CHAPTER THIRTEEN
Conformational Analysis of G Protein-Coupled Receptor Signaling by Hydrogen/Deuterium Exchange Mass Spectrometry Sheng Li*, Su Youn Lee†, Ka Young Chung†,1 *Department of Medicine, University of California at San Diego, San Diego, California, USA † School of Pharmacy, Sungkyunkwan University, Suwon, Republic of Korea 1 Corresponding author: e-mail address: [email protected]
Contents 1. Introduction 1.1 Structural study of GPCRs with related signaling proteins 1.2 HDX-MS for conformational studies of GPCRs with related signaling proteins 2. Experimental Procedure 2.1 Overall HDX-MS procedure 2.2 Protein sample preparation 2.3 Hydrogen/deuterium exchange procedure 2.4 Quench and denaturation 2.5 Fragmentation by pepsin 2.6 LC–MS/MS procedure 2.7 Data analysis 3. Conclusion and Perspectives Acknowledgments References
262 262 264 266 266 267 270 271 272 272 274 275 276 276
Abstract Conformational change and protein–protein interactions are two major mechanisms of membrane protein signal transduction, including G protein-coupled receptors (GPCRs). Upon agonist binding, GPCRs change conformation, resulting in interaction with downstream signaling molecules such as G proteins. To understand the precise signaling mechanism, studies have investigated the structural mechanism of GPCR signaling using X-ray crystallography, nuclear magnetic resonance (NMR), or electron paramagnetic resonance. In addition to these techniques, hydrogen/deuterium exchange mass spectrometry (HDX-MS) has recently been used in GPCR studies. HDX-MS measures the rate at which peptide amide hydrogens exchange with deuterium in the solvent. Exposed or flexible regions have higher exchange rates and excluded or ordered regions have lower exchange rates. Therefore, HDX-MS is a useful tool for studying Methods in Enzymology, Volume 557 ISSN 0076-6879 http://dx.doi.org/10.1016/bs.mie.2014.12.004
#
2015 Elsevier Inc. All rights reserved.
261
262
Sheng Li et al.
protein–protein interfaces and conformational changes after protein activation or protein–protein interactions. Although HDX-MS does not give high-resolution structures, it analyzes protein conformations that are difficult to study with X-ray crystallography or NMR. Furthermore, conformational information from HDX-MS can help in the crystallization of X-ray crystallography by suggesting highly flexible regions. Interactions between GPCRs and downstream signaling molecules are not easily analyzed by X-ray crystallography or NMR because of the large size of the GPCR-signaling molecule complexes, hydrophobicity, and flexibility of GPCRs. HDX-MS could be useful for analyzing the conformational mechanism of GPCR signaling. In this chapter, we discuss details of HDX-MS for analyzing GPCRs using the β2AR-G protein complex as a model system.
1. INTRODUCTION 1.1 Structural study of GPCRs with related signaling proteins In 2012, the Nobel Prize in Chemistry was awarded to two American scientists, Brian K. Kobilka and Robert J. Lefkowtiz, for “studies of G-proteincoupled receptors (GPCRs).” GPCRs are the largest family of membrane proteins with approximately 800 identified in the human genome. GPCRs perform vital signaling functions in vision and olfactory perception and are involved in signal transduction in the metabolic, endocrine, neuromuscular, and CNS systems. Many GPCRs are involved in various diseases, and 40% of drugs on the market targets GPCRs to treat diseases including heart failure, peptic ulcers, prostate carcinoma, hypertension, pain, and bronchial asthma (Garland, 2013; Shoichet & Kobilka, 2012). Classical drug discovery approaches targeting GPCRs select and optimize compounds that interact with specific receptors using high-throughput screening (HTS) assays. However, HTS drug discovery efforts often fail because of the high degree of homology in the orthosteric ligand-binding sites among GPCRs, which makes finding a selective drug for a specific receptor subtype difficult (Shoichet & Kobilka, 2012). An alternative approach to regulating specific GPCRs is to design drugs based on the structural information of GPCRs (Shoichet & Kobilka, 2012). Characterization of the conformation and structural dynamics of proteins is critical for understanding the molecular mechanism of proteins in normal physiological and abnormal pathological processes. All GPCRs share a common seven-transmembrane α-helical structure with an extracellular N-terminus and an intracellular C-terminus connected by three extracellular loops (ECLs) and three intracellular loops (ICLs; Katritch, Cherezov, & Stevens, 2013; Venkatakrishnan et al., 2013).
HDX-MS for GPCR Signaling Analysis
263
Upon activation by agonists, GPCRs undergo conformational changes that induce interactions between the intracellular parts of GPCRs and downstream signaling molecules such as G proteins and β-arrestins (Katritch et al., 2013; Lefkowitz, Rajagopal, & Whalen, 2006; Venkatakrishnan et al., 2013). The physiological or pathological outcome of GPCR-mediated signaling depends on the signaling molecules with which the GPCRs interact. Therefore, to design a drug targeting a GPCR, it is important to understand which structural characteristics provide the selective binding of ligands to a specific GPCR (ligand specificity) and the selective interaction of GPCRs with specific signaling molecules (signaling specificity; Garland, 2013; Shoichet & Kobilka, 2012). Therefore, an effort has been made to define the mechanism of the ligand specificity in the structural studies of various GPCRs. Other studies have tried to define the signaling specificity by analyzing the conformational mechanisms of GPCRs interacting with downstream signaling molecules such as G proteins or β-arrestins. Tremendous efforts have focused on defining high-resolution GPCR structures (Zhao & Wu, 2012). Tools and techniques have been developed for better crystallization as discussed in this book: protein engineering (Chapter 2. A novel screening approach for optimal and functional fusion of T4 lysozyme in GPCRs), improved detergents (Chapter 4. Amphiphatic agents for membrane protein study), addition of antibodies (Section II. Generation and use of antibody fragments against membrane proteins), and improved crystallization conditions (Section IV. Crystallization of membrane proteins). As a result, great progress has been made in defining X-ray crystal structures of GPCRs during the past 7 years (Flight, 2013). A few GPCR crystal structures have been determined including agonist/antagonist-bound forms, a G protein-bound form, and different subtypes (Katritch et al., 2013; Rasmussen et al., 2011). However, the precise mechanism of GPCR-signaling specificity is not fully understood because of several limitations of X-ray crystallography. Intracellular loop 3 (ICL3) of GPCRs is not well defined in the X-ray crystal structures. ICL3 is reported to be involved in the activation, internalization, desensitization, and oligomerization of GPCRs and is an important binding site for signaling molecules including G proteins, β-arrestins, G proteincoupled receptor kinase (GRK), and regulator of G protein signaling (Bernstein et al., 2004; Borroto-Escuela, Correia, et al., 2010; BorrotoEscuela, Garcia-Negredo, Garriga, Fuxe, & Ciruela, 2010; Hashimoto et al., 2008; Hu et al., 2012; Kang et al., 2005; Lee, Ptasienski, PalsRylaarsdam, Gurevich, & Hosey, 2000; Pao & Benovic, 2005; Peverelli et al., 2008; Singer-Lahat, Liu, Wess, & Felder, 1996). ICL3 is highly
264
Sheng Li et al.
variable in length and sequence, and this might contribute to the diverse signaling mechanism of GPCRs (Katritch, Cherezov, & Stevens, 2012), but ICL3 in X-ray crystal structures of GPCRs is often missing or replaced by other proteins such as T4 lysozyme (Katritch et al., 2012; Venkatakrishnan et al., 2013). Another limitation is that only one X-ray crystal structure of a GPCR-signaling molecule complex (β2Adrenoceptor-Gs complex) is available (Rasmussen et al., 2011). This makes it hard to identify the structural mechanism of signaling specificity. Moreover, dynamic conformational changes during GPCR-signaling protein complex formation and dissociation, which are not easily analyzed by X-ray crystallography, have not yet been systemically studied. Studying the sequential structural events of GPCR-signaling protein complex formation would contribute to an in-depth understanding of the structural mechanism of GPCR-signaling activation. Therefore, studying the structure of GPCRs with signaling molecules by not only X-ray crystallography but also other methods is necessary.
1.2 HDX-MS for conformational studies of GPCRs with related signaling proteins Although X-ray crystallography and NMR spectroscopy are standard techniques for obtaining high-resolution protein structures, both have limitations. A major limitation of X-ray crystallography is the crystallization process. Some proteins such as membrane proteins that have large hydrophobic regions are not suitable for crystallization. Moreover, GPCRs are dynamic, which hampers the crystallization process. Another limitation is that flexible regions are often missing in defined structures such as ECL or ICL regions of GPCRs. X-ray crystallography provides structural information of only static protein states, and the crystallization environment is often not physiological. This is particularly true when studying GPCR-signaling protein complexes, which are unstable under the extreme precipitating conditions required for protein crystallography. On the other hand, NMR spectroscopy allows analysis of protein structures in more physiological, aqueous conditions. Current NMR spectroscopy technology is, however, not suitable for determining high-resolution structures of proteins larger than 40 kDa, although NMR is useful for tracking local conformational changes in specific regions of GPCRs by labeling the regions (Liu, Horst, Katritch, Stevens, & Wuthrich, 2012; Ratnala et al., 2007; Tapaneeyakorn, Goddard, Oates, Willis, & Watts, 2011). Both X-ray crystallography and NMR spectroscopy require large amounts of proteins, and obtaining structures of heterogeneous complexes, particularly in dynamic conformational states, is challenging.
HDX-MS for GPCR Signaling Analysis
265
Hydrogen/deuterium exchange mass spectrometry (HDX-MS) measures the rates at which peptide amide hydrogens exchange with deuterium in solvent. In folded proteins, the exchange rate depends on the position of the amide hydrogen (Marcsisin & Engen, 2010; Percy, Rey, Burns, & Schriemer, 2012). Exposed or highly dynamic regions show rapid exchange rates while excluded, and rigid regions show slow exchange rates (Marcsisin & Engen, 2010; Percy et al., 2012). Step 2 in Fig. 1 illustrates this phenomenon showing that the deuterium uptake occurs earlier in the flexible and exposed regions than in the ordered and excluded regions. Thus, the exchange rate provides information on the structural properties of a folded protein and contact sites between two proteins or a protein and small molecules (Marcsisin & Engen, 2010; Percy et al., 2012).
Figure 1 Theory and overall step-by-step procedure for HDX-MS regarding GPCR analysis
266
Sheng Li et al.
Although HDX-MS does not provide high-resolution structures, there are several advantages of HDX-MS over other techniques. HDX-MS requires relatively small amounts of proteins and can analyze large protein assemblies that are not amenable to other techniques. Sample preparation can be done under more physiological conditions, making the investigation of dynamic structural changes possible. For this reason, HDX-MS was adopted to analyze GPCR-G protein or GPCR-β-arrestin structures (Chung et al., 2011; Orban et al., 2012; Shukla et al., 2014). For HDX-MS, the hydrogen/deuterium exchange is initiated by adding deuterated buffer to the target protein samples, and the reaction is stopped by adding quench buffer to the reaction mixture. Because deuterium is 1 Da heavier than hydrogen, the incorporation of deuterium is measured by mass spectrometry (MS). Hydrogen/deuterium exchange is performed under physiological conditions in deuterated buffer. The effects of protein conformational change are probed during this step. GPCRs are difficult to analyze by MS, similar to other membrane proteins (Rabilloud, 2009). However, a few studies have analyzed GPCR conformations using HDX-MS, specifically the conformation of β2-adrenergic receptor (β2AR) and rhodopsin (Chung et al., 2011; Orban et al., 2012; Shukla et al., 2014; West et al., 2011; Zhang et al., 2010). The Griffin group and the Stevens group conducted a well-planned study to analyze β2AR by optimizing the experimental conditions of the HDX-MS procedure (West et al., 2011; Zhang et al., 2010). The Kobilka group and the Palczewski group analyzed the conformation of the β2AR-Gs complex and rhodopsin-transducin complex, respectively (Chung et al., 2011; Orban et al., 2012). The Kobilka group and the Lefkowitz group analyzed the conformation of the β2AR-β-arrestin1 complex (Shukla et al., 2014). Here, we discuss detailed HDX-MS procedures and factors to consider for better HDX-MS results for GPCRs. We use the β2AR-G protein complex as a model system.
2. EXPERIMENTAL PROCEDURE 2.1 Overall HDX-MS procedure (Fig. 1) A general procedure for HDX-MS is divided into six steps (Fig. 1): (1) protein sample preparation; (2) hydrogen/deuterium exchange; (3) quenching and denaturation; (4) fragmentation by pepsin; (5) liquid chromatography and tandem mass spectrometry (LC/MS); and (6) data analysis. For
HDX-MS for GPCR Signaling Analysis
267
HDX-MS, proteins are prepared in a buffer condition that preserves the protein conformation. For example, to analyze conformational changes upon β2AR-Gs complex formation, the β2AR-Gs complex, β2AR alone, and Gs alone are prepared in buffer conditions in which the β2AR-Gs complex is stabilized. Prepared proteins are subjected to hydrogen/deuterium exchange by adding a large volume of deuterated buffer. The exchange rate is monitored after stopping hydrogen/deuterium exchange at various time points by adding quench buffer. The quench buffer not only stops the exchange reaction but also denatures the proteins, and quenched denatured proteins are fragmented by enzymatic digestion by pepsin. Fragmented peptides are separated by LC, and their molecular weight is measured by MS. Peptides from nondeuterated proteins are identified by matching experimental masses with theoretical masses of peptic fragments from the protein. Once peptides are identified, deuterium uptake levels are calculated for each peptide from the deuterated proteins using commercially available or homemade software. The first four steps of HDX-MS greatly affect data quality because of the challenge of maintaining the functional conformation of the β2AR-Gs complex and getting quality peptides from membrane proteins such as GPCRs. We will not discuss how to make the stable β2AR-Gs complex as it is outside the scope of this chapter. Instead, we will discuss how to get good-quality HDX-MS data from well-prepared protein samples, focusing on GPCRs. Detailed procedures and special cautions for GPCR analysis by HDX-MS are discussed below.
2.2 Protein sample preparation (Fig. 1, step 1) 1. Protein stocks. β2AR, Gs, or β2AR-Gs complex is solubilized in protein buffer (20 mM HEPES, pH 7.5, 100 mM NaCl, 100 μM tris(2carboxyethyl)phosphine (TCEP)) supplemented with either 0.1% n-dodecyl-b-D-maltopyranoside (DDM) or 0.02% maltose neopentyl glycol-3 (MNG-3). The protein concentration is optimized by testing proteins at various concentrations and adjusted to 70 μM. Caution 2-1: Protein stocks and all buffers for HDX-MS should be filtered because particles can block LC/MS columns and tubing. Frozen/thawed protein stocks should be filtered before hydrogen/ deuterium exchange because freezing can induce protein aggregation. Caution 2-2: Many membrane proteins including GPCRs have glycosylation in the extracellular parts of the protein. Glycosylation
268
Sheng Li et al.
increases the peptide molecular weight to a various extent and adds complexity in the data analysis. Thus, deglycosylation is recommended for GPCRs before making protein stocks. This simplifies data analysis of MS-based peptide identification and increases the likelihood of higher sequence coverage of the extracellular part of GPCRs. GPCR deglycosylation can be done with peptide-Nglycosidase F, an endoglycosidase that releases N-linked oligosaccharides. Upon deglycosylation, asparagine becomes aspartate, so asparagine should be changed to aspartate when uploading protein sequences to search engines for the identification of peptides from nondeuterated proteins using theoretical masses. Released oligosaccharides can be removed by size-exclusion chromatography or affinity-tag purification. Caution 2-3: Stock concentrations or protein amounts for HDX-MS should be optimized, depending on the sensitivity of the mass spectrometer. Caution 2-4: GPCRs are hydrophobic membrane proteins, and detergents are used to extract and solubilize membrane proteins into an aqueous buffer. For GPCR solubilization, DDM has been widely used in a number of structural analyses including X-ray crystallography and NMR studies (Chung et al., 2012; Rasmussen et al., 2007). Recently, new detergents such as MNG-3 have been developed for more efficient stabilization of GPCRs (Chapter 5; Chae, Gotfryd, et al., 2010, Chae, Rasmussen, et al., 2010; Chung et al., 2012). These new detergents have been successfully used for X-ray crystallography, NMR, and other biochemical studies of GPCRs (Chung et al., 2012, 2011; Kruse et al., 2012; Rasmussen et al., 2011; Westfield et al., 2011). Detergents, however, often cause low sequence coverage of membrane proteins because of various reasons, including reduction of the sensitivity of MS (Rabilloud, 2009). Therefore, the compatibility of detergents for HDX-MS should be tested. For the study of the β2AR-G protein complex, we compared the sequence coverage of β2AR prepared in DDM with the sequence coverage of β2AR in MNG-3. MNG-3 stabilizes the β2AR-G protein complex better than DDM, but DDM gave better sequence coverage than MNG-3 (Fig. 2A and B). Caution 2-5: When protein solutions are concentrated by a spin concentrator, detergents will also be concentrated. Increased detergent
Figure 2 Optimization of sequence coverage of β2AR. (A) Sequence coverage when β2AR is solubilized in DDM. (B) Sequence coverage when β2AR is solubilized in MNG-3. (C) Sequence coverage when β2AR is solubilized in DDM, and guanidine is added in the quench buffer. Dark blue (black color in the print version) bars, peptic fragments; TM, transmembrane region; ICL, intracellular loop region, ECL; extracellular loop region. Red (gray color in the print version) arrows, connection between cysteines.
270
Sheng Li et al.
concentrations might affect the quality of mass spectra. Therefore, attention is required when proteins are concentrated. The details about this issue are discussed in Caution 3-2.
2.3 Hydrogen/deuterium exchange procedure (Fig. 1, step 2) 1. Hydrogen/deuterium exchange buffer (D2O buffer) and nonexchange buffer (H2O buffer). An advantage of HDX-MS is that the hydrogen/deuterium exchange process can be done in the buffer in which the protein stock is prepared unless this buffer severely interferes with LC/MS analysis. Thus, D2O and H2O buffers are similar to the protein buffer described earlier except with lower detergent concentrations. D2O buffer contains 20 mM HEPES, pH 7.5, 100 mM NaCl, 100 μM TCEP with either 0.05% DDM or 0.0015% MNG in D2O. H2O buffer contains 20 mM HEPES, pH 7.5, 100 mM NaCl, 100 μM TCEP with either 0.05% DDM or 0.0015% MNG in H2O. Caution 3-1: Because pH meters give inaccurate pD, the pH of the D2O buffer is adjusted with highly concentrated HEPES stock solution (1 M HEPES, pH 7.5). The same concentrated HEPES stock solution is used to adjust the pH of the H2O buffer. Another way to measure pD using a pH meter is by adding 0.4, the pD correction factor, to the pH meter reading. pD ¼ pH meter reading + 0:4 Caution 3-2: Factors that affect protein conformation should be considered when making the D2O and H2O buffers. The detergent concentration should be above the critical micelle concentration (CMC) to keep the GPCRs solubilized. The detergent concentration in D2O or H2O buffer should be lower than in the protein buffer but above the CMC because detergents tend to have deleterious effects on LC/MS analysis such as reducing MS sensitivity. If a protein stock is concentrated in a spin concentrator, detergents can be omitted from D2O and H2O buffers to minimize the deleterious effects of detergents because spin concentrators often increase detergent concentration. If ligands are used to form a specific GPCR conformation, the ligands might need to be added to D2O and H2O buffers to maintain the specific active or inactive conformation. The amount of ligand can be adjusted
HDX-MS for GPCR Signaling Analysis
271
by considering the Kd value of the ligand and the protein dilution factor after addition of D2O or H2O buffers. 2. Hydrogen/deuterium exchange. Several automated systems such as H/D-X PAL (LEAP technologies) mix and incubate protein stocks with D2O or H2O buffers and quench buffer. Automated systems are convenient but relatively expensive for small academic laboratories. Hydrogen/deuterium exchange can also be done manually by the following procedure. Hydrogen/deuterium exchange is initiated by mixing 1.5-μl protein stock (70 μM) with 4.5-μl ice-cold D2O buffer on ice. To obtain the deuterium uptake rate, mixed samples are incubated for 10, 100, 1000, or 10,000 s on ice, and reactions are stopped with 15-μl ice-cold quench buffer. The quench step is described in detail in Section 2.4. For nondeuterated samples, 1.5-μl protein stock is mixed with 4.5-μl H2O buffer and 15-μl quench buffer. Caution 3-3: Hydrogen/deuterium exchange should be on ice in a cold room (4 °C) to minimize back exchange. Exchange rate decreases as temperature is lowered, and thus back exchange during and after quenching is reduced at lower temperatures. Pipette tips and pipettes should be cooled in the cold room before use. Caution 3-4: Exchange duration can be optimized after initial experiments. If the deuterium uptake level in highly exchangeable regions is not saturated at 10,000 s, extend the exchange duration. If most regions show quick saturation, shorten the exchange duration. Caution 3-5: The volume ratio between protein stock and D2O buffer that is described here is 1:3. Therefore, the maximum deuterium uptake should not be more than 75%. When analyzing deuterium uptake mass spectra, the dilution factor should be considered.
2.4 Quench and denaturation (Fig. 1, step 3) 1. Quench and denaturation. Hydrogen/deuterium exchange rate is dependent on temperature and pH (Marcsisin & Engen, 2010); the lower the temperature, the slower the exchange rate. Hydrogen/deuterium exchange is catalyzed by both acid and base. The exchange rate is lowest at around pH 2.5 and increases as pH increases or decreases. To slow the exchange, the temperature can be lowered to 0 °C, and the pH of the quench buffer can be adjusted to pH 2.5. The exchange reaction is stopped by addition of ice-cold quench buffer at pH 2.5.
272
Sheng Li et al.
Quenched proteins might need to be denatured for effective peptic digestion. The denaturing step should be quick to minimize possible back-exchange. The denaturing step, therefore, can be done during the quench step by adding denaturants (e.g., urea or guanidine) to the quench buffer. Most GPCRs have disulfide bonds, which should be broken by reducing agents such as dithiothreitol or TCEP for effective digestion and simplification of data analysis for peptide identification. The composition of quench buffer is described later. For β2AR-Gs analysis, 15-μl ice-cold quench buffer was added to 6-μl exchange reaction mixture (1.5-μl protein stock, 4.5 μl D2O or H2O buffer) at 10, 100, 1000, or 10,000 s. Caution 4-1: If quenched samples will not immediately be analyzed, samples can be snap frozen with dry ice or liquid nitrogen and stored at 80 °C until use. Caution 4-2: GPCRs tend to aggregate with urea or guanidine, so the effects of denaturants on sequence coverage should be tested. The addition of guanidine reduced the sequence coverage of β2AR (Fig. 2A and C) but not Gs. Therefore, guanidine was not added in the quench buffer for β2AR-Gs analysis. 2. Quench buffer. Quench buffer is composed of 0.1 M NaH2PO4, pH 2.4, 20 mM TCEP, and 16.6% glycerol. Caution 4-3: The quench buffer pH should be adjusted to a final pH of about 2.5 after addition of quench buffer to the protein reaction sample.
2.5 Fragmentation by pepsin (Fig. 1, step 4) 1. Pepsin digestion. Pepsin is used as the digestion enzyme because quenched samples should be kept at pH 2.5. Quenched samples or frozen/thawed quenched samples are immediately passed through an immobilized pepsin column (16 μl bed volume, with porcine pepsin [Sigma]) at a flow rate of 20 μl/min of 0.05% trifluoroacetic acid (TFA).
2.6 LC–MS/MS procedure (Fig. 1, step 5) 1. HDX-MS instrument setup (Fig. 3). The instrumental configuration consists of microelectrically actuated high-pressure switching valves (EMHA, from VICI Valco Instruments,
HDX-MS for GPCR Signaling Analysis
273
Figure 3 Instrumental configuration of pepsin-column-conjugated LC/MS.
Houston, TX; G1163A 2port/10 position microvalve) connected to pumps, columns, and other parts with PEEK tubing (Upchurch Scientific), in-house-made pepsin columns (pepsin or fungal protease XIII), C18 Trap and analytical columns, and an extensively modified AS3000 autosampler (Spectra Physics) for low-temperature operation. All but 10 sample vial positions are filled with powdered dry ice to keep the autosampler basin at 42 °C for 10 h. Two Shimadzu LC-10AD HPLC pumps and one Agilent 1100 series binary nanopump are operated by Chemstation software (Agilent Tech Inc., Santa Clara, CA). One LC-10AD pump delivers 0.05% TFA aqueous buffer to pass protein samples over the pepsin column and onto a C18 Trap column. The other pump delivers the same buffer to back flush the pepsin column after sample digestion. An Agilent binary pump delivers solvents for linear gradient elution from the reverse-phase HPLC column (Pump A: 0.05% TFA in water; pump B: 20% water, 80% acetonitrile, 0.01% TFA). The timing and sequence of pump and valve operations are controlled by programs run in Chemstation, with an interface provided by an external contact interface board (Agilent). To minimize—back exchange, precise temperature control of the instrument is achieved by putting the autosampler, valves, connecting plumbing, and columns in a high thermal-capacity refrigerator
274
Sheng Li et al.
maintained at 3.8 °C. Components with no contact with pure D2O are immersed in ice water. MS analysis is carried out by a Finnigan LCQ Classic mass spectrometer with ESI voltage at 4.5 kV and capillary temperature of 200 °C. 2. General operational procedure. Sets of autosampler vials containing quenched, functionally deuterated samples (stored at 80 °C) are placed in the dry ice-containing sample basin of the autosampler. Samples are kept at dry ice temperature until they are melted over 1 min by the autosampler at 4–5 °C during sample preparation. Samples are injected into the immobilized pepsin column coupled to 20AL support material from PerSeptive Biosystems. Samples are passed over the pepsin column (0.05% TFA at 20 μl/min) for 30–40 s. Digested peptides are collected on a C18 Trap column (Magic C18AQ, 3u, 200 A˚, 0.2 2 mm; Michrom BioResources Inc.) for desalting (1 min at 60 μl/min) and separated on an analytical C18 column (Magic C18AQ, 3u, 200 A˚, 0.2 50 mm) by a linear gradient of 8–48% B over 30 min, with solvent A, 0.05% TFA, and solvent B: 80% acetonitrile, 20% water, and 0.01% TFA. Eluant is directed to the mass spectrometer in positive ion mode over the m/z window 20–2000. Source parameters are optimized to minimize back exchange while maintaining ionization efficiency as spray voltage 4.5 kV and heated capillary temperature 200 °C (Walters, Ricciuti, Mayne, & Englander, 2012). Data are acquired in either MS profile mode or data-dependent MS/MS mode. A dynamic exclusion list is applied to the three most intense peaks in each MS/MS scan to allow additional data to be obtained on more peptide ions. Caution 6-1: Some GPCR ligands are not soluble in aqueous solution and aggregate, which can block LC tubing.
2.7 Data analysis (Fig. 1, step 6) HDX-MS data processing includes sequence-specific peptide identification and determination of deuterium incorporation for each peptide. 1. Sequence-specific peptide identification. To identify pepsin-generated peptides, CID-based MS/MS runs are acquired using nondeuterated control samples, and lists of candidate peptides are generated by SEQUEST software. All tentative peptides are confirmed using specialized data reduction software (DXMS Explorer, Sierra Analytics Inc., Modesto, CA).
HDX-MS for GPCR Signaling Analysis
275
Caution 7-1: Even with well-controlled sample preparation and instrumental setup, identified peptides vary from sample to sample. Combining peptide lists from different sets of nondeuterated samples increases the chance of higher sequence coverage for deuterated samples. For β2AR-Gs analysis, peptide lists are generated by combining peptide lists from three nondeuterated samples. 2. Determination of deuterium incorporation Currently, the rate-limiting step in HDX-MS studies is deuterium incorporation analysis. Software has been developed to accelerate data analysis (Pascal et al., 2012; Slysz et al., 2009; Weis, Engen, & Kass, 2006; Zhang, Zhang, & Xiao, 2012); however, calculation of the centroid mass for each identified peptide and extraction of deuterium incorporation information from the large quantity of data still relies on manual intervention, especially for low-yield peptides. In β2ARGs analysis, the centroids of isotopic envelopes of all peptides were measured using DXMS Explorer, which fully automates the centroiding procedure. Corrections for back exchange use the formula of Zhang and Smith (1993). Deuterium incorporation ¼ ðmðP ÞmðN ÞÞ=ðmðF ÞmðN ÞÞ MaxD where m(P), m(N), and m(F) are the centroid value of partially deuterated, nondeuterated, and fully deuterated peptides, respectively. MaxD is the calculated maximum deuterium incorporation for the corresponding peptides. Deuteration levels of peptides are further sublocalized using overlapping peptides.
3. CONCLUSION AND PERSPECTIVES HDX-MS has been proved to be a nice tool to study the conformation of GPCR-downstream signaling molecule complexes such as β2AR-Gs and β2AR-β-arrestin1 (Chung et al., 2011; Shukla et al., 2014). We provided detailed experimental procedures for HDX-MS for GPCRs. However, HDX-MS for studying GPCRs has room to improve. In previous studies (Chung et al., 2011; Shukla et al., 2014), information about a GPCR (e.g., β2AR) is limited because of low sequence coverage. However, conformational information about downstream signaling molecules (e.g., Gs and β-arrestin1) is rich. One of the main reasons for poor sequence coverage of GPCRs is choice of detergents as discussed in Fig. 2. β2AR-Gs study used MNG-3, and β2AR-β-arrestin1 study used lauryl maltose neopentyl glycol
276
Sheng Li et al.
for receptor solubilization because these detergents are better than DDM at stabilizing β2AR-downstream signaling molecule complexes. However, DDM is better for good sequence coverage of β2AR (Fig. 2). Therefore, new detergents that are compatible with HDX-MS and stabilize GPCRdownstream signaling molecule complexes are needed. With optimal detergents and conditions, as discussed in this chapter, HDX-MS could be a powerful tool for studying the structures of GPCR-downstream signaling molecule complexes.
ACKNOWLEDGMENTS This work was supported by the basic science research program (NRF-2012R1A1A1039220) and medical research center program (NRF-2012R1A5A2A28671860) of the National Research Foundation of Korea.
REFERENCES Bernstein, L. S., Ramineni, S., Hague, C., Cladman, W., Chidiac, P., Levey, A. I., et al. (2004). RGS2 binds directly and selectively to the M1 muscarinic acetylcholine receptor third intracellular loop to modulate Gq/11alpha signaling. Journal of Biological Chemistry, 279, 21248–21256. Borroto-Escuela, D. O., Correia, P. A., Perez Alea, M., Narvaez, M., Garriga, P., Fuxe, K., et al. (2010). Impaired M(3) muscarinic acetylcholine receptor signal transduction through blockade of binding of multiple proteins to its third intracellular loop. Cellular Physiology and Biochemistry, 25, 397–408. Borroto-Escuela, D. O., Garcia-Negredo, G., Garriga, P., Fuxe, K., & Ciruela, F. (2010). The M(5) muscarinic acetylcholine receptor third intracellular loop regulates receptor function and oligomerization. Biochimica et Biophysica Acta, 1803, 813–825. Chae, P. S., Gotfryd, K., Pacyna, J., Miercke, L. J., Rasmussen, S. G., Robbins, R. A., et al. (2010). Tandem facial amphiphiles for membrane protein stabilization. Journal of the American Chemical Society, 132, 16750–16752. Chae, P. S., Rasmussen, S. G., Rana, R. R., Gotfryd, K., Chandra, R., Goren, M. A., et al. (2010). Maltose-neopentyl glycol (MNG) amphiphiles for solubilization, stabilization and crystallization of membrane proteins. Nature Methods, 7, 1003–1008. Chung, K. Y., Kim, T. H., Manglik, A., Alvares, R., Kobilka, B. K., & Prosser, R. S. (2012). Role of detergents in conformational exchange of a G protein-coupled receptor. Journal of Biological Chemistry, 287, 36305–36311. Chung, K. Y., Rasmussen, S. G., Liu, T., Li, S., DeVree, B. T., Chae, P. S., et al. (2011). Conformational changes in the G protein Gs induced by the beta2 adrenergic receptor. Nature, 477, 611–615. Flight, M. H. (2013). Drug discovery: Structure-led design. Nature, 502, S50–S52. Garland, S. L. (2013). Are GPCRs still a source of new targets? Journal of Biomolecular Screening, 18, 947–966. Hashimoto, Y., Morisawa, K., Saito, H., Jojima, E., Yoshida, N., & Haga, T. (2008). Muscarinic M4 receptor recycling requires a motif in the third intracellular loop. Journal of Pharmacology and Experimental Therapeutics, 325, 947–953. Hu, J., Thor, D., Zhou, Y., Liu, T., Wang, Y., McMillin, S. M., et al. (2012). Structural aspects of M(3) muscarinic acetylcholine receptor dimer formation and activation. FASEB Journal, 26, 604–616.
HDX-MS for GPCR Signaling Analysis
277
Kang, H., Lee, W. K., Choi, Y. H., Vukoti, K. M., Bang, W. G., & Yu, Y. G. (2005). Molecular analysis of the interaction between the intracellular loops of the human serotonin receptor type 6 (5-HT6) and the alpha subunit of GS protein. Biochemical and Biophysical Research Communications, 329, 684–692. Katritch, V., Cherezov, V., & Stevens, R. C. (2012). Diversity and modularity of G proteincoupled receptor structures. Trends in Pharmacological Sciences, 33, 17–27. Katritch, V., Cherezov, V., & Stevens, R. C. (2013). Structure-function of the G proteincoupled receptor superfamily. Annual Review of Pharmacology and Toxicology, 53, 531–556. Kruse, A. C., Hu, J., Pan, A. C., Arlow, D. H., Rosenbaum, D. M., Rosemond, E., et al. (2012). Structure and dynamics of the M3 muscarinic acetylcholine receptor. Nature, 482, 552–556. Lee, K. B., Ptasienski, J. A., Pals-Rylaarsdam, R., Gurevich, V. V., & Hosey, M. M. (2000). Arrestin binding to the M(2) muscarinic acetylcholine receptor is precluded by an inhibitory element in the third intracellular loop of the receptor. Journal of Biological Chemistry, 275, 9284–9289. Lefkowitz, R. J., Rajagopal, K., & Whalen, E. J. (2006). New roles for beta-arrestins in cell signaling: Not just for seven-transmembrane receptors. Molecular Cell, 24, 643–652. Liu, J. J., Horst, R., Katritch, V., Stevens, R. C., & Wuthrich, K. (2012). Biased signaling pathways in beta2-adrenergic receptor characterized by 19F-NMR. Science, 335, 1106–1110. Marcsisin, S. R., & Engen, J. R. (2010). Hydrogen exchange mass spectrometry: What is it and what can it tell us? Analytical and Bioanalytical Chemistry, 397, 967–972. Orban, T., Jastrzebska, B., Gupta, S., Wang, B., Miyagi, M., Chance, M. R., et al. (2012). Conformational dynamics of activation for the pentameric complex of dimeric G protein-coupled receptor and heterotrimeric G protein. Structure, 20, 826–840. Pao, C. S., & Benovic, J. L. (2005). Structure/function analysis of alpha2A-adrenergic receptor interaction with G protein-coupled receptor kinase 2. Journal of Biological Chemistry, 280, 11052–11058. Pascal, B. D., Willis, S., Lauer, J. L., Landgraf, R. R., West, G. M., Marciano, D., et al. (2012). HDX workbench: Software for the analysis of H/D exchange MS data. Journal of the American Society for Mass Spectrometry, 23, 1512–1521. Percy, A. J., Rey, M., Burns, K. M., & Schriemer, D. C. (2012). Probing protein interactions with hydrogen/deuterium exchange and mass spectrometry—A review. Analytica Chimica Acta, 721, 7–21. Peverelli, E., Mantovani, G., Calebiro, D., Doni, A., Bondioni, S., Lania, A., et al. (2008). The third intracellular loop of the human somatostatin receptor 5 is crucial for arrestin binding and receptor internalization after somatostatin stimulation. Molecular Endocrinology, 22, 676–688. Rabilloud, T. (2009). Membrane proteins and proteomics: Love is possible, but so difficult. Electrophoresis, 30(Suppl. 1), S174–S180. Rasmussen, S. G., Choi, H. J., Rosenbaum, D. M., Kobilka, T. S., Thian, F. S., Edwards, P. C., et al. (2007). Crystal structure of the human beta2 adrenergic G-protein-coupled receptor. Nature, 450, 383–387. Rasmussen, S. G., DeVree, B. T., Zou, Y., Kruse, A. C., Chung, K. Y., Kobilka, T. S., et al. (2011). Crystal structure of the beta2 adrenergic receptor-Gs protein complex. Nature, 477, 549–555. Ratnala, V. R., Kiihne, S. R., Buda, F., Leurs, R., de Groot, H. J., & DeGrip, W. J. (2007). Solid-state NMR evidence for a protonation switch in the binding pocket of the H1 receptor upon binding of the agonist histamine. Journal of the American Chemical Society, 129, 867–872. Shoichet, B. K., & Kobilka, B. K. (2012). Structure-based drug screening for G-proteincoupled receptors. Trends in Pharmacological Sciences, 33, 268–272.
278
Sheng Li et al.
Shukla, A. K., Westfield, G. H., Xiao, K., Reis, R. I., Huang, L. Y., Tripathi-Shukla, P., et al. (2014). Visualization of arrestin recruitment by a G-protein-coupled receptor. Nature, 512(7513), 218–222. Singer-Lahat, D., Liu, J., Wess, J., & Felder, C. C. (1996). The third intracellular domain of the m3 muscarinic receptor determines coupling to calcium influx in transfected Chinese hamster ovary cells. FEBS Letters, 386, 51–54. Slysz, G. W., Baker, C. A., Bozsa, B. M., Dang, A., Percy, A. J., Bennett, M., et al. (2009). Hydra: Software for tailored processing of H/D exchange data from MS or tandem MS analyses. BMC Bioinformatics, 10, 162. Tapaneeyakorn, S., Goddard, A. D., Oates, J., Willis, C. L., & Watts, A. (2011). Solutionand solid-state NMR studies of GPCRs and their ligands. Biochimica et Biophysica Acta, 1808, 1462–1475. Venkatakrishnan, A. J., Deupi, X., Lebon, G., Tate, C. G., Schertler, G. F., & Babu, M. M. (2013). Molecular signatures of G-protein-coupled receptors. Nature, 494, 185–194. Walters, B. T., Ricciuti, A., Mayne, L., & Englander, S. W. (2012). Minimizing back exchange in the hydrogen exchange-mass spectrometry experiment. Journal of the American Society for Mass Spectrometry, 23, 2132–2139. Weis, D. D., Engen, J. R., & Kass, I. J. (2006). Semi-automated data processing of hydrogen exchange mass spectra using HX-express. Journal of the American Society for Mass Spectrometry, 17, 1700–1703. West, G. M., Chien, E. Y., Katritch, V., Gatchalian, J., Chalmers, M. J., Stevens, R. C., et al. (2011). Ligand-dependent perturbation of the conformational ensemble for the GPCR beta2 adrenergic receptor revealed by HDX. Structure, 19, 1424–1432. Westfield, G. H., Rasmussen, S. G., Su, M., Dutta, S., DeVree, B. T., Chung, K. Y., et al. (2011). Structural flexibility of the G alpha s alpha-helical domain in the beta2adrenoceptor Gs complex. Proceedings of the National Academy of Sciences of the United States of America, 108, 16086–16091. Zhang, X., Chien, E. Y., Chalmers, M. J., Pascal, B. D., Gatchalian, J., Stevens, R. C., et al. (2010). Dynamics of the beta2-adrenergic G-protein coupled receptor revealed by hydrogen-deuterium exchange. Analytical Chemistry, 82, 1100–1108. Zhang, Z., & Smith, D. L. (1993). Determination of amide hydrogen exchange by mass spectrometry: A new tool for protein structure elucidation. Protein Science, 2, 522–531. Zhang, Z., Zhang, A., & Xiao, G. (2012). Improved protein hydrogen/deuterium exchange mass spectrometry platform with fully automated data processing. Analytical Chemistry, 84, 4942–4949. Zhao, Q., & Wu, B. L. (2012). Ice breaking in GPCR structural biology. Acta Pharmacologica Sinica, 33, 324–334.
Conformational Analysis of G Protein-Coupled Receptor Signaling by Hydrogen/Deuterium Exchange Mass Spectrometry Sheng Li*, Su Youn Lee†, Ka Young Chung†,1 *Department of Medicine, University of California at San Diego, San Diego, California, USA † School of Pharmacy, Sungkyunkwan University, Suwon, Republic of Korea 1 Corresponding author: e-mail address: [email protected]
Contents 1. Introduction 1.1 Structural study of GPCRs with related signaling proteins 1.2 HDX-MS for conformational studies of GPCRs with related signaling proteins 2. Experimental Procedure 2.1 Overall HDX-MS procedure 2.2 Protein sample preparation 2.3 Hydrogen/deuterium exchange procedure 2.4 Quench and denaturation 2.5 Fragmentation by pepsin 2.6 LC–MS/MS procedure 2.7 Data analysis 3. Conclusion and Perspectives Acknowledgments References
262 262 264 266 266 267 270 271 272 272 274 275 276 276
Abstract Conformational change and protein–protein interactions are two major mechanisms of membrane protein signal transduction, including G protein-coupled receptors (GPCRs). Upon agonist binding, GPCRs change conformation, resulting in interaction with downstream signaling molecules such as G proteins. To understand the precise signaling mechanism, studies have investigated the structural mechanism of GPCR signaling using X-ray crystallography, nuclear magnetic resonance (NMR), or electron paramagnetic resonance. In addition to these techniques, hydrogen/deuterium exchange mass spectrometry (HDX-MS) has recently been used in GPCR studies. HDX-MS measures the rate at which peptide amide hydrogens exchange with deuterium in the solvent. Exposed or flexible regions have higher exchange rates and excluded or ordered regions have lower exchange rates. Therefore, HDX-MS is a useful tool for studying Methods in Enzymology, Volume 557 ISSN 0076-6879 http://dx.doi.org/10.1016/bs.mie.2014.12.004
#
2015 Elsevier Inc. All rights reserved.
261
262
Sheng Li et al.
protein–protein interfaces and conformational changes after protein activation or protein–protein interactions. Although HDX-MS does not give high-resolution structures, it analyzes protein conformations that are difficult to study with X-ray crystallography or NMR. Furthermore, conformational information from HDX-MS can help in the crystallization of X-ray crystallography by suggesting highly flexible regions. Interactions between GPCRs and downstream signaling molecules are not easily analyzed by X-ray crystallography or NMR because of the large size of the GPCR-signaling molecule complexes, hydrophobicity, and flexibility of GPCRs. HDX-MS could be useful for analyzing the conformational mechanism of GPCR signaling. In this chapter, we discuss details of HDX-MS for analyzing GPCRs using the β2AR-G protein complex as a model system.
1. INTRODUCTION 1.1 Structural study of GPCRs with related signaling proteins In 2012, the Nobel Prize in Chemistry was awarded to two American scientists, Brian K. Kobilka and Robert J. Lefkowtiz, for “studies of G-proteincoupled receptors (GPCRs).” GPCRs are the largest family of membrane proteins with approximately 800 identified in the human genome. GPCRs perform vital signaling functions in vision and olfactory perception and are involved in signal transduction in the metabolic, endocrine, neuromuscular, and CNS systems. Many GPCRs are involved in various diseases, and 40% of drugs on the market targets GPCRs to treat diseases including heart failure, peptic ulcers, prostate carcinoma, hypertension, pain, and bronchial asthma (Garland, 2013; Shoichet & Kobilka, 2012). Classical drug discovery approaches targeting GPCRs select and optimize compounds that interact with specific receptors using high-throughput screening (HTS) assays. However, HTS drug discovery efforts often fail because of the high degree of homology in the orthosteric ligand-binding sites among GPCRs, which makes finding a selective drug for a specific receptor subtype difficult (Shoichet & Kobilka, 2012). An alternative approach to regulating specific GPCRs is to design drugs based on the structural information of GPCRs (Shoichet & Kobilka, 2012). Characterization of the conformation and structural dynamics of proteins is critical for understanding the molecular mechanism of proteins in normal physiological and abnormal pathological processes. All GPCRs share a common seven-transmembrane α-helical structure with an extracellular N-terminus and an intracellular C-terminus connected by three extracellular loops (ECLs) and three intracellular loops (ICLs; Katritch, Cherezov, & Stevens, 2013; Venkatakrishnan et al., 2013).
HDX-MS for GPCR Signaling Analysis
263
Upon activation by agonists, GPCRs undergo conformational changes that induce interactions between the intracellular parts of GPCRs and downstream signaling molecules such as G proteins and β-arrestins (Katritch et al., 2013; Lefkowitz, Rajagopal, & Whalen, 2006; Venkatakrishnan et al., 2013). The physiological or pathological outcome of GPCR-mediated signaling depends on the signaling molecules with which the GPCRs interact. Therefore, to design a drug targeting a GPCR, it is important to understand which structural characteristics provide the selective binding of ligands to a specific GPCR (ligand specificity) and the selective interaction of GPCRs with specific signaling molecules (signaling specificity; Garland, 2013; Shoichet & Kobilka, 2012). Therefore, an effort has been made to define the mechanism of the ligand specificity in the structural studies of various GPCRs. Other studies have tried to define the signaling specificity by analyzing the conformational mechanisms of GPCRs interacting with downstream signaling molecules such as G proteins or β-arrestins. Tremendous efforts have focused on defining high-resolution GPCR structures (Zhao & Wu, 2012). Tools and techniques have been developed for better crystallization as discussed in this book: protein engineering (Chapter 2. A novel screening approach for optimal and functional fusion of T4 lysozyme in GPCRs), improved detergents (Chapter 4. Amphiphatic agents for membrane protein study), addition of antibodies (Section II. Generation and use of antibody fragments against membrane proteins), and improved crystallization conditions (Section IV. Crystallization of membrane proteins). As a result, great progress has been made in defining X-ray crystal structures of GPCRs during the past 7 years (Flight, 2013). A few GPCR crystal structures have been determined including agonist/antagonist-bound forms, a G protein-bound form, and different subtypes (Katritch et al., 2013; Rasmussen et al., 2011). However, the precise mechanism of GPCR-signaling specificity is not fully understood because of several limitations of X-ray crystallography. Intracellular loop 3 (ICL3) of GPCRs is not well defined in the X-ray crystal structures. ICL3 is reported to be involved in the activation, internalization, desensitization, and oligomerization of GPCRs and is an important binding site for signaling molecules including G proteins, β-arrestins, G proteincoupled receptor kinase (GRK), and regulator of G protein signaling (Bernstein et al., 2004; Borroto-Escuela, Correia, et al., 2010; BorrotoEscuela, Garcia-Negredo, Garriga, Fuxe, & Ciruela, 2010; Hashimoto et al., 2008; Hu et al., 2012; Kang et al., 2005; Lee, Ptasienski, PalsRylaarsdam, Gurevich, & Hosey, 2000; Pao & Benovic, 2005; Peverelli et al., 2008; Singer-Lahat, Liu, Wess, & Felder, 1996). ICL3 is highly
264
Sheng Li et al.
variable in length and sequence, and this might contribute to the diverse signaling mechanism of GPCRs (Katritch, Cherezov, & Stevens, 2012), but ICL3 in X-ray crystal structures of GPCRs is often missing or replaced by other proteins such as T4 lysozyme (Katritch et al., 2012; Venkatakrishnan et al., 2013). Another limitation is that only one X-ray crystal structure of a GPCR-signaling molecule complex (β2Adrenoceptor-Gs complex) is available (Rasmussen et al., 2011). This makes it hard to identify the structural mechanism of signaling specificity. Moreover, dynamic conformational changes during GPCR-signaling protein complex formation and dissociation, which are not easily analyzed by X-ray crystallography, have not yet been systemically studied. Studying the sequential structural events of GPCR-signaling protein complex formation would contribute to an in-depth understanding of the structural mechanism of GPCR-signaling activation. Therefore, studying the structure of GPCRs with signaling molecules by not only X-ray crystallography but also other methods is necessary.
1.2 HDX-MS for conformational studies of GPCRs with related signaling proteins Although X-ray crystallography and NMR spectroscopy are standard techniques for obtaining high-resolution protein structures, both have limitations. A major limitation of X-ray crystallography is the crystallization process. Some proteins such as membrane proteins that have large hydrophobic regions are not suitable for crystallization. Moreover, GPCRs are dynamic, which hampers the crystallization process. Another limitation is that flexible regions are often missing in defined structures such as ECL or ICL regions of GPCRs. X-ray crystallography provides structural information of only static protein states, and the crystallization environment is often not physiological. This is particularly true when studying GPCR-signaling protein complexes, which are unstable under the extreme precipitating conditions required for protein crystallography. On the other hand, NMR spectroscopy allows analysis of protein structures in more physiological, aqueous conditions. Current NMR spectroscopy technology is, however, not suitable for determining high-resolution structures of proteins larger than 40 kDa, although NMR is useful for tracking local conformational changes in specific regions of GPCRs by labeling the regions (Liu, Horst, Katritch, Stevens, & Wuthrich, 2012; Ratnala et al., 2007; Tapaneeyakorn, Goddard, Oates, Willis, & Watts, 2011). Both X-ray crystallography and NMR spectroscopy require large amounts of proteins, and obtaining structures of heterogeneous complexes, particularly in dynamic conformational states, is challenging.
HDX-MS for GPCR Signaling Analysis
265
Hydrogen/deuterium exchange mass spectrometry (HDX-MS) measures the rates at which peptide amide hydrogens exchange with deuterium in solvent. In folded proteins, the exchange rate depends on the position of the amide hydrogen (Marcsisin & Engen, 2010; Percy, Rey, Burns, & Schriemer, 2012). Exposed or highly dynamic regions show rapid exchange rates while excluded, and rigid regions show slow exchange rates (Marcsisin & Engen, 2010; Percy et al., 2012). Step 2 in Fig. 1 illustrates this phenomenon showing that the deuterium uptake occurs earlier in the flexible and exposed regions than in the ordered and excluded regions. Thus, the exchange rate provides information on the structural properties of a folded protein and contact sites between two proteins or a protein and small molecules (Marcsisin & Engen, 2010; Percy et al., 2012).
Figure 1 Theory and overall step-by-step procedure for HDX-MS regarding GPCR analysis
266
Sheng Li et al.
Although HDX-MS does not provide high-resolution structures, there are several advantages of HDX-MS over other techniques. HDX-MS requires relatively small amounts of proteins and can analyze large protein assemblies that are not amenable to other techniques. Sample preparation can be done under more physiological conditions, making the investigation of dynamic structural changes possible. For this reason, HDX-MS was adopted to analyze GPCR-G protein or GPCR-β-arrestin structures (Chung et al., 2011; Orban et al., 2012; Shukla et al., 2014). For HDX-MS, the hydrogen/deuterium exchange is initiated by adding deuterated buffer to the target protein samples, and the reaction is stopped by adding quench buffer to the reaction mixture. Because deuterium is 1 Da heavier than hydrogen, the incorporation of deuterium is measured by mass spectrometry (MS). Hydrogen/deuterium exchange is performed under physiological conditions in deuterated buffer. The effects of protein conformational change are probed during this step. GPCRs are difficult to analyze by MS, similar to other membrane proteins (Rabilloud, 2009). However, a few studies have analyzed GPCR conformations using HDX-MS, specifically the conformation of β2-adrenergic receptor (β2AR) and rhodopsin (Chung et al., 2011; Orban et al., 2012; Shukla et al., 2014; West et al., 2011; Zhang et al., 2010). The Griffin group and the Stevens group conducted a well-planned study to analyze β2AR by optimizing the experimental conditions of the HDX-MS procedure (West et al., 2011; Zhang et al., 2010). The Kobilka group and the Palczewski group analyzed the conformation of the β2AR-Gs complex and rhodopsin-transducin complex, respectively (Chung et al., 2011; Orban et al., 2012). The Kobilka group and the Lefkowitz group analyzed the conformation of the β2AR-β-arrestin1 complex (Shukla et al., 2014). Here, we discuss detailed HDX-MS procedures and factors to consider for better HDX-MS results for GPCRs. We use the β2AR-G protein complex as a model system.
2. EXPERIMENTAL PROCEDURE 2.1 Overall HDX-MS procedure (Fig. 1) A general procedure for HDX-MS is divided into six steps (Fig. 1): (1) protein sample preparation; (2) hydrogen/deuterium exchange; (3) quenching and denaturation; (4) fragmentation by pepsin; (5) liquid chromatography and tandem mass spectrometry (LC/MS); and (6) data analysis. For
HDX-MS for GPCR Signaling Analysis
267
HDX-MS, proteins are prepared in a buffer condition that preserves the protein conformation. For example, to analyze conformational changes upon β2AR-Gs complex formation, the β2AR-Gs complex, β2AR alone, and Gs alone are prepared in buffer conditions in which the β2AR-Gs complex is stabilized. Prepared proteins are subjected to hydrogen/deuterium exchange by adding a large volume of deuterated buffer. The exchange rate is monitored after stopping hydrogen/deuterium exchange at various time points by adding quench buffer. The quench buffer not only stops the exchange reaction but also denatures the proteins, and quenched denatured proteins are fragmented by enzymatic digestion by pepsin. Fragmented peptides are separated by LC, and their molecular weight is measured by MS. Peptides from nondeuterated proteins are identified by matching experimental masses with theoretical masses of peptic fragments from the protein. Once peptides are identified, deuterium uptake levels are calculated for each peptide from the deuterated proteins using commercially available or homemade software. The first four steps of HDX-MS greatly affect data quality because of the challenge of maintaining the functional conformation of the β2AR-Gs complex and getting quality peptides from membrane proteins such as GPCRs. We will not discuss how to make the stable β2AR-Gs complex as it is outside the scope of this chapter. Instead, we will discuss how to get good-quality HDX-MS data from well-prepared protein samples, focusing on GPCRs. Detailed procedures and special cautions for GPCR analysis by HDX-MS are discussed below.
2.2 Protein sample preparation (Fig. 1, step 1) 1. Protein stocks. β2AR, Gs, or β2AR-Gs complex is solubilized in protein buffer (20 mM HEPES, pH 7.5, 100 mM NaCl, 100 μM tris(2carboxyethyl)phosphine (TCEP)) supplemented with either 0.1% n-dodecyl-b-D-maltopyranoside (DDM) or 0.02% maltose neopentyl glycol-3 (MNG-3). The protein concentration is optimized by testing proteins at various concentrations and adjusted to 70 μM. Caution 2-1: Protein stocks and all buffers for HDX-MS should be filtered because particles can block LC/MS columns and tubing. Frozen/thawed protein stocks should be filtered before hydrogen/ deuterium exchange because freezing can induce protein aggregation. Caution 2-2: Many membrane proteins including GPCRs have glycosylation in the extracellular parts of the protein. Glycosylation
268
Sheng Li et al.
increases the peptide molecular weight to a various extent and adds complexity in the data analysis. Thus, deglycosylation is recommended for GPCRs before making protein stocks. This simplifies data analysis of MS-based peptide identification and increases the likelihood of higher sequence coverage of the extracellular part of GPCRs. GPCR deglycosylation can be done with peptide-Nglycosidase F, an endoglycosidase that releases N-linked oligosaccharides. Upon deglycosylation, asparagine becomes aspartate, so asparagine should be changed to aspartate when uploading protein sequences to search engines for the identification of peptides from nondeuterated proteins using theoretical masses. Released oligosaccharides can be removed by size-exclusion chromatography or affinity-tag purification. Caution 2-3: Stock concentrations or protein amounts for HDX-MS should be optimized, depending on the sensitivity of the mass spectrometer. Caution 2-4: GPCRs are hydrophobic membrane proteins, and detergents are used to extract and solubilize membrane proteins into an aqueous buffer. For GPCR solubilization, DDM has been widely used in a number of structural analyses including X-ray crystallography and NMR studies (Chung et al., 2012; Rasmussen et al., 2007). Recently, new detergents such as MNG-3 have been developed for more efficient stabilization of GPCRs (Chapter 5; Chae, Gotfryd, et al., 2010, Chae, Rasmussen, et al., 2010; Chung et al., 2012). These new detergents have been successfully used for X-ray crystallography, NMR, and other biochemical studies of GPCRs (Chung et al., 2012, 2011; Kruse et al., 2012; Rasmussen et al., 2011; Westfield et al., 2011). Detergents, however, often cause low sequence coverage of membrane proteins because of various reasons, including reduction of the sensitivity of MS (Rabilloud, 2009). Therefore, the compatibility of detergents for HDX-MS should be tested. For the study of the β2AR-G protein complex, we compared the sequence coverage of β2AR prepared in DDM with the sequence coverage of β2AR in MNG-3. MNG-3 stabilizes the β2AR-G protein complex better than DDM, but DDM gave better sequence coverage than MNG-3 (Fig. 2A and B). Caution 2-5: When protein solutions are concentrated by a spin concentrator, detergents will also be concentrated. Increased detergent
Figure 2 Optimization of sequence coverage of β2AR. (A) Sequence coverage when β2AR is solubilized in DDM. (B) Sequence coverage when β2AR is solubilized in MNG-3. (C) Sequence coverage when β2AR is solubilized in DDM, and guanidine is added in the quench buffer. Dark blue (black color in the print version) bars, peptic fragments; TM, transmembrane region; ICL, intracellular loop region, ECL; extracellular loop region. Red (gray color in the print version) arrows, connection between cysteines.
270
Sheng Li et al.
concentrations might affect the quality of mass spectra. Therefore, attention is required when proteins are concentrated. The details about this issue are discussed in Caution 3-2.
2.3 Hydrogen/deuterium exchange procedure (Fig. 1, step 2) 1. Hydrogen/deuterium exchange buffer (D2O buffer) and nonexchange buffer (H2O buffer). An advantage of HDX-MS is that the hydrogen/deuterium exchange process can be done in the buffer in which the protein stock is prepared unless this buffer severely interferes with LC/MS analysis. Thus, D2O and H2O buffers are similar to the protein buffer described earlier except with lower detergent concentrations. D2O buffer contains 20 mM HEPES, pH 7.5, 100 mM NaCl, 100 μM TCEP with either 0.05% DDM or 0.0015% MNG in D2O. H2O buffer contains 20 mM HEPES, pH 7.5, 100 mM NaCl, 100 μM TCEP with either 0.05% DDM or 0.0015% MNG in H2O. Caution 3-1: Because pH meters give inaccurate pD, the pH of the D2O buffer is adjusted with highly concentrated HEPES stock solution (1 M HEPES, pH 7.5). The same concentrated HEPES stock solution is used to adjust the pH of the H2O buffer. Another way to measure pD using a pH meter is by adding 0.4, the pD correction factor, to the pH meter reading. pD ¼ pH meter reading + 0:4 Caution 3-2: Factors that affect protein conformation should be considered when making the D2O and H2O buffers. The detergent concentration should be above the critical micelle concentration (CMC) to keep the GPCRs solubilized. The detergent concentration in D2O or H2O buffer should be lower than in the protein buffer but above the CMC because detergents tend to have deleterious effects on LC/MS analysis such as reducing MS sensitivity. If a protein stock is concentrated in a spin concentrator, detergents can be omitted from D2O and H2O buffers to minimize the deleterious effects of detergents because spin concentrators often increase detergent concentration. If ligands are used to form a specific GPCR conformation, the ligands might need to be added to D2O and H2O buffers to maintain the specific active or inactive conformation. The amount of ligand can be adjusted
HDX-MS for GPCR Signaling Analysis
271
by considering the Kd value of the ligand and the protein dilution factor after addition of D2O or H2O buffers. 2. Hydrogen/deuterium exchange. Several automated systems such as H/D-X PAL (LEAP technologies) mix and incubate protein stocks with D2O or H2O buffers and quench buffer. Automated systems are convenient but relatively expensive for small academic laboratories. Hydrogen/deuterium exchange can also be done manually by the following procedure. Hydrogen/deuterium exchange is initiated by mixing 1.5-μl protein stock (70 μM) with 4.5-μl ice-cold D2O buffer on ice. To obtain the deuterium uptake rate, mixed samples are incubated for 10, 100, 1000, or 10,000 s on ice, and reactions are stopped with 15-μl ice-cold quench buffer. The quench step is described in detail in Section 2.4. For nondeuterated samples, 1.5-μl protein stock is mixed with 4.5-μl H2O buffer and 15-μl quench buffer. Caution 3-3: Hydrogen/deuterium exchange should be on ice in a cold room (4 °C) to minimize back exchange. Exchange rate decreases as temperature is lowered, and thus back exchange during and after quenching is reduced at lower temperatures. Pipette tips and pipettes should be cooled in the cold room before use. Caution 3-4: Exchange duration can be optimized after initial experiments. If the deuterium uptake level in highly exchangeable regions is not saturated at 10,000 s, extend the exchange duration. If most regions show quick saturation, shorten the exchange duration. Caution 3-5: The volume ratio between protein stock and D2O buffer that is described here is 1:3. Therefore, the maximum deuterium uptake should not be more than 75%. When analyzing deuterium uptake mass spectra, the dilution factor should be considered.
2.4 Quench and denaturation (Fig. 1, step 3) 1. Quench and denaturation. Hydrogen/deuterium exchange rate is dependent on temperature and pH (Marcsisin & Engen, 2010); the lower the temperature, the slower the exchange rate. Hydrogen/deuterium exchange is catalyzed by both acid and base. The exchange rate is lowest at around pH 2.5 and increases as pH increases or decreases. To slow the exchange, the temperature can be lowered to 0 °C, and the pH of the quench buffer can be adjusted to pH 2.5. The exchange reaction is stopped by addition of ice-cold quench buffer at pH 2.5.
272
Sheng Li et al.
Quenched proteins might need to be denatured for effective peptic digestion. The denaturing step should be quick to minimize possible back-exchange. The denaturing step, therefore, can be done during the quench step by adding denaturants (e.g., urea or guanidine) to the quench buffer. Most GPCRs have disulfide bonds, which should be broken by reducing agents such as dithiothreitol or TCEP for effective digestion and simplification of data analysis for peptide identification. The composition of quench buffer is described later. For β2AR-Gs analysis, 15-μl ice-cold quench buffer was added to 6-μl exchange reaction mixture (1.5-μl protein stock, 4.5 μl D2O or H2O buffer) at 10, 100, 1000, or 10,000 s. Caution 4-1: If quenched samples will not immediately be analyzed, samples can be snap frozen with dry ice or liquid nitrogen and stored at 80 °C until use. Caution 4-2: GPCRs tend to aggregate with urea or guanidine, so the effects of denaturants on sequence coverage should be tested. The addition of guanidine reduced the sequence coverage of β2AR (Fig. 2A and C) but not Gs. Therefore, guanidine was not added in the quench buffer for β2AR-Gs analysis. 2. Quench buffer. Quench buffer is composed of 0.1 M NaH2PO4, pH 2.4, 20 mM TCEP, and 16.6% glycerol. Caution 4-3: The quench buffer pH should be adjusted to a final pH of about 2.5 after addition of quench buffer to the protein reaction sample.
2.5 Fragmentation by pepsin (Fig. 1, step 4) 1. Pepsin digestion. Pepsin is used as the digestion enzyme because quenched samples should be kept at pH 2.5. Quenched samples or frozen/thawed quenched samples are immediately passed through an immobilized pepsin column (16 μl bed volume, with porcine pepsin [Sigma]) at a flow rate of 20 μl/min of 0.05% trifluoroacetic acid (TFA).
2.6 LC–MS/MS procedure (Fig. 1, step 5) 1. HDX-MS instrument setup (Fig. 3). The instrumental configuration consists of microelectrically actuated high-pressure switching valves (EMHA, from VICI Valco Instruments,
HDX-MS for GPCR Signaling Analysis
273
Figure 3 Instrumental configuration of pepsin-column-conjugated LC/MS.
Houston, TX; G1163A 2port/10 position microvalve) connected to pumps, columns, and other parts with PEEK tubing (Upchurch Scientific), in-house-made pepsin columns (pepsin or fungal protease XIII), C18 Trap and analytical columns, and an extensively modified AS3000 autosampler (Spectra Physics) for low-temperature operation. All but 10 sample vial positions are filled with powdered dry ice to keep the autosampler basin at 42 °C for 10 h. Two Shimadzu LC-10AD HPLC pumps and one Agilent 1100 series binary nanopump are operated by Chemstation software (Agilent Tech Inc., Santa Clara, CA). One LC-10AD pump delivers 0.05% TFA aqueous buffer to pass protein samples over the pepsin column and onto a C18 Trap column. The other pump delivers the same buffer to back flush the pepsin column after sample digestion. An Agilent binary pump delivers solvents for linear gradient elution from the reverse-phase HPLC column (Pump A: 0.05% TFA in water; pump B: 20% water, 80% acetonitrile, 0.01% TFA). The timing and sequence of pump and valve operations are controlled by programs run in Chemstation, with an interface provided by an external contact interface board (Agilent). To minimize—back exchange, precise temperature control of the instrument is achieved by putting the autosampler, valves, connecting plumbing, and columns in a high thermal-capacity refrigerator
274
Sheng Li et al.
maintained at 3.8 °C. Components with no contact with pure D2O are immersed in ice water. MS analysis is carried out by a Finnigan LCQ Classic mass spectrometer with ESI voltage at 4.5 kV and capillary temperature of 200 °C. 2. General operational procedure. Sets of autosampler vials containing quenched, functionally deuterated samples (stored at 80 °C) are placed in the dry ice-containing sample basin of the autosampler. Samples are kept at dry ice temperature until they are melted over 1 min by the autosampler at 4–5 °C during sample preparation. Samples are injected into the immobilized pepsin column coupled to 20AL support material from PerSeptive Biosystems. Samples are passed over the pepsin column (0.05% TFA at 20 μl/min) for 30–40 s. Digested peptides are collected on a C18 Trap column (Magic C18AQ, 3u, 200 A˚, 0.2 2 mm; Michrom BioResources Inc.) for desalting (1 min at 60 μl/min) and separated on an analytical C18 column (Magic C18AQ, 3u, 200 A˚, 0.2 50 mm) by a linear gradient of 8–48% B over 30 min, with solvent A, 0.05% TFA, and solvent B: 80% acetonitrile, 20% water, and 0.01% TFA. Eluant is directed to the mass spectrometer in positive ion mode over the m/z window 20–2000. Source parameters are optimized to minimize back exchange while maintaining ionization efficiency as spray voltage 4.5 kV and heated capillary temperature 200 °C (Walters, Ricciuti, Mayne, & Englander, 2012). Data are acquired in either MS profile mode or data-dependent MS/MS mode. A dynamic exclusion list is applied to the three most intense peaks in each MS/MS scan to allow additional data to be obtained on more peptide ions. Caution 6-1: Some GPCR ligands are not soluble in aqueous solution and aggregate, which can block LC tubing.
2.7 Data analysis (Fig. 1, step 6) HDX-MS data processing includes sequence-specific peptide identification and determination of deuterium incorporation for each peptide. 1. Sequence-specific peptide identification. To identify pepsin-generated peptides, CID-based MS/MS runs are acquired using nondeuterated control samples, and lists of candidate peptides are generated by SEQUEST software. All tentative peptides are confirmed using specialized data reduction software (DXMS Explorer, Sierra Analytics Inc., Modesto, CA).
HDX-MS for GPCR Signaling Analysis
275
Caution 7-1: Even with well-controlled sample preparation and instrumental setup, identified peptides vary from sample to sample. Combining peptide lists from different sets of nondeuterated samples increases the chance of higher sequence coverage for deuterated samples. For β2AR-Gs analysis, peptide lists are generated by combining peptide lists from three nondeuterated samples. 2. Determination of deuterium incorporation Currently, the rate-limiting step in HDX-MS studies is deuterium incorporation analysis. Software has been developed to accelerate data analysis (Pascal et al., 2012; Slysz et al., 2009; Weis, Engen, & Kass, 2006; Zhang, Zhang, & Xiao, 2012); however, calculation of the centroid mass for each identified peptide and extraction of deuterium incorporation information from the large quantity of data still relies on manual intervention, especially for low-yield peptides. In β2ARGs analysis, the centroids of isotopic envelopes of all peptides were measured using DXMS Explorer, which fully automates the centroiding procedure. Corrections for back exchange use the formula of Zhang and Smith (1993). Deuterium incorporation ¼ ðmðP ÞmðN ÞÞ=ðmðF ÞmðN ÞÞ MaxD where m(P), m(N), and m(F) are the centroid value of partially deuterated, nondeuterated, and fully deuterated peptides, respectively. MaxD is the calculated maximum deuterium incorporation for the corresponding peptides. Deuteration levels of peptides are further sublocalized using overlapping peptides.
3. CONCLUSION AND PERSPECTIVES HDX-MS has been proved to be a nice tool to study the conformation of GPCR-downstream signaling molecule complexes such as β2AR-Gs and β2AR-β-arrestin1 (Chung et al., 2011; Shukla et al., 2014). We provided detailed experimental procedures for HDX-MS for GPCRs. However, HDX-MS for studying GPCRs has room to improve. In previous studies (Chung et al., 2011; Shukla et al., 2014), information about a GPCR (e.g., β2AR) is limited because of low sequence coverage. However, conformational information about downstream signaling molecules (e.g., Gs and β-arrestin1) is rich. One of the main reasons for poor sequence coverage of GPCRs is choice of detergents as discussed in Fig. 2. β2AR-Gs study used MNG-3, and β2AR-β-arrestin1 study used lauryl maltose neopentyl glycol
276
Sheng Li et al.
for receptor solubilization because these detergents are better than DDM at stabilizing β2AR-downstream signaling molecule complexes. However, DDM is better for good sequence coverage of β2AR (Fig. 2). Therefore, new detergents that are compatible with HDX-MS and stabilize GPCRdownstream signaling molecule complexes are needed. With optimal detergents and conditions, as discussed in this chapter, HDX-MS could be a powerful tool for studying the structures of GPCR-downstream signaling molecule complexes.
ACKNOWLEDGMENTS This work was supported by the basic science research program (NRF-2012R1A1A1039220) and medical research center program (NRF-2012R1A5A2A28671860) of the National Research Foundation of Korea.
REFERENCES Bernstein, L. S., Ramineni, S., Hague, C., Cladman, W., Chidiac, P., Levey, A. I., et al. (2004). RGS2 binds directly and selectively to the M1 muscarinic acetylcholine receptor third intracellular loop to modulate Gq/11alpha signaling. Journal of Biological Chemistry, 279, 21248–21256. Borroto-Escuela, D. O., Correia, P. A., Perez Alea, M., Narvaez, M., Garriga, P., Fuxe, K., et al. (2010). Impaired M(3) muscarinic acetylcholine receptor signal transduction through blockade of binding of multiple proteins to its third intracellular loop. Cellular Physiology and Biochemistry, 25, 397–408. Borroto-Escuela, D. O., Garcia-Negredo, G., Garriga, P., Fuxe, K., & Ciruela, F. (2010). The M(5) muscarinic acetylcholine receptor third intracellular loop regulates receptor function and oligomerization. Biochimica et Biophysica Acta, 1803, 813–825. Chae, P. S., Gotfryd, K., Pacyna, J., Miercke, L. J., Rasmussen, S. G., Robbins, R. A., et al. (2010). Tandem facial amphiphiles for membrane protein stabilization. Journal of the American Chemical Society, 132, 16750–16752. Chae, P. S., Rasmussen, S. G., Rana, R. R., Gotfryd, K., Chandra, R., Goren, M. A., et al. (2010). Maltose-neopentyl glycol (MNG) amphiphiles for solubilization, stabilization and crystallization of membrane proteins. Nature Methods, 7, 1003–1008. Chung, K. Y., Kim, T. H., Manglik, A., Alvares, R., Kobilka, B. K., & Prosser, R. S. (2012). Role of detergents in conformational exchange of a G protein-coupled receptor. Journal of Biological Chemistry, 287, 36305–36311. Chung, K. Y., Rasmussen, S. G., Liu, T., Li, S., DeVree, B. T., Chae, P. S., et al. (2011). Conformational changes in the G protein Gs induced by the beta2 adrenergic receptor. Nature, 477, 611–615. Flight, M. H. (2013). Drug discovery: Structure-led design. Nature, 502, S50–S52. Garland, S. L. (2013). Are GPCRs still a source of new targets? Journal of Biomolecular Screening, 18, 947–966. Hashimoto, Y., Morisawa, K., Saito, H., Jojima, E., Yoshida, N., & Haga, T. (2008). Muscarinic M4 receptor recycling requires a motif in the third intracellular loop. Journal of Pharmacology and Experimental Therapeutics, 325, 947–953. Hu, J., Thor, D., Zhou, Y., Liu, T., Wang, Y., McMillin, S. M., et al. (2012). Structural aspects of M(3) muscarinic acetylcholine receptor dimer formation and activation. FASEB Journal, 26, 604–616.
HDX-MS for GPCR Signaling Analysis
277
Kang, H., Lee, W. K., Choi, Y. H., Vukoti, K. M., Bang, W. G., & Yu, Y. G. (2005). Molecular analysis of the interaction between the intracellular loops of the human serotonin receptor type 6 (5-HT6) and the alpha subunit of GS protein. Biochemical and Biophysical Research Communications, 329, 684–692. Katritch, V., Cherezov, V., & Stevens, R. C. (2012). Diversity and modularity of G proteincoupled receptor structures. Trends in Pharmacological Sciences, 33, 17–27. Katritch, V., Cherezov, V., & Stevens, R. C. (2013). Structure-function of the G proteincoupled receptor superfamily. Annual Review of Pharmacology and Toxicology, 53, 531–556. Kruse, A. C., Hu, J., Pan, A. C., Arlow, D. H., Rosenbaum, D. M., Rosemond, E., et al. (2012). Structure and dynamics of the M3 muscarinic acetylcholine receptor. Nature, 482, 552–556. Lee, K. B., Ptasienski, J. A., Pals-Rylaarsdam, R., Gurevich, V. V., & Hosey, M. M. (2000). Arrestin binding to the M(2) muscarinic acetylcholine receptor is precluded by an inhibitory element in the third intracellular loop of the receptor. Journal of Biological Chemistry, 275, 9284–9289. Lefkowitz, R. J., Rajagopal, K., & Whalen, E. J. (2006). New roles for beta-arrestins in cell signaling: Not just for seven-transmembrane receptors. Molecular Cell, 24, 643–652. Liu, J. J., Horst, R., Katritch, V., Stevens, R. C., & Wuthrich, K. (2012). Biased signaling pathways in beta2-adrenergic receptor characterized by 19F-NMR. Science, 335, 1106–1110. Marcsisin, S. R., & Engen, J. R. (2010). Hydrogen exchange mass spectrometry: What is it and what can it tell us? Analytical and Bioanalytical Chemistry, 397, 967–972. Orban, T., Jastrzebska, B., Gupta, S., Wang, B., Miyagi, M., Chance, M. R., et al. (2012). Conformational dynamics of activation for the pentameric complex of dimeric G protein-coupled receptor and heterotrimeric G protein. Structure, 20, 826–840. Pao, C. S., & Benovic, J. L. (2005). Structure/function analysis of alpha2A-adrenergic receptor interaction with G protein-coupled receptor kinase 2. Journal of Biological Chemistry, 280, 11052–11058. Pascal, B. D., Willis, S., Lauer, J. L., Landgraf, R. R., West, G. M., Marciano, D., et al. (2012). HDX workbench: Software for the analysis of H/D exchange MS data. Journal of the American Society for Mass Spectrometry, 23, 1512–1521. Percy, A. J., Rey, M., Burns, K. M., & Schriemer, D. C. (2012). Probing protein interactions with hydrogen/deuterium exchange and mass spectrometry—A review. Analytica Chimica Acta, 721, 7–21. Peverelli, E., Mantovani, G., Calebiro, D., Doni, A., Bondioni, S., Lania, A., et al. (2008). The third intracellular loop of the human somatostatin receptor 5 is crucial for arrestin binding and receptor internalization after somatostatin stimulation. Molecular Endocrinology, 22, 676–688. Rabilloud, T. (2009). Membrane proteins and proteomics: Love is possible, but so difficult. Electrophoresis, 30(Suppl. 1), S174–S180. Rasmussen, S. G., Choi, H. J., Rosenbaum, D. M., Kobilka, T. S., Thian, F. S., Edwards, P. C., et al. (2007). Crystal structure of the human beta2 adrenergic G-protein-coupled receptor. Nature, 450, 383–387. Rasmussen, S. G., DeVree, B. T., Zou, Y., Kruse, A. C., Chung, K. Y., Kobilka, T. S., et al. (2011). Crystal structure of the beta2 adrenergic receptor-Gs protein complex. Nature, 477, 549–555. Ratnala, V. R., Kiihne, S. R., Buda, F., Leurs, R., de Groot, H. J., & DeGrip, W. J. (2007). Solid-state NMR evidence for a protonation switch in the binding pocket of the H1 receptor upon binding of the agonist histamine. Journal of the American Chemical Society, 129, 867–872. Shoichet, B. K., & Kobilka, B. K. (2012). Structure-based drug screening for G-proteincoupled receptors. Trends in Pharmacological Sciences, 33, 268–272.
278
Sheng Li et al.
Shukla, A. K., Westfield, G. H., Xiao, K., Reis, R. I., Huang, L. Y., Tripathi-Shukla, P., et al. (2014). Visualization of arrestin recruitment by a G-protein-coupled receptor. Nature, 512(7513), 218–222. Singer-Lahat, D., Liu, J., Wess, J., & Felder, C. C. (1996). The third intracellular domain of the m3 muscarinic receptor determines coupling to calcium influx in transfected Chinese hamster ovary cells. FEBS Letters, 386, 51–54. Slysz, G. W., Baker, C. A., Bozsa, B. M., Dang, A., Percy, A. J., Bennett, M., et al. (2009). Hydra: Software for tailored processing of H/D exchange data from MS or tandem MS analyses. BMC Bioinformatics, 10, 162. Tapaneeyakorn, S., Goddard, A. D., Oates, J., Willis, C. L., & Watts, A. (2011). Solutionand solid-state NMR studies of GPCRs and their ligands. Biochimica et Biophysica Acta, 1808, 1462–1475. Venkatakrishnan, A. J., Deupi, X., Lebon, G., Tate, C. G., Schertler, G. F., & Babu, M. M. (2013). Molecular signatures of G-protein-coupled receptors. Nature, 494, 185–194. Walters, B. T., Ricciuti, A., Mayne, L., & Englander, S. W. (2012). Minimizing back exchange in the hydrogen exchange-mass spectrometry experiment. Journal of the American Society for Mass Spectrometry, 23, 2132–2139. Weis, D. D., Engen, J. R., & Kass, I. J. (2006). Semi-automated data processing of hydrogen exchange mass spectra using HX-express. Journal of the American Society for Mass Spectrometry, 17, 1700–1703. West, G. M., Chien, E. Y., Katritch, V., Gatchalian, J., Chalmers, M. J., Stevens, R. C., et al. (2011). Ligand-dependent perturbation of the conformational ensemble for the GPCR beta2 adrenergic receptor revealed by HDX. Structure, 19, 1424–1432. Westfield, G. H., Rasmussen, S. G., Su, M., Dutta, S., DeVree, B. T., Chung, K. Y., et al. (2011). Structural flexibility of the G alpha s alpha-helical domain in the beta2adrenoceptor Gs complex. Proceedings of the National Academy of Sciences of the United States of America, 108, 16086–16091. Zhang, X., Chien, E. Y., Chalmers, M. J., Pascal, B. D., Gatchalian, J., Stevens, R. C., et al. (2010). Dynamics of the beta2-adrenergic G-protein coupled receptor revealed by hydrogen-deuterium exchange. Analytical Chemistry, 82, 1100–1108. Zhang, Z., & Smith, D. L. (1993). Determination of amide hydrogen exchange by mass spectrometry: A new tool for protein structure elucidation. Protein Science, 2, 522–531. Zhang, Z., Zhang, A., & Xiao, G. (2012). Improved protein hydrogen/deuterium exchange mass spectrometry platform with fully automated data processing. Analytical Chemistry, 84, 4942–4949. Zhao, Q., & Wu, B. L. (2012). Ice breaking in GPCR structural biology. Acta Pharmacologica Sinica, 33, 324–334.
