of effectors, but like their cognate GPCRs, their mechanisms of activation and deactivation are likely conserved throughout the family. Numerous x-ray crystal structures have revealed how the binding of GDP, GDP-AlF4− or GTP γS to the G α subunit influences the conformations of three nearby surface loops known as switches I–III ( 4, 6, 29). These switch regions form a contiguous surface for effector binding (Figure 3b) (30).
While rhodopsin is by far the most extensively characterized GPCR, its cognate G protein Gt does not heterologously express to levels sufficient for biochemical characterization, which limits techniques that require mutagenesis. However, the large quantities of endogenously expressed transducin in bovine retina has allowed the native protein to be studied and crystallized. For techniques requiring mutagenesis, the close homolog of the Gαt subunit, Gαi (with 67% identity and 82% similarity to Gαt) is frequently used to understand general mechanisms of heterotrimeric G protein activation (31–34). Gαi can be activated by rhodopsin, is easily expressed in bacteria, and can be reconstituted with Gβγ to form a fully functional chimeric system (4–6).
The rate-limiting step in GPCR signaling is receptor-catalyzed GDP release from the Gα subunit of the heterotrimer. A combination of site-directed mutagenesis (36–39), peptide-mapping (3, 40–43), and chemical crosslinking (42–45) demonstrated that the α4–β6 loop, α3–β5 loop, N-terminus, and C-terminus of the Gα subunit as well as the C-termini of the G β G γ subunits are implicated in binding to rhodopsin. While these motifs are distantly arranged in the primary sequence, they form a contiguous surface on the folded protein (Figure 3a) (46, 47).
The α5 helix, located at the C-terminus of the Gα subunit, is perhaps the best-studied region of receptor interaction and independently binds within a pocket on activated receptor (1). Multiple studies have suggested that receptor induced rotation and translation of the α5 helix is a requirement for GDP release from Gα (26), and movement of the α5 helix dipole may be important for weakening the interactions between the bound nucleotide and Gα (43).
Much less is understood about how receptor interactions with the Gβγ subunits contribute to G protein activation. Until recently, two major hypotheses existed: the gearshift model (50) and the lever-arm model (51). Mutagenesis was used to support both hypotheses. The gearshift model proposes that the Gβ subunit rotates into the GTP-binding domain of the Gα subunit and pushes the helical domain away from the bound nucleotide. In comparison, the lever-arm model proposes that nucleotide exchange is catalyzed by induced tilt of the Gβγ subunits, which act as a lever to open the binding pocket through interactions with switch II. New insights into the mechanism of G protein activation were revealed by two recent crystal structures of activated β2 adrenergic receptor (β2AR). The first structure was that of the activated form of the receptor stabilized by llama antibodies termed nanobodies (52). The second structure was that of the activated receptor stabilized by its physiologically-relevant signaling partner, the Gαsβγ heterotrimer (Gs) (53). Both structures exhibited conformational changes in transmembrane helices 5 and 6. The β2AR-Gs structure additionally confirmed the stabilizing effects of G αβγ association on the agonist-bound state of the receptor and the presence of a translation of the a5 helix of Gα. Unanticipated observations revealed by the β2AR-Gs structure include ordering of intracellular loop 2 on β2AR, a lack of extensive contacts between the receptor and the Gβγ subunits of Gαsβγ, and a large displacement of the helical domain relative to the Ras-like GTPase domain of Gαs. These data, taken together, offer a possible mechanism for receptor-catalyzed GDP release from Gαβγ that is largely consistent with the models suggested by biochemical and structural data obtained thus far.
Large quantities of pure protein are essential for studying biophysical properties of GPCRs and heterotrimeric G proteins. Therefore, the identification of mechanisms of GPCR signaling has been facilitated by reproducible methods for obtaining and purifying rhodopsin, transducin, and Gαi. Spectroscopic monitoring is commonly used to follow the activation state in these proteins, while x-ray crystal structures have provided snapshots of stable signaling intermediates.
The metarhodopsin II assay can be used to determined the stability of light-activated rhodopsin, meta II, in the conditions in which it has been purified. Chimeric, heterotrimeric G protein (Gαiβγ) can be substituted for Gt to characterize the contribution of mutations on protein complex stability.
While multiple methods of crystallization exist (sitting drop, hanging drop, batch, etc), the hanging drop vapor diffusion method is the most commonly used (Figure 8). In this method, a closed system is designed in which a droplet of purified protein mixed with reservoir solution, usually in a 50:50 ratio, is allowed to equilibrate with a solution contained in a reservoir below. As water evaporates from the droplet hanging above the reservoir solution, the solubility of the purified protein decreases, the concentration of precipitant increases, and nucleation of crystals occurs. The system is then maintained at equilibrium until crystallization is complete.
Table 2 summarizes the conditions used to determine the structures of a subset of 33 proteins known to interact with rhodopsin. These can be used to guide your own vapor diffusion experiment using Subheading 7.
The rate of receptor-catalyzed nucleotide exchange can be measured by monitoring the increase in intrinsic tryptophan fluorescence in heterotrimeric G proteins in the presence of light activated rhodopsin. This assay can be applied to Gt or recombinant Gαi reconstituted with Gβγ (see Note 63 ).
1Bovine eyes can be obtained from your local slaughterhouse.
2All solutions that include volatile reagents (DTT or βME) or protease inhibitors with short half-lives (PMSF) should be prepared fresh each time the solution is made.
3PMSF rapidly degrades in aqueous solutions where its half-life is approximately 110 minutes (56). Prepare stock solutions of this protease inhibitor in organic solvents such as acetone or isopropanol.
4The pH of all buffers containing Tris-Cl listed in protocols throughout this chapter should be adjusted with HCl.
5Many detergents are available in two grades. SOL-grade detergents are less pure and less expensive. These can be used for solubilization of membrane proteins but may be too heterogeneous to support crystallization. ANAGRADE detergents are of higher purity and should be used for techniques requiring homogenous solutions, like x-ray crystallography.
6All buffers used for chromatographic techniques should be filtered to remove large particles and dust. Buffers containing detergents should be filtered in sufficiently large bottles (use a 1 L bottle for filtering 500 mL of buffer) so that the bubbles generated during filtering will not be suctioned into the vacuum pump. If you are using Millipore Centricon® filters, add DDM after filtering. DDM will concentrate in Millipore Centricon® filters, which can affect the final concentration of detergent in your buffer.
7It is important to use high quality, siliconized or plastic cover slips versus non-siliconized glass cover slips. While more expensive, the hydrophobic surface of siliconized cover slips promotes protein droplets more conducive to crystallization.
8GTP can be hydrolyzed in aqueous solutions. Prepare this buffer right before use and add GTP from a fresh stock.
9Ampicillin is usually prepared as a 1000X (100 mg/mL) stock solution that can withstand long-term storage at −20°C. The stock solution can be made in either a water solution or a 70% ethanol solution. The latter will not freeze at −20°C, thus decreasing the wait time for thawing the reagent during experiments.
10The rhodopsin-Gt complex is light sensitive. Protect the gel filtration column from ambient light by performing gel filtration experiment under dim red light or by wrapping column in aluminum foil.
11The half-life of a rhodopsin-Gt complex in the presence of detergent micelles is approximately 30 minutes. Supplementing the gel filtration buffer or eluted protein fractions with all-trans retinal can improve the half-life of the complex (57).
12If the volume of the retinal slurry exceeds the volume held by the maximum number of centrifuge tubes that will fit in your rotor, leave the pellet intact, layer remaining slurry on top of the pellet, and repeat the centrifugation step until all of the retinal slurry is centrifuged as described. Pool the fresh supernatant with supernatant from the previous round of centrifugation.
13Sucrose Solution 1 is the least dense, while Sucrose Solution 3 is the densest. To ensure that you maintain a distinct interface, be slow to dispense each sucrose solution into each of the centrifuge tubes and slow to remove the pipette tip from the centrifuge tube once the solution has been dispensed.
14The sucrose layer below the ROS membrane layer and the pellet at the bottom of the centrifuge tube contain contaminants. To avoid these contaminants, remove the top layer by pipetting slowly and without disrupting the gradient.
15Homogenize all resuspended fractions by inserting and removing the plunger into the glass homogenizer at least 10 times. This will ensure adequate separation of the ROS membranes from other retinal components.
16Extra caution is required when balancing ultracentrifuge tubes. The weight of each tube should be within 0.1 g of each other.
17The pellet in this step of the purification contains the ROS membranes. It is very soft and may decant with the supernatant. Instead of decanting, pipette the supernatant off of the pellet to control the amount of pellet loss.
18Detergent in this step of the membrane wash will extract rhodopsin from the membrane pellet and result in a loss of protein. It is important to use a solution without any detergent when washing this membrane pellet.
19The volume of Solubilization Buffer should be proportional to the initial volume of urea-washed ROS membranes used in your ConA purification so that the final concentration of DDM after receptor solubilization is less than 40 mM.
20Be careful to avoid getting air on the column, which can damage the resin and prevent the elution buffer from accessing the air pockets (thus decreasing yield). Attach the syringe to the column using a “wet-connect” technique by adding buffer to an adaptor connected to the top of the column and filling the syringe tube with a small volume of buffer once it is connected to the column before the column plug on the bottom end is removed.
21It is important to equilibrate the ConA resin for at least 2 days in a buffer containing 0.5 mM (or more) DDM before it is used for the first time (SOL-GRADE detergent is acceptable for this). Most chromatography columns contain hydrophobic binding sites that will strip the detergent micelles from protein if they are not filled. This results in aggregation of membrane proteins and large decreases in yield.
22The detergent concentration in the solubilized rhodopsin sample needs to be sufficiently decreased by dilution with ConA Binding Buffer. While the effects of high detergent concentrations on ConA affinity chromatography are unknown, it is possible that the high concentration of DDM may interfere with rhodopsin-ConA interactions.
23The rhodopsin-containing load fraction needs to be looped over the conA column during the loading step because the kinetics of binding are slow.
24ConA-purified rhodopsin samples contain the sugars α-D methylglucoside and sucrose, which can stick to concentrator filters. During concentration, wash the sides of the filter with the concentrated protein sample before refilling with unconcentrated, purified protein to minimize the amount of protein stuck to the filter.
25While the sample is being light-activated with bursts of bright light, remember to close your eyes to preserve night-vision.
26The concentration measured in this experiment is for 4 μL of protein diluted to 100 μL (1:25 dilution). Be sure to account for such dilution factors when calculating the concentration. Multiply the concentration obtained using equation 3 by 25 to obtain the actual rhodopsin concentration in your undiluted sample.
27This photoactivation technique activates only 10% of the total rhodopsin population in your sample.
28Filter all buffers and protein samples before using in the vapor diffusion experiment to remove dust particles.
29The temperature of the reagents should match that of the crystallization system. Drastic changes in temperature often interfere with crystallization.
30Be careful to avoid air bubbles during pipetting. Watching your pipette as solutions are dispensed and stopping before air is dispensed can prevent bubble formation in your protein drop. If you do introduce an air bubble into the crystallization reaction, these can be removed by touching the droplet gently with a clean, dry pipette tip.
31Crystallization can be temperature sensitive. If you do not have a dedicated crystallization incubator, place the crystallization trays into a room with low temperature variation and away from windows, as sunlight can cause dramatic local temperature changes.
32Physical vibrations, such as those from air compressors, can impair crystallization. Store crystallization trays in a room that lacks signification vibrations. Avoid disrupting trays for a least one day after having set them up.
33Because rhodopsin undergoes a conformational change upon activation, dark-adapted rhodopsin crystals must be examined under dim red light, either by using a red filter fitted to a light microscope or with an infrared camera.
34The crystallization condition for obtaining the highest quality crystals can be optimized by slightly varying the components in the condition. This includes testing alternative precipitants, alternative salts, buffer pH, temperature, concentrations of each chemical, protein concentrations, hanging drop size, and the method of crystallization used. For example, in the hanging drop vapor diffusion example given in Figure 14a, the buffer identity and concentration are held constant, but salt and precipitant concentrations are varied. Condition optimization is often performed in 24-well crystallization trays with each well containing a unique crystallization condition. Figure 14b is an example of a optimization screen.
35While the majority of protein crystals exhibit the best diffraction when they are large, the best crystals of rhodopsin were very small and needle-shaped. The diffraction from rhodopsin crystals is inconsistent, with only a small percentage of optically equivalent crystals exhibiting reasonable diffraction.
36If Gt is to be separated into GDP-bound Gαt and Gβγ, concentrate the Gt-containing supernatant to ~7–8 mL. Add MgSO4 to a final concentration of 40 mM and purify the protein sample on a Blue Sepharose gel filtration column (GE Healthcare). Collect the fractions corresponding to Gα and Gβγ separately, and analyze their purity by SDS-PAGE.
37It is preferable to use cells that have been transformed with fresh plasmid DNA (less than 3 months old) immediately before cell expression. Both glycerol stocks of transformed cells and older DNA (even stored at −20 °C) commonly result in lower protein yields for unknown reasons.
38Wrap parafilm around the bacterial plate, to keep moisture and air out, and store at 4°C. Colonies from this plate can be used for protein growths for up to 1 week. After 1 week, yields may be compromised.
39Most cell cultures of Gα proteins expressed in BL-21 E. coli will reach an OD600 of 0.3–0.4 within 2–3 hours post-inoculation when grown at room temperature, however, this is dependent on the Gα mutant and expression system used. Monitor the absorbance carefully, as late induction can affect the yield.
40Induction for longer than 18 hours decreases the yield.
41Cell pellets can be stored at −80°C. Freezing and thawing the pellets has been shown to actually improve the protein yield by contributing to cell lysis.
42When resuspending the resin, gently invert the bottle until all of the resin is in solution. Shaking can cause the resin to break apart into small pieces, called fines, which can clog fritted filters and interfere with affinity chromatography. To remove fines, re-mix the resin in the desired buffer and allow it to settle for ~5 minutes. This is enough time to allow the beads to settle, but the small particles remain in solution. Decant the resin supernatant, add fresh buffer, and re-mix by inverting. Repeat the process until the resin settles within 5 minutes and leaves behind a clear supernatant.
43Commercially available resin is usually stored in ethanol, which will precipitate the protein if it is not completely exchanged prior to use. Perform 3–5 successive washes of the resin slurry using 10 column volumes of Loading Buffer per wash or by continuous flow of Loading Buffer over resin in a gravity flow column to remove the ethanol.
44While G proteins are stable at room temperature, it is beneficial to perform all steps of purification at 4°C.
45If anion exchange is performed at room temperature, be sure to collect protein fractions on ice.
46A slow, low salt buffer gradient followed by a fast, high salt buffer gradient is needed to separate Gα proteins from contaminants by anion exchange chromatography.
47Purified Gα is best resolved on 10% polyacrylamide gels.
48Filter all buffers and protein samples prior to gel filtration chromatography as large particles and aggregates can irreversibly damage both the column and the instrument.. Small volumes of buffer or protein can be spin-filtered on a Spin-X column, while large volumes can be vacuum filtered.
49Gα subunits more stable in Tris-Cl buffers than HEPES buffers. The inclusion of a guanine nucleotide is critical for the stability of the Gα subunit, and GDP is most frequently used.
50For achieving the best resolution, use a sample volume of 500 μL or less for gel filtration purification of proteins on a 24 mL Superdex 200 10/300 GL gel filtration column.
51Gαi purified on a column equilibrated in 50 mM Tris-Cl pH 7.5 buffer containing 200 mM NaCl will elute at 15.5 mL. This is not necessarily the case for Gαi purified in other buffers. If using other buffers, the gel filtration column should be calibrated in that buffer using standards.
52The Gt heterotrimer and Gαi are both best resolved on a 10% polyacrylamide gel. Of the three Gt subunits, Gγ is the smallest (approximately 8 kDa) and is not observed on 10% acrylamide gels.
53AlF4− is highly unstable in solution for extended periods of time, and must be formed by mixing AlCl3 and NaF immediately before intrinsic tryptophan fluorescence is measured.
54In the case of limiting quantities of protein, this assay can also be performed in a 96-well plate using 5 μL of protein and 250 μL of 1X Quick Start™ Bradford dye.
55Mix the dye thoroughly by inverting the bottle multiple times before use so that all samples develop evenly.
56Use a clean pipette tip for each sample to avoid cross-contamination, as this will affect the fit of a regression line to your data.
57In order to determine the appropriate dilution of your protein of unknown concentration, try loading different dilutions of the purified sample on a gel and analyze by SDS-PAGE. Compare the intensity of each dilution to that of a protein sample of known concentration. You will want to pick a dilution that falls into the range of the protein standard concentrations.
58The Bradford dye is light sensitive. Keep the samples protected from ambient light during the incubation step.
59If substituting buffers, do not use Tris-Cl or any buffer that contains a primary amine, as these will interfere with the cross-linking reaction.
60For the reasons discussed in Note 21, it is important to equilibrate the gel filtration column in buffer containing detergent for a minimum of 48 hours prior to the first use. To save money on detergent, a fixed volume of buffer may be looped over the column.
61The given elution volumes are for rhodopsin-Gt complexes purified on a Superdex 200 10/300 GL gel filtration column equilibrated in Rho-Gt Buffer. This is not necessarily the volume at which the complex will elute if purified in different buffers or different detergents.
62Many membrane proteins will aggregate upon boiling in SDS-PAGE buffer. Room temperature incubation is sufficient.
63To reconstitute Gt using recombinant Gα and GPγ, first mix purified proteins in equimolar amounts and incubate on ice for 10 minutes prior to addition to the sample cuvette containing buffer.
64The Activation Buffer should not contain GDP. The presence of free GDP in solution will inhibit GTPγS binding.