Binding Site Characterization of AM1336, a Novel Covalent Inverse

Subscriber access provided by UNIV OF PITTSBURGH

Article

Binding Site Characterization of AM1336, a Novel Covalent Inverse Agonist at Human Cannabinoid 2 Receptor, Using Mass Spectrometric Analysis Srikrishnan Mallipeddi, Simion Kreimer, Nikolai Zvonok, V. Kiran Vemuri, Barry L. Karger, Alexander R. Ivanov, and Alexandros Makriyannis J. Proteome Res., Just Accepted Manuscript • Publication Date (Web): 04 Apr 2017 Downloaded from http://pubs.acs.org on April 4, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Binding Site Characterization of AM1336, a Novel Covalent Inverse Agonist at Human Cannabinoid 2 Receptor, Using Mass Spectrometric Analysis Srikrishnan Mallipeddi,† Simion Kreimer,‡ Nikolai Zvonok,† Kiran Vemuri,† Barry L. Karger, ‡ Alexander R. Ivanov,‡ Alexandros Makriyannis,†‡* †Center for Drug Discovery, Northeastern University, Boston and Department of Pharmaceutical Sciences, Northeastern University, Boston ‡Barnett Institute, Department of Chemistry and Chemical Biology, Northeastern University, Boston

*Please address all correspondence regarding this manuscript to: Alexandros Makriyannis Center for Drug Discovery Northeastern University 360 Huntington Avenue Boston, MA 02115 USA Phone: 617-373-4200; Fax: 617-373-7493 e-mail: [email protected]

1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT Cannabinoid 2 receptor (CB2R), a Class A G-protein coupled receptor (GPCR), is a promising drug target in a wide array of pathological conditions. Rational drug design has been hindered due to our poor understanding of the structural features involved in ligand binding. Binding of a high-affinity biarylpyrazole inverse agonist AM1336 to a library of the human CB2 receptor (hCB2R) cysteinesubstituted mutants provided indirect evidence that two cysteines in transmembrane helix-7 (H7) were critical for the covalent attachment. Here, we used proteomics analysis of the hCB2R with bound AM1336 to directly identify peptides with covalently attached ligand and applied in-silico modeling for visualization of the ligand-receptor interactions.

The hCB2R, with affinity tags (FlaghCB2His6), was produced in a baculovirus-insect cell expression system and purified as a functional receptor using immunoaffinity chromatography. Using mass spectrometry-based bottom-up proteomic analysis of the hCB2R-AM1336 we identified a peptide with AM1336 attached to the cysteine C284(7.38) in H7. The hCB2R homology model in lipid bilayer accommodated covalent attachment of AM1336 to C284(7.38), supporting both biochemical and mass spectrometric data. This work consolidates proteomics data and in-silico modeling, and integrates with our ligand-assisted protein structure (LAPS) experimental paradigm to assist in structure-based design of cannabinoid antagonist/inverse agonists.

Keywords: GPCR; Cannabinoid receptor; Mass spectrometry-based proteomics; Ligand-binding; Protein structure.

Introduction The endocannabinoid system is a ubiquitous lipid signaling system in the human body and has been identified to have critical functions in a diverse set of physiological processes including lipid and energy metabolism, appetite, immune activation, emotional reactivity and pain perception. The

2


Page 2 of 25

Page 3 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


endocannabinoid system encompasses two Class-A G-protein coupled receptors designated as cannabinoid 1 (CB1R) and cannabinoid 2 (CB2R) receptors, a family of endogenous lipid signaling molecules, including anandamide (AEA) and 2-arachidonoylglycerol (2-AG), related biosynthetic and metabolic enzymes and fatty-acid transporters.

The hCB2R is a seven-transmembrane protein that belongs to the Class A GPCR family and is encoded by the CNR2 gene in chromosome 1 (1p36.11).1 It has a modest sequence similarity of 44% overall (68% within the transmembrane regions) with the CB1R and is expressed predominantly in the periphery in the human body.1 CB2R expresses predominantly in immune cells, such as monocytes, macrophages, T-lymphocytes and B-lymphocytes,2 gastrointestinal system3 and to a lower extent in microglia in the CNS. Therefore, CB2R modulation has been proposed as a therapy for the treatment or management of a range of disease conditions including neuropathic pain4, chronic inflammatory pain,5 multiple sclerosis6, amyotrophic lateral sclerosis7, Huntington’s disease,8 stroke,9 inflammatory bowel diseases,10 liver cirrhosis,11 osteoporosis12 and cancer.13 Although, a number of promising drug candidates were reported to produce favorable results in preclinical models, CB2R ligands have fared poorly in human clinical trials. There is therefore a growing need for novel approaches for the structure-based drug design and development of CB2R ligands that are highly efficacious yet have minimal side effects.

Owing to the high flexibility and very lipophilic transmembrane domains in GPCRs, threedimensional structural analysis had proven to be difficult. However, due the advances in lipid-cubic phases crystallization techniques and introduction of structural modifications, over 30 unique, highresolution, GPCR crystal structures have been reported.14 More recently, two crystal structures of human CB1R with ligands AM653815 and taranabant16 respectively, have been reported. However, the atomic structure of hCB2R has remained elusive and calls for the use of alternative techniques to obtain direct structural information of fully intact CB2R. To address this issue, we developed a comprehensive strategy - Ligand Assisted Protein Structure (LAPS) that encompasses both experimental evidence from

3



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 25

biochemical and biophysical techniques and in-silico predictions would facilitate the design and optimization of therapeutically attractive CB2R-selective ligands with minimal off-targets. LAPS experimental approach that involves the integration of four elements - functionalized covalent ligandprobe development, point mutation-based pharmacological analysis, MS-based proteomic analysis and computer modeling of probe-receptor complex was developed and reported.16, approach,

we

have

previously

shown

that

19

Using the LAPS

(−)-7′-isothiocyanato-11-hydroxy-1′,1′-

dimethylheptylhexahydrocannabinol (AM841), a high-affinity CB2R agonist, interacts covalently with C257(6.47) of the conserved CWxP motif in transmembrane helix 6 (H6).17, 18 This approach proved to be a powerful technique for the characterization of critical receptor residues that interact with a chemically and functionally diverse set of ligands, at the level of specific amino acid residues.

Over the years, a large number of distinct classes of synthetic cannabinoid ligands,19 including classical cannabinoids (∆9-THC), non-classical cannabinoids (CP55940), aminoalkylindoles (WIN55212-A) and biarylpyrazoles (SR144528), have been developed and reported.20 Rimonabant, a CB1Rselective biarylpyrazole analogue, was approved for treatment of obesity in patients with associated risk factors, such as type 2 diabetes or dyslipidaemia but was later withdrawn from the market due to severe side effects (depression and suicidality). SR144528 (biarylpyrazole analogue) as well as AM630 (aminoalkylindole analogue) were the first nanomolar affinity CB2R selective ligands that function as inverse agonists.21, 22 Previously, we strategically functionalized the aliphatic side chain of biarylpyrazole analogue with an electrophilic isothiocyanate (-NCS) chemical group to generate novel covalently interacting AM1336 ligand. At the physiological pH of 7.4, ligands functionalized with isothiocyanate have been shown previously to selectively and covalently modify proteins at nearby nucleophilic thiol groups of cysteines. Using systematic covalent binding studies on multiple site-directed cysteine mutants of hCB2R overexpressed in HEK293 cells, we identified that the two cysteines - C284(7.38) and C288(7.42) were critical for the covalent attachment of AM1336 to the hCB2R.23 In this study, we use proteomic mass spectrometric (MS) analysis to provide direct evidence of this covalent interaction

4


Page 5 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


between AM1336 and H7 of hCB2R. To further understand and visualize the other interactions of AM1336 in the binding pocket of hCB2R, we performed receptor docking studies and molecular dynamics simulations in a lipid bilayer system.

Materials and Methods Reagents and Materials: Standard laboratory chemicals were purchased from Sigma-Aldrich. (St. Louis, MO) and Fisher Chemical (Pittsburgh, PA) if not otherwise specified. Coomassie G-250 stain, Laemmli electrophoresis sample buffer, PVDF membrane, molecular weight markers and 1D SDS-PAGE gels were from Bio-Rad (Hercules, CA). ANTI-FLAG® M2 affinity gel for FLAG-tag based protein immunoaffinity purification was purchased from Sigma-Aldrich. Trypsin Gold, MS grade, was purchased from Promega (Madison, WI). n-Dodecyl-β-d-maltoside (DDM) and 5-cyclohexyl-1-pentyl-β-d-maltoside (CYMAL5) were purchased from Anatrace (Maumee, OH). [3H]CP55940 was provided by the National Institute on Drug Abuse, National Institutes of Health (Bethesda, MD). AM1336 was synthesized at the Center for Drug Discovery, Northeastern University (Boston, MA).

Baculovirus/Insect Cell Expression of FlaghCB2his6: Sf21 insect cells were transfected with bacmid-FlaghCB2his6 DNA according to the Bac-to-Bac baculovirus expression kit’s (Invitrogen) recommended protocol. Briefly, Sf21 cells cultured in Sf-900™ II SFM media containing 2% fetal bovine serum (FBS), when in log phase (1.5-2.5 x 106 cells/ml) with greater than 95% viability, were transfected with different amounts of recombinant bacmid-FlaghCB2his6 DNA using Cellfectin® II reagent. The supernatant (viral stock, P1) were collected and stored at 4 °C, protected from light. The cell pellets were used to test for the expression of FlaghCB2his6 receptor. The cells were resuspended in lysis buffer (300 mM NaCl, 50 mM sodium phosphate, 10% glycerol, 1% dodecyl maltoside, pH 8.0) and lysed on ice by three, 33 s sonication cycles, each cycle consisting of 1 s bursts at 50 W separated by a 5 s interval. The lysate was centrifuged at 16,000x g for 15 min and

5



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

supernatant and pellets were analyzed separately using SDS-PAGE or western blot analysis as detailed below.

The P1 viral stock was filtered through a 0.22 µm Millex-GS syringe filter unit (Millipore, Billerica, MA) and used for further infection steps. Different quantities of filtered P1 viral stock was used for infecting Sf21 cells in log phase and incubated at 27 °C for 5 days (until ~80% cell lysis).

SDS-PAGE and Western Blot Analysis Samples were preincubated at room temperature for 10 min in Laemmli sample buffer containing 5% β-mercaptoethanol. Protein samples were resolved on AnykD™ Mini-PROTEAN® TGX™ precast polyacrylamide gels. Proteins were transferred to PVDF membranes, and the membranes were prepared for immunodetection following the procedures outlined in the QIAexpress Detection and Assay Handbook (QIAGEN, Valencia, CA). Membranes were incubated with a 1:10,000 dilution of Penta-His antibody horseradish peroxidase (HRP) conjugate (Sigma) or a 1:10,000 dilution of monoclonal mouse ANTI-FLAG® M2 antibody (Sigma) followed by incubation in a 1:10,000 dilution of goat anti-mouse IgG-HRP (Sigma-Aldrich). Protein bands were visualized using the ECL western blotting reagents (GE Healthcare, Piscataway, NJ). The image was captured using the Fluorchem SPTM (Alpha Innotech Santa Clara, California) system.

Membrane Preparation from Sf21 Cells Sf21 insect cell pellets (1.5 g) were suspended in 20 ml of TME [25 mM Tris base, 5 mM MgCl2, 1 mM ethylenediaminetetraacetic acid (EDTA), pH 7.4], lysed under nitrogen cavitation for 30 min at 1,000 psi on ice and the cell lysate was centrifuged at 2,000 g for 5 min. The pellet was washed three times with 20 ml of TME and centrifuged. The combined supernatants from each wash were ultracentrifuged at 100,000 g for 45 min at 4 °C. The resultant membrane pellet was used immediately or

6


Page 6 of 25

Page 7 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


stored at −80 °C. Membrane protein was quantified with a Bradford dye-binding method (Bio-Rad Laboratories).

Saturation Binding Assay with FlaghCB2his6 Membrane Preparations: Saturation binding assays were performed in a 96-well format. Membranes containing ~25 µg of proteins were added to each assay well. [3H]CP55940 was diluted in TME-BSA buffer to yield ligand concentrations ranging from 0.625 to 20 nM. Non-specific binding was assayed in the presence of 5 µM unlabeled CP-55,940. The assay was performed at 30 °C for 1 h with gentle agitation. The resultant material was transferred to Unifilter GF/B filter plates and unbound ligand was removed using a Packard Filtermate-196 Cell Harvester. Filter plates were washed four times with ice-cold wash buffer (50 mM Tris-base, 5 mM MgCl2 containing 0.5% BSA, pH 7.4). 40 µl of MicroscintTM 20 scintillation fluid was added to each well and the plates were counted using TopCount NXT™ Microplate Scintillation and Luminescence Counter. The data obtained was processed using Microsoft Excel and Prism 5. All concentration points were performed in triplicate and data points used for plotting are baseline corrected. Bmax and Kd values were calculated by nonlinear regression using Graphpad Prism version 5.03 (one site-binding analysis equation Y = Bmax× X/(Kd + X)) on a Windows platform.

For the saturation binding assays with purified FlaghCB2his6, 96-well GF/B filtration plates were pre-treated with 0.5% polyethylenimine (PEI) for 3 hours at 4 °C and washed twice with 200 µl of binding buffer (BB - 25 mM Tris base, 5 mM MgCl2, 1 mM ethylenediaminetetraacetic acid (EDTA), 0.1% BSA, pH 7.4).

Immunoaffinity Purification of FlaghCB2his6 Receptor Cell pellets (125 mg) were suspended in 1.0 ml lysis buffer (50 mM sodium phosphate, 300 mM NaCl, 10% glycerol, 1% dodecyl-β-d-maltoside, pH 7.4) and lysed on ice by two, 33 s sonication cycles, each cycle consisting of 1 second bursts at 50W separated by a 5 second interval. The lysate was

7



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

centrifuged at 16,000x g for 15 min (to remove cell debris) and the supernatant was collected. An equal volume of purification buffer (50 mM sodium phosphate, 300 mM NaCl, 10% glycerol, pH 7.4) was added to the supernatant. This mixture was then added to pre-equilibrated (in 50% lysis buffer, 50% purification buffer) 0.2 ml ANTI-FLAG® M2 affinity gel and incubated for 2 hours at 4 °C on a rotating wheel. The resin was washed twice with 1 ml of wash buffer (purification buffer with 0.2% dodecyl-β-dmaltoside) and FlaghCB2his6 was eluted three times with 0.2 ml of wash buffer containing 150 µg/ml of FLAG® peptide. SDS-PAGE and western blotting analysis were performed on the aliquot of samples taken during purification according to the procedure previously described.

Covalent Attachment of AM1336 to Purified FlaghCB2his6 To 30 µl (~3 µg, 2.4 µM) of purified FlaghCB2his6 receptor 1 µl of 100 µM AM1336 (final concentration 3.3 M; ~1.3-fold excess) was added and incubated for 1 hour at room temperature. The excess of AM1336 was inactivated with DDT during reduction step.

Samples Preparation of FlaghCB2his6 for MS Analysis The samples for LC-MS/MS analysis were prepared using the previously reported standard procedures.17 Briefly, 20 µl (~2 µg, 1.6 µM) of untreated or AM1336 treated purified FlaghCB2his6 receptor was reduced using 20 mM dithiothreitol (DTT) and alkylated using 50 mM iodoacetamide (IAM) by incubating with each at RT for 1 h. The mixture was then desalted into 25 mM ammonium bicarbonate, pH 8.0 with 0.05% CYMAL5 using Micro BioSpin 6 columns (Bio-Rad) according to the manufacturer’s recommended protocol. The sample was then subjected to overnight digestion with Trypsin Gold, MS-grade (Promega, Madison, WI) at a ratio of 1:20 (enzyme to substrate) at 37 °C. The digests were analyzed immediately or stored at -80 °C until further processing.

8


Page 8 of 25

Page 9 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


LC-MS/MS Analysis of FlaghCB2his6 and FlaghCB2his6-AM1336 Trypsin Digested Samples The peptides from trypsin digested FlaghCB2his6 and FlaghCB2his6-AM1336 samples were analyzed by LC-MS/MS (NanoLC Ultra2D, Eksigent) coupled to a LTQ Orbitrap XL with ETD (Thermo Scientific). Briefly, 5 µl of FlaghCB2his6 or FlaghCB2his6-AM1336 tryptic digest was loaded onto a 50 micron ID in-house polymerized polystyrene-divinyl benzene C10 derivitized monolithic (10 cm) trap column with a flow rate of 600 nl/min using a loading solvent (mobile phase A). The mobile Phase A consisted of 0.1% formic acid in 5% acetonitrile and mobile phase B consisted of 0.1% formic acid in 95% acetonitrile. Peptide separation was performed on a 40 cm long, 50 µm ID, in-house packed column with Magic C18 3 µm beads and 150 angstrom pores, using a linear gradient from 100% A: 0% B to 5% A: 95% B over 60 min at a flow rate of 100 nl/min. Electrospray ionization was carried out using a nanoESI source (New Objective) and a distal coated 20 µm ID capillary pulled to a 5 µm tip (New Objective, MA) and butt-to-butt connected to the outlet of the separation. The peptide spectra were obtained in positive-ion mode and analyzed using the PEAKS software (Bioinformatics Solutions, Waterloo, ON, Canada). Additional searches were performed using Proteome Discoverer 1.4 software (Thermo Scientific).

Targeted Identification of AM1336-H7 Peptide Using LC-MS/MS Analysis of FlaghCB2his6 and FlaghCB2his6-AM1336 Trypsin Digested Samples Due to poor solubility and possibly low ionization efficiency of the highly hydrophobic peptides from transmembrane helices, particularly with attached AM1336 ligand, targeted LC-MS/MS analysis was used for peptide profiling. TFE was added to 30% and 1 µL of sample was injected directly on to a 50 µm x 20 cm column packed in house with C8 3.5 µm ID beads (Waters, MA). The sample was desalted for 10 minutes in 2% acetonitrile and acetonitrile was ramped to 40% over 10 minutes, followed by a 30 minute linear gradient to 95% acetonitrile. The acetonitrile was held for 20 minutes and returned to 2% in 1 minute and the column was equilibrated for 19 minutes. The gradient was delivered at 200 nl/min flow rate. The sample was sprayed using a distal coated 20 µm I.D, 10 µm opening tip (New

9



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 25

Objective). Alternating targeted single ion monitoring and parallel reaction monitoring (t-SIM PRM) with a 2 m/z wide window centered around 1548.28 m/z were acquired on a QExactive (Thermo Fisher Scientific). All scans were acquired at 70,000 resolution (at 400 Th) with AGC set to 3e6 and 240 ms maximum injection time, combining 4 microscans to generate each spectrum. The presented spectra represent several microscans averaged over the entire LC-MS/MS peak eluting at 52.55-52.89 minutes.

In-silico Modeling Studies with the hCB2R The hCB2R homology model, created based on the mutation-stabilized Turkey beta-1 adrenergic receptor (PDB: 2y00) structure bound by partial agonist dobutamine, was obtained from the GPCRDB. The homology model was energy minimized by the steepest-descent and conjugate-gradient algorithms and then subjected to refinement (relaxation) by all-atom molecular dynamics (MD) simulation using AMBER03 force field in phosphatidyl-ethanolamine (PEA) lipid bilayer using YASARA software. The ligands for docking were energy minimized and cleaned for right bond orders and bond lengths. The ligands were docked into 10 nm2 cell covering the predicted GPCR binding pocket in the energy minimized hCB2R model using Autodock VINA algorithm. The poses with the highest binding energies were further analyzed. The docked AM1336 was covalently attached to either of the helix 7 cysteines and energy minimized using YASARA force field. The covalently docked AM1336-hCB2R model was subjected to MD simulation for >2 ns in phosphatidyl-ethanolamine (PEA) lipid bilayer as described previously. Additionally, the Calpha atoms of the transmembrane domains were fixed, while the residues - Phe87, Phe91, Phe94, Asp101, Phe106, Lys109, Ile110, Val113, Phe117, Glu181, Leu182, Phe183, Trp194, Phe197, Trp258, Val261, Met265, Phe281 and Ser285 and all residue side chains were allowed to be flexible. The 2D protein-ligand interaction diagrams were generated using the LIGPLOT+ v1.4 as previously reported.24

Results and Discussion Expression and Purification of FlaghCB2his6 Receptor

10


Page 11 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


To generate the functional receptor in sufficient quantities for MS-based bottom-up proteomics studies, the hCB2R was overexpressed in Spodoptera frugiperda (Sf21) cells using the Bac-to-Bac baculovirus expression system (Invitrogen). Consistent with previously published work, a full-length hCB2R construct encoding a N-terminal FLAG-tag and a C-terminal 6His-tag was chosen for recombinant receptor expression. All steps of the receptor expression following FlaghCB2his6 bacmid transfection into Sf21 cells and subsequent virus amplification were monitored using anti-His western blot analysis. Based on prior empirical detergent screen, FlaghCB2his6 was solubilized in buffer containing mild, non-ionic, MS-compatible detergent DDM, previously shown to be suitable for efficient extraction of functional GPCRs. Under optimized expression conditions, more than 70-80% of the FlaghCB2his6 receptor was extracted with detergent into solution and separated from insoluble fractions (inclusion bodies) using high-speed centrifugation (Figure 1A). Direct solubilization of FlaghCB2his6 from the intact Sf21 cells, rather than from membrane preparation, resulted in reduced receptor losses and eliminated time-consuming procedure of cell membrane preparation. FlaghCB2his6 was purified by immunoaffinity chromatography on ANTI-FLAG M2 affinity resin and analyzed using anti-His western blotting and coomassie staining of SDS-PAGE gel (Figure 1A, B). In agreement with prior results, the FlaghCB2his6 in eluate was estimated to be >80% pure. Upon storage at -80 °C for >24 hours, along with receptor monomer, dimer and higher order oligomeric states were observed, indicating time-dependent protein instability in detergent, as observed in the case of other GPCRs.

Ligand Binding Competency of FlaghCB2his6 To ascertain the functional activity of the expressed recombinant protein, saturation radioligand binding experiment using [3H]CP55940, was performed with the FlaghCB2his6-Sf21 membrane preparation. The assay revealed the level of FlaghCB2his6 receptor to be 0.5 pmol/mg (Bmax) of total protein in membrane preparation (Figure 2A). The Kd of 0.62 nM also correlated well with the previously reported25 values (0.7-3.7 nM) for the wild-type CB2R preparations from native and overexpression systems, indicating the proper folding of FlaghCB2his6 receptor.

11



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 25

Figure 1. Analysis of FlaghCB2his6 purification. (A) Anti-His western blot analysis: 2. Detergentsolubilized proteins from Sf21 cells expressing FlaghCB2his6; 3. Unbound to FLAG M2 affinity resin proteins; 4-5. FLAG M2 affinity resin washes; 6-8. FlaghCB2his6 receptor eluate 1, 3 and 5. (B) Coomassie-stained SDS-PAGE: 10. FlaghCB2his6 receptor eluate 1. 1, 9 - the protein ladder.

The ligand binding competency of the purified FlaghCB2his6 receptor was ascertained using a modified radioligand binding experiment where the filter plates were pre-treated with polyethelenimine to help retain solubilized receptor. The assay showed the presence of 20.3 pmol/mg of purified FlaghCB2his6 in elution buffer (Figure 2B). This confirmed that the detergent-solubilized, purified, FlaghCB2his6 receptor remained functionally active and can be used for further characterization studies. Further, when purified receptor was tested in saturation radioligand binding assay in the presence of 50 nM AM1336 (~10 times the Ki value obtained from CP55940 radioligand displacement assay using hCB2R membrane preparations from mammalian HEK293 cells) the Bmax decreased to 5.4 pmol/mg (Figure 2b). This decrease in Bmax indicates 73% receptor labeling with AM1336 and is consistent with

12


Page 13 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the previously reported data for hCB2R labeling by ligand. This experiment confirmed that purified FlaghCB2his6 receptor could be efficiently labeled with AM1336 for MS-based proteomics analysis.

Figure 2: Saturation binding of [3H]CP55940 ligand to FlaghCB2his6-Sf21 membrane preparations (A) and purified FlaghCB2his6 in DDM detergent micelles (B) in the presence of 50 nM AM1336 or DMSO. The experimental conditions for radioligand binding assay are detailed in the methods section.

Bottom-up Proteomic Analysis of FlaghCB2his6 and FlaghCB2his6-AM1336 Using LC-MS We used previously optimized methods of hCB2R sample preparation for proteomic analysis.17 The FlaghCB2his6 and FlaghCB2his6-AM1336 samples were reduced and alkylated using dithiothreitol (DTT) and iodoacetamide (IAM), respectively, desalted on Micro Biospin columns into 25 mM ammonium bicarbonate, pH 8.0 containing 0.05% CYMAL5 and digested overnight with trypsin. Although the desalting columns led to some protein losses, we found that they efficiently removed salts and excess of detergent. The MS-compatible detergent, CYMAL5 was added to exchange buffer to prevent protein aggregation and precipitation. The MS analysis of this sample, performed using the above detailed LC-MS procedure, yielded a protein sequence identity of ~90% to full-length wild-type hCB2R (Figure 3). The H7 peptides with different modifications including carbamidomethylation of the

13



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 25

cysteines, deamidation of the asparagines and oxidation of the methionines, with the exception of AM1336 attachment were identified in FlaghCB2his6 and FlaghCB2his6-AM1336 digested samples. We assumed that the presence of the covalently attached ligand in the resulting highly hydrophobic peptides may significantly decrease their solubility and hence, concentration in aqueous solvent systems, which can make them undetectable in conventional non-targeted proteomic profiling analyses.

Figure 3: Bottom-up proteomic analysis of FlaghCB2his6 receptor tryptic digest and peptide coverage using an LTQ Orbitrap MS. The peptides identified using both high mass accuracy MS1 measurements and MS/MS analysis are indicated in blue and using only high mass accuracy MS1 measurements are in

14


Page 15 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


grey, respectively. The protein modifications – carbamidomethylation (+57.02) (c), deamidation (+0.98) (d) and oxidation (+15.99) (o) are also indicated.

Targeted AM1336-H7 Peptides Analysis of Trypsin Digested FlaghCB2his6 and FlaghCB2his6AM1336 Using LC-MS/MS Compounds containing isothiocyanate functional group reacted at neutral pH exclusively with cysteine sulfhydryl group through nucleophilic addition. Indirect evidence from the mutation-based ligand binding studies indicated that both cysteines C284(7.38) and C288(7.42) in H7 of the hCB2R are critical for the covalent interaction with AM1336.16 To define directly the amino acid residues in receptor that modified with AM1336 we used LTQ Orbitrap MS/MS for analysis of the FlaghCB2his6 receptor tryptic peptides following treatment with DMSO or AM1336. Since the hydrophobic peptides in receptor tryptic digests are prone to precipitatation, up to 30% of trifluoroethanol was added to the samples before injection to enhance solubilization of the preparation. NanoLC separations were performed on a shorter (20 versus 40 cm), less hydrophobic (C8 versus C18) column with a two-step acetonitrile gradient (240%, 10 min; 40-95%, 30 min versus 5-95%, 60 min). Due to poor separation of the highly hydrophobic peptides from transmembrane helices under the tested conditions, particularly peptides with AM1336 attached, targeted analysis was used to identify ligand modified peptides. The high-resolution MS spectra of H7 (AFAFCSMLCLINSMVNPVIYALR) tryptic peptide in DMSO-treated FlaghCB2his6 receptor sample are shown in Figure 4A. The monoisotopic mass of the 3+ charge peak corresponding to the naïve H7 peptide where both cysteines are carbamidomethylated, was observed at m/z 897.449 Th (observed and predicted monoisotopic masses are 2689.349 and 2689.325 Da, mass error of 8.8 ppm). In AM1336treated samples, the monoisotopic mass of the 3+ charge peak tentatively assigned to AM1336-H7 peptide was observed at m/z 1032.523 Th (Figure 4B; observed and predicted monoisotopic masses are 3094.571 and 3094.549 Da; mass error 7.2 ppm). The AM1336 modified H7 peptide was shown to have a ligand at one cysteine and the second cysteine was carbamidomethylated, while both cysteines are carbomidomethylated in peptide without ligand. The observed and calculated mass differences between

15



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 25

these two peptides are 405.222 and 405.223 Da (462.245-57.021 Da; error -3.7 ppm), respectively that strongly indicated addition of AM1336 to one of the two cysteines in H7. The signal intensity of H7 peptide modified with AM1336 (3.91x103) was considerably lower than that from the unmodified peptide (2.89x104), This is attributed to increase in the hydrophobicity of the modified peptide and possible decrease in the ionization efficiency due to the addition of lipophilic ligand, thus resulting in significant losses during sample preparation and LC-MS/MS analysis (Figure 4A-B).

A

B NL: 2.89 E4 RT: 49.38

NL: 3.91 E3 RT: 50.62

Figure 4: High-resolution LC-MS/MS spectra of the hCB2R transmembrane helix 7 peptide (AFAFCSMLCLINSMVNPVIYALR) obtained from analysis of tryptic digests of purified FlaghCB2his6 treated with DMSO (monoisotopic m/z = 897.4497 Th, z=3) (A) or purified FlaghCB2his6 treated with AM1336 (monoisotopic m/z = 1032.5238 Th, z=3) (B). The mass difference is equivalent to the mass of AM1336.

MS/MS fragmentation of triple charged ions of H7 peptide m/z 897.4497 Th (both cysteines carbamidomethylated) and m/z 1032.5238 Th (one cysteine carbamidomethylated and one modified with AM1336) was used for identification of amino acid residue with attached AM1336) (Supplementary Figure 1). The mass difference (406.7507 Da) between the y19 fragments from naïve H7 (2252.5738 Da) (y192+, m/z 1127.2869 Th) and AM1336-H7 (2659.3245 Da) (y193+, m/z 887.4415 Th) peptides indicated

16


Page 17 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the presence of ligand on either C284(7.38) or C288(7.42) (Figure 5). However, no difference in mass between b10 and b8 fragments from unlabeled H7 (m/z +273.1853 Th) or AM1336-H7 (m/z +273.0725 Th) peptides excluded attachment of ligand to C288(7.42).

NL: 1.61 E3 RT: 49.40

b152+

y9+ y8+

b7+ b5+

b222+ y192+

y4+

b132+

y7+

b10+

b192+

b11+

b8+ y3+ b4+

y5+

b182+

b4+

y12+

b9+

b14+

b13+

b15+

b16+

b152+

NL: 6.25 E1 RT: 50.61 b192+

y8+

b222+ b162+

y4+

y193+

b173+

c5+ + y10+ y11

y7+ b5+ y9+

Figure

5:

High-resolution

MS/MS

b8+

b11+ y12+

fragmentation

spectra

b10+

of

helix

7

tryptic

peptide

(AFAFCSMLCLINSMVNPVIYALR): DMSO-treated (Top); AM1336-treated hCB2R (Bottom). The most structurally informative MS/MS fragments are labeled in blue.

Further, the mass difference of m/z +273.185 Th and m/z +273.072 Th, between b8+ and b10+ fragments in both DMSO-treated and AM1336-treated H7 peptides, respectively, corresponds to the combined mass of L287(7.41) and carbamidomethylated C288(7.42). Similarly, no significant shift was observed in mass difference between b8+ and b11+ fragments in both DMSO-treated and AM1336-treated H7 peptides and it corresponded to the L287(7.41), carbaminomethylated C288(7.42) and I289(7.43). The bottom-up proteomics analysis of the hCB2R-AM1336 directly indicated that AM1336 covalently attached to the

17



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 25

C284(7.38) in transmembrane H7. These results are not fully congruent with our previous indirect radioligand binding studies in which hCB2 cysteine mutants treated with AM1336 suggested that both cysteines C284(7.38) and C288(7.42) in H7 are involved in covalent binding of AM1336 with each cysteine having approximately equal probability for ligand attachment. This discrepancy of AM1336 attachment between the data obtained from our LC-MS/MS work when compared to our mutation based data may have any of the following explanations: (1). The preparations of recombinant hCB2Rs expressed in insect and mammalian cells probably have membrane species related differences that affect the ligand binding motifs; (2). Moreover, for the mutation-based experiments AM1336 covalent attachment to hCB2R was performed with receptor incorporated in the native membrane, while in our MS studies the reaction was conducted in solubilized receptor preparations.

Molecular Modeling of AM1336 Binding to hCB2R Collective data from other GPCRs, based on biochemical and modeling studies, have proposed that full cannabinoid agonists such as CP55490 interact with transmembrane helices 3-6-7 of hCB2R and activate intracellular receptor signaling. Using the LAPS approach, we previously reported that AM841, a classical cannabinoid agonist with an isothiocyanate moiety at the end of its alkyl tail, covalently attached to C257(6.47) of the CWxP hinge motif in transmembrane helix 6.18 Similarly, previous reports predicted that in case of SR144528, a CB2R-selective inverse agonist, the amide moiety interacts with transmembrane helix 3/7 and the aromatic rings interact near transmembrane helix 3-5-6 region of the receptor.26 The X-ray crystal structure of GPCR with greater sequence identity to CB2R - turkey beta-1 adrenergic receptor (PDB 2y00) was used to create a CB2R homology model. Figure 6 represents the docking results for AM1336 ligand when flexibly docked into the predicted hCB2R binding pocket. As predicted from biochemical data, the AM1336 binds similarly to hCB2R as SR144528 with pi-pi interactions between the aromatic rings and W258(6.48) and F197(5.46) with the adamantyl moiety packed against H7 and the alkyl tail ending with isothiocyanate group pointing towards C284(7.38).

18


Page 19 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


A

B Trp258

ECL2

ECL1

Ser193

ECL3

Trp194

Val113 Val261

C15

Phe87

C22

S C16 C23

C10

N

C9 C

AM1336

N2

C18 C19

N C

I

CA C8

O

CB

N1

SG

C14

V

AM1336 C13

C24

Leu182

C11

Cys284 VI

VII

C12

CYS284

C25 O

Met265

N3

Ile110

C3

Phe281

C4 ---C7 C6 C20

----

ICL1

C2 C17

C21

Lys109

C-terminal Val277

Figure 6: In-silico modeling results of AM1336-hCB2R complex. A) AM1336 (in elemental colors) covalently attached to C284(7.38) in human CB2R following ~2 ns of energy minimization with a molecular dynamics simulation in PE lipid bilayer membrane. The helices – H6, H7, H2 and H4 of hCB2R and C7.38(284) and C7.42(288) are labeled accordingly. H1, 4 and 5 are hidden for better visualization of the binding pocket. Residues that make critical interactions with AM1336 - K109(3.28), I110(3.29) and F281(7.35) are labeled in magenta. B) Ligand binding site of hCB2R with AM1336 attached at C284(7.38) respectively as predicted by molecular modeling when depicted in 2D. The conserved residues between both models are labeled (red circles) and the site of hydrophobic interactions are labeled at each atom on the ligand structure. The hydrogen bonding interactions are shown as green dotted lines and distance is indicated in angstrom.

We then attached the docked AM1336 through the sulfhydryl group of C284(7.38) and energy minimized the AM1336-hCB2R complex in a solvated lipid bilayer membrane as described under

19



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 25

“Methods”. The final structure revealed that AM1336 showed similar hydrophobic interactions to that of SR144528 with residues I110(3.29), S193(5.42), W194(5.43), M265(6.55) and S285(7.39) (Figure 6). Additionally, the movement of the adamantyl group into the pocket between H2, H7 and ECL2 indicating that large bulky groups can be accommodated at this position without significant changes to ligand affinity. It has been suggested that S112(3.31), E181(ECL2), K109(3.28) and F197(5.46) are critical for CB2R selectivity in SR14452826, 27 but were not found to be directly interacting with AM1336, consistent with its poor CBR subtype selectivity. S285(7.39) has previously been shown to be critical for binding of classical CB agonists such as HU243 in both CB1R and CB2R.28 Interestingly, it is found to make critical interactions with both SR144528 and AM1336 in our modeling experiments. The 2D ligand plot reveals that residues F87(2.57), V113(3.32), W258(6.58), V261(6.51) and F281(7.35), consistent with previous reports, are involved in hydrophobic interactions with AM1336 and part of the binding pocket for cannabinergic ligands at hCB2R.26

Conclusion The hCB2R agonists have generated significant interest as candidates for the treatment of neuropathic and inflammatory pain and for neuroprotective properties in neurodegenerative disorders.29 By contrast, CB2R antagonists/inverse agonists have been shown promising for immunomodulation therapies30 and in the treatment of bone disorders,31 however the latter remains controversial.32 To improve and accelerate design of novel, selective and potent ligands aimed to investigate further CB2Rbased therapeutic applications, a better understanding of the receptor structure is necessary.

The past decade has been very successful for X-ray structure determination of different classes of GPCRs in their active or inactive states.14 From the active-state structures, certain critical changes have been observed that are consistent among most class A GPCRs; this includes the breakage of H3-H6 salt bridge, the outward movement of H6, the inward movement of H7 and decrease and increase in the orthosteric binding pocket and G-protein binding pocket volumes respectively.33, 34 The recent hCB1R

20


Page 21 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


crystal structures provide further insight into the specific structural features of receptors in the cannabinoid family. However, since the 3D structure of the hCB2R is not available, alternate biophysical techniques could be a useful starting point for the design of new drugs with predictable functions.

Interestingly, the pharmacological profiling of cysteine to serine or alanine substituted hCB2R mutants in mammalian cell lines indirectly revealed that both cysteines C284(7.38) and C288(7.42) in transmembrane H7 are critical for the covalent attachment of AM1336, a potent biarylpyrazole CB2R inverse agonist. In this work MS analysis of the hCB2R treated with AM1336 directly identified ligand attachment to cysteine C284(7.38) of the H7, the important component of the ligand binding motif at the hCB2R. Together with the computational modeling data, this could prove to be a useful guide to our structure-based cannabinoid drug design platform.

Using our LAPS approach we have been able to identify the direct attachment of AM841, a CB2R agonist, to C257(6.47) in H6 and AM1336, a CB2R inverse agonist, to C284(7.38) in H7, respectively. Further investigation may be required to completely understand the significance of these covalent ligands attachments on the functional profile of the CB2R. A potentially useful approach may include the design of selective hCB2R ligands with required functional profile (agonist or antagonist) by targeting specific helices of the cannabinoid receptor.

Conflict of interest: The authors declare no conflicts of interest regarding the subject of this paper.

Acknowledgements: This work has been supported by grants DA 3801 and DA 9158 from the USA National Institutes of Health, National Institute on Drug Abuse.

21



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abbreviations: ∆9-THC - ∆9 Tetrahydrocannabinol 2-AG - 2-arachidonoylglycerol AEA - Anandamide CB1R - Cannabinoid 1 receptor CB2R - Cannabinoid 2 receptor DTT - Dithiothreitol EDTA - Ethylenediaminetetraacetic acid FBS - Fetal bovine serum GPCR - G-protein coupled receptor H7 - Transmembrane helix-7 IAM - Iodoacetamide LAPS - Ligand-Assisted Protein Structure MS – Mass Spectrometry PEI - Polyethylenimine

References: 1. Munro, S., Thomas, K. L. & Abu-Shaar, M. Molecular characterization of a peripheral receptor for cannabinoids. Nature 365, 61–65 (1993). 2. Galiègue, S. et al. Expression of Central and Peripheral Cannabinoid Receptors in Human Immune Tissues and Leukocyte Subpopulations, Expression of Central and Peripheral Cannabinoid Receptors in Human Immune Tissues and Leukocyte Subpopulations. European Journal of Biochemistry, European Journal of Biochemistry 232, 232, 54, 54–61, 61 (1995). 3. Wright, K. L., Duncan, M. & Sharkey, K. A. Cannabinoid CB2 receptors in the gastrointestinal tract: a regulatory system in states of inflammation. Br J Pharmacol 153, 263–270 (2008).

22


Page 22 of 25

Page 23 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


4. Malan Jr, T. P. et al. CB2 cannabinoid receptor agonists: pain relief without psychoactive effects? Current Opinion in Pharmacology 3, 62–67 (2003). 5. Guindon, J. & Hohmann, A. G. Cannabinoid CB2 receptors: a therapeutic target for the treatment of inflammatory and neuropathic pain. British Journal of Pharmacology 153, 319–334 (2008). 6. Pertwee, R. G. Cannabinoids and multiple sclerosis. Pharmacology & Therapeutics 95, 165–174 (2002). 7. Shoemaker, J. L., Seely, K. A., Reed, R. L., Crow, J. P. & Prather, P. L. The CB2 cannabinoid agonist AM-1241 prolongs survival in a transgenic mouse model of amyotrophic lateral sclerosis when initiated at symptom onset. J. Neurochem. 101, 87–98 (2007). 8. Palazuelos, J. et al. Microglial CB2 cannabinoid receptors are neuroprotective in Huntington’s disease excitotoxicity. Brain 132, 3152–3164 (2009). 9. Zhang, M. et al. Cannabinoid CB2 receptor activation decreases cerebral infarction in a mouse focal ischemia/reperfusion model. J Cereb Blood Flow Metab 27, 1387–1396 (2007). 10. Massa, F. et al. The endogenous cannabinoid system protects against colonic inflammation. J Clin Invest 113, 1202–1209 (2004). 11. Teixeira-Clerc, F. et al. CB1 cannabinoid receptor antagonism: a new strategy for the treatment of liver fibrosis. Nat Med 12, 671–676 (2006). 12. Karsak, M. et al. Cannabinoid receptor type 2 gene is associated with human osteoporosis. Hum. Mol. Genet. 14, 3389–3396 (2005). 13. Bifulco, M. & Di Marzo, V. Targeting the endocannabinoid system in cancer therapy: a call for further research. Nat. Med. 8, 547–550 (2002). 14. Isberg, V. et al. GPCRdb: an information system for G protein-coupled receptors. Nucl. Acids Res. 44, D356–D364 (2016). 15. Hua, T. et al. Crystal Structure of the Human Cannabinoid Receptor CB1. Cell 167, 750–762.e14 (2016).

23



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

16. Shao, Z. et al. High-resolution crystal structure of the human CB1 cannabinoid receptor. Nature 540, 602–606 (2016). 17. Picone, R. P. et al. (-)-7′-Isothiocyanato-11-Hydroxy-1′,1′-Dimethylheptylhexahydrocannabinol (AM841), a High-Affinity Electrophilic Ligand, Interacts Covalently with a Cysteine in Helix Six and Activates the CB1 Cannabinoid Receptor. Mol Pharmacol 68, 1623–1635 (2005). 18. Szymanski, D. W. et al. Mass spectrometry-based proteomics of human cannabinoid receptor 2: covalent cysteine 6.47(257)-ligand interaction affording megagonist receptor activation. J. Proteome Res. 10, 4789–4798 (2011). 19. Vemuri, V. K. & Makriyannis, A. Medicinal chemistry of cannabinoids. Clin. Pharmacol. Ther. 97, 553–558 (2015). 20. Pertwee, R. G. et al. International Union of Basic and Clinical Pharmacology. LXXIX. Cannabinoid Receptors and Their Ligands: Beyond CB1 and CB2. Pharmacol Rev 62, 588–631 (2010). 21. Rinaldi-Carmona, M. et al. SR 144528, the First Potent and Selective Antagonist of the CB2 Cannabinoid Receptor. J Pharmacol Exp Ther 284, 644–650 (1998). 22. Makriyannis, A. 2012 Division of Medicinal Chemistry Award Address. Trekking the Cannabinoid Road: A Personal Perspective. J. Med. Chem. 57, 3891–3911 (2014). 23. Mercier, R. W. et al. hCB2 Ligand-Interaction Landscape: Cysteine Residues Critical to Biarylpyrazole Antagonist Binding Motif and Receptor Modulation. Chemistry & Biology 17, 1132– 1142 (2010). 24. Laskowski, R. A. & Swindells, M. B. LigPlot+: multiple ligand-protein interaction diagrams for drug discovery. J Chem Inf Model 51, 2778–2786 (2011). 25. Felder, C. C. et al. Comparison of the pharmacology and signal transduction of the human cannabinoid CB1 and CB2 receptors. Mol. Pharmacol. 48, 443–450 (1995). 26. Feng, Z. et al. Modeling, Molecular Dynamics Simulation, and Mutation Validation for Structure of Cannabinoid Receptor 2 Based on Known Crystal Structures of GPCRs. J. Chem. Inf. Model. 54, 2483–2499 (2014).

24


Page 24 of 25

Page 25 of 25

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


27. Poso, A. & Huffman, J. W. Targeting the cannabinoid CB2 receptor: modelling and structural determinants of CB2 selective ligands. Br J Pharmacol 153, 335–346 (2008). 28. Rhee, M.-H. Functional Role of Serine Residues of Transmembrane Dopamin VII in Signal Transduction of CB2 Cannabinoid Receptor. Journal of Veterinary Science 3, 185–191 29. Dhopeshwarkar, A. & Mackie, K. CB2 Cannabinoid Receptors as a Therapeutic Target—What Does the Future Hold? Mol Pharmacol 86, 430–437 (2014). 30. Lunn, C. A. et al. Biology and therapeutic potential of cannabinoid CB2 receptor inverse agonists. British Journal of Pharmacology 153, 226–239 (2008). 31. Idris, A. I. et al. Regulation of bone mass, bone loss and osteoclast activity by cannabinoid receptors. Nat Med 11, 774–779 (2005). 32. Ofek, O. et al. Peripheral cannabinoid receptor, CB2, regulates bone mass. PNAS 103, 696–701 (2006). 33. Venkatakrishnan, A. J. et al. Molecular signatures of G-protein-coupled receptors. Nature 494, 185– 194 (2013). 34. Mallipeddi, S., Janero, D. R., Zvonok, N. & Makriyannis, A. Functional selectivity at G-protein coupled receptors: Advancing cannabinoid receptors as drug targets. Biochemical Pharmacology 128, 1–11 (2017).

For TOC only

y22

High-resolu on MS/MS fragmenta on spectra of hCB2R helix 7 tryp c pep de

y19

y1

Ala Phe Ala Phe Cys Ser Met Leu Cys Leu Ile Asn Ser Met Val Asn Pro Val Ile Tyr Ala Leu Arg b8

b1

b10 b11

b22

Cys284 (AM1336)

Cys288 (No ligand)

y193+ + AM1336

b8+

b11+ b10+

25


Binding Site Characterization of AM1336, a Novel Covalent Inverse

Recommend Documents