www.sciencemag.org/cgi/content/full/science.1202793/DC1

Supporting Online Material for Structure of an Agonist-Bound Human A2A Adenosine Receptor Fei Xu, Huixian Wu, Vsevolod Katritch, Gye Won Han, Kenneth A. Jacobson, Zhan-Guo Gao, Vadim Cherezov, Raymond C. Stevens* *To whom correspondence should be addressed. E-mail: [email protected]

Published 10 March 2011 on Science Express DOI: 10.1126/science.1202793

This PDF file includes: Materials and Methods Tables S1 to S3 Figs. S1 to S4 References Legend for Movie S1 Additional supplemental material for this manuscript includes the following: Movie S1

Materials and Methods Molecular biology for generation of Spodoptera frugiperda (Sf9) expressed A2AAR-WT, A2AAR-T4L and A2AAR-T4L-C constructs Construction of A2AAR-WT was completed utilizing standard PCR techniques to amplify the wild type (WT) human A2AAR gene (www.cDNA.org) using modified PCR primers encoding exogenous restriction sites AscI at the 5’, and FseI at the 3’ termini. This A2AAR-WT gene was sub-cloned into expression vector pBac5b-830400, which was modified from the commercially available pBac5 vector (EMD biosciences) with inserted expression cassettes containing HA signal sequence, FLAG epitope tag, precission protease site and 10x histidine tag (1). Subcloning into pcDNA3.1(-) was achieved using PCR with primer pairs encoding restriction sites BamHI at the 5’ and HindIII at the 3’ termini of pBac5b+830400+A2AAR with subsequent ligation into the corresponding restriction sites found in pcDNA3.1(-). A2AAR-T4L construction involved a two-step cloning strategy with the first step using splicing by overlap extension (SOE) PCR (2) to insert a cysteine-free bacteriophage T4 lysozyme (T4L C54T, C97A) (3) within the intracellular loop 3 (ICL3) of A2AAR-WT (replacing Lys209-Ala221). The second step utilized standard PCR techniques to amplify the resulting A2AAR-T4L fusion and subcloned into pBac5b vector. Subcloning into pcDNA3.1(-) followed the same protocol as for A2AAR-WT. Generation of A2AAR-T4L-C is the result of a ligation between A2AAR-T4L and A2AAR-C (317-412). Both were digested in two separate restriction digest reactions using Bsu36I and PciI restriction enzymes. After digestion the larger fragment of A2AAR-T4L was treated as the vector while the smaller fragment of A2AAR-C was used as the insert. Standard cloning methods were implemented and the resulting A2AAR-T4L-C fusion was subcloned into pcDNA3.1(-).

2

Expression and purification of A2AAR-T4L-C for crystallization High-titer recombinant baculovirus (>108 viral particles per mL) was obtained following transfection

protocol

from

Expression

Systems

(http://www.expressionsystems.com).

Briefly,

recombinant baculoviruses were generated by co-transfecting 2 g of transfer plasmid containing the target coding sequence with 0.5 g of SapphireTM baculovirus DNA (Orbigen) into Sf9 cells using 6 L of FuGENE 6 Transfection Reagent (Roche) and Transfection Medium (Expression Systems). Cell suspension was incubated for 4 days while shaking at 27 °C. P0 viral stock was isolated after 4 days and used to produce high-titer baculovirus stock. Viral titers were performed by flow cytometric method after staining cells with gp64-PE (Expression Systems) (4). Sf9 cells at a cell density of 2-3 x 106 cells/mL were infected with P2 virus at MOI of 3. Cells were harvested by centrifugation at 48 hours post infection and stored at -80 °C until use. Insect cell membranes were disrupted by thawing frozen cell pellets in a hypotonic buffer containing 10 mM HEPES (pH 7.5), 10 mM MgCl2, 20 mM KCl and protease inhibitor cocktail (Roche). Extensive washing of the isolated raw membranes was performed by repeated centrifugation in the same hypotonic buffer (~2-3 times), and then in a high osmotic buffer containing 1.0 M NaCl, 10 mM HEPES (pH 7.5), 10 mM MgCl2, 20 mM KCl (~3-4 times), followed by Dounce homogenization to resuspend the membranes in fresh wash buffer thereby removing soluble and membrane associated proteins from suspension of membranes. Purified membranes were resuspended in 10 mM HEPES (pH 7.5), 10 mM MgCl2, 20 mM KCl, and 40% glycerol then flash-frozen with liquid nitrogen and stored at -80 °C until further use. Prior to solubilization, purified membranes were thawed on ice in the presence of 4 mM theophylline (Sigma), 2.0 mg/mL iodoacetamide (Sigma), and protease inhibitor cocktail (Roche). After incubation for 30 min at 4 °C membranes were solubilized by incubation in the presence of 0.5% (w/v) DDM (Anatrace) and 0.1% (w/v) cholesteryl hemisuccinate (CHS; Sigma) for ~2.5-4 hours at 4 °C. The 3

unsolubilized material was removed by centrifugation at 150,000 x g for 45 min. The supernatant was incubated with TALON IMAC resin (Clontech) in the buffer containing 50 mM HEPES (pH 7.5), 800 mM NaCl, 0.5% (w/v) DDM, 0.1% (w/v) CHS, and 20 mM imidazole. After overnight binding the resin was washed with ten column volumes of 50 mM HEPES (pH 7.5), 800 mM NaCl, 10% (v/v) glycerol, 25 mM imidazole, 0.1% (w/v) DDM, 0.02% (w/v) CHS, 10 mM MgCl 2, 8 mM ATP (Sigma) and 25 M UK-432097 (Axon Medchem, prepared as 100 mM stock in DMSO), followed by four column volumes of 50 mM HEPES (pH 7.5), 800 mM NaCl, 10% (v/v) glycerol, 50 mM imidazole, 0.05% (w/v) DDM, 0.01% (w/v) CHS and 25 M UK-432097. The receptor was eluted with 25 mM HEPES (pH 7.5), 800 mM NaCl, 10% (v/v) glycerol, 220 mM imidazole, 0.025% (w/v) DDM, 0.005% (w/v) CHS and 25 M UK-432097 in a minimal volume. Purified receptor in the presence of UK-432097 was concentrated from ~0.4 mg/mL to 60 mg/mL with a 100 kDa molecular weight cut-off Vivaspin concentrator (GE Healthcare). Receptor purity and monodispersity was followed using SDS-PAGE and analytical sizeexclusion chromatography (aSEC).

In meso crystallization of A2AAR with UK-432097 Protein samples of A2AAR in complex with UK-432097 were reconstituted into lipidic cubic phase (LCP) by mixing with molten lipid in a mechanical syringe mixer (5). The protein-LCP mixture contained 40% (w/w) protein solution, 54% (w/w) monoolein (Sigma) and 6% (w/w) cholesterol (Avanti Polar Lipids).

LCP crystallization trials were performed using an in meso crystallization robot as

previously described (6). 96-well glass sandwich plates (Paul Marienfeld GmbH, Germany) were filled with 40 nL protein-laden LCP boluses overlaid by 0.8 L of precipitant solution in each well and sealed with a glass coverslip. Crystallization set-ups were performed at room temperature (~20 °C). Plates were incubated and imaged at 20 °C using an automated incubator/imager (RockImager 1000, Formulatrix). Data-collection quality crystals were obtained in precipitant condition containing 0.1 M sodium citrate 4

(pH 5.0-5.5), 30% (v/v) PEG400, 150-200 mM MgCl2. Crystals appeared about 6 hours after crystallization set-ups and continued growing to full size (~150 m x 20 m x 5 m) within 1 week. Crystals were harvested directly from LCP matrix using MiTeGen micromounts and flash frozen in liquid nitrogen.

Data collection and structure determination X-ray diffraction data were collected on the 23ID-D beamline (GM/CA-CAT) at the Advanced Photon Source (Argonne, IL) using a 10 m minibeam with a MarMosaic 300 CCD detector (Table S1). Crystals embedded in LCP became invisible in the opaque mesophase upon flash-freezing into liquid nitrogen, and a similar rastering and data-collection strategy was followed as previously described (7) for about 60 crystal samples. Although the crystals were relatively long (~150 m), it was not possible to collect data at multiple positions on the same crystal, due to crystal defects accumulated at the crystal ends and, therefore, only one single position per crystal that gave the best diffraction pattern was selected for data collection. Due to the fast onset of radiation damage when using a 1 s exposure of unattenuated beam, data collection was limited to 10 – 30 frames per crystal. Data were integrated, scaled and merged using HKL2000 (8). The data was anisotropic with the best reflection diffracting further than 2.4 Å in the c* direction and lower in the a* and b* direction. A 98.0% complete data set at 2.7 Å resolution was obtained by merging data collected from 20 crystals. A2AAR/UK-432097 crystals were isomorphous with A2AAR/ZM241385 crystals (PDB ID: 3EML). The initial structure was obtained by molecular replacement (MR) with the program PHASER (9), using two independent search models of A2AAR (without ligand) and T4L from the A2AAR/ZM241385 structure. All refinements were performed with the REFMAC5 (10) and autoBUSTER (11) followed by manual examination and rebuilding of the refined coordinates in the program COOT (12) using both sigma-A weighted |2Fo-Fc| and |Fo-Fc| maps, as well as omit maps. 5

Thermal stability assay Thermal stability assays (Table S2) were performed as previously described (13). Purified A2AAR-T4L-C bound to different agonists were screened using the thio-specific fluorochrome N-[4-(7diethylamino-4-methyl-3-coumarinyl)phenyl]maleimide

(CPM)

for

stability

profiling.

Melting

temperatures (Tm) were obtained by fitting the data with a Boltzmann sigmoidal function using Prism (GraphPad Software).

Radioligand binding and functional assays Ligand binding assay: Membranes were isolated from Sf9 cells expressing the A2AAR-T4L-C construct as previously described (1). Briefly, cell pellets were suspended in ice-cold 25 mM HEPES (pH 7.5) lysis buffer, containing protease inhibitor cocktail (Roche) and homogenized with 20 strokes using a Dounce homogenizer. Cellular debris and nucleoli were removed by centrifugation at 400 x g for 5 min at 4 °C, and the supernatant was collected. Crude plasma membranes were isolated by centrifugation of the supernatant at 150,000 x g for 60 min at 4 °C, and were further washed three times by alternating centrifugation and resuspension in 25 mM HEPES (pH 7.5), 1.0 M NaCl. Prior to the ligand binding assays, the membrane pellets were resuspended in ligand binding buffer (TME: 50 mM Tris-HCl, pH 7.4, 10 mM MgCl2 and 1 mM EDTA). For competition binding experiments, 1 nM [3H]ZM241385 (50 Ci/mmol) (ARC Inc., St. Louis, USA) was incubated with crude plasma membranes (2 g/ml of total proteins) and with increasing concentrations of compounds in 200 l TME buffer at 25 °C for 60 min. Binding reactions were terminated by filtration through Whatman GF/B filters under reduced pressure using an MT-24 cell harvester (Brandell, Gaithersburg, MD, USA). Filters were washed three times with 6

ice-cold buffer. Radioactivity was determined using a Tri-Carb liquid scintillation counter (Perkin Elmer, Shelton, CT, USA). Non-specific binding was determined using 10 μM SCH58261 (Tocris). Protein concentration was determined using a protein assay kit (Bio-Rad, Hercules, CA, USA). All experiments were performed in duplicate, and were repeated at least two more times independently. The ability of an allosteric modulator NaCl (14) to shift competition curves of three agonists (UK-432097, NECA (Tocris) and CI-936 (prepared in KJ’s lab)) and one antagonist (SCH58261 (Tocris)) were tested using TME buffer. Characterization of agonist property of UK-432097 in comparison to other known A2AAR agonists using a cyclic AMP accumulation assay: CHO cells expressing the human A2AAR were seeded in 96-well plates and incubated at 37 °C overnight. The medium was removed the following day and replaced with DMEM containing 50 mM HEPES, 10 μM rolipram and 3 units/mL adenosine deaminase. After 30 min incubation time, increasing concentrations of agonists (UK-432097, NECA, CGS21680 (Tocris) and CI936) were added and incubated for another 30 min. The medium was then removed, and the cells were lysed with 100 μL of 0.1 M HCl. The cyclic AMP level was measured with an Amersham cyclic AMP EIA kit following the instructions provided with the kit. The OD values were measured with a SpectraMax M5 plate reader at 450 nm. In each separate assay, the percentage agonist efficacies of agonists at various concentrations were normalized based on maximum effect of 10 M CGS21680 (100%). The experiments were performed in triplicate and were repeated at least two more times in triplicate independently.

7

Supplementary Tables: Table S1. Data collection and refinement statistics. Highest resolution shell is shown in parentheses. Structure

A2AAR_UK-432097 Data collection

Number of crystals Space group Cell dimensions a, b, c (Å) β (◦)

20 P21

Number of reflections measured Number of unique reflections Resolution (Å) Rmerge* Mean I/σ(I) Completeness (%) Redundancy Refinement # Resolution (Å) Number of reflections (test set) Rwork / Rfree Number of atoms All Proteins Ligand (UK-432097) Lipids and waters

103,017 16,713 29.6 – 2.7 (2.8 – 2.7) 0.129 (0.715) 14.6 (1.8) 98.0 (91.3) 6.2 (3.6)

47.8, 78.9, 86.6 100.5

29.6 – 2.7 15,770 (840) 0.217 / 0.273 3,568 3,480 57 31

Average B value (Å2) All A2a T4 lysozyme Ligand (UK-432097) Lipids and waters

63 64 63 52 77

RMSD Bond lengths (Å) Bond angles (°)

0.011 1.27

Ramachandran plot statistics (%)╪ (excluding Gly, Pro) Most favored regions 91.5 Additionally allowed regions 8.3 Generously allowed regions 0.3 Disallowed regions 0.0 * Rmerge = Σhkl |I(hkl) - ‹I(hkl)›|/Σhkl‹I(hkl)›, where ‹I(hkl)› is the mean of the symmetry-equivalent reflections of I(hkl). ╪ Ramachandran plot was calculated by using Procheck v3.4.4 (15).

8

Table S2. Direct contact (within 4 Å) between A2AAR and UK-432097.

Hydrogen bonds A2AAR

UK-432097

Distance (Å)

Tyr2717.36 (OH)

O6

2.7

Glu169 (ECL2) (OE1)

N8

2.8

Glu169 (ECL2) (OE2)

N9

2.7

HOH1204

O5

3.1

Asn2536.55 (OD1)

N6

3.0

Asn2536.55 (ND2)

N3

3.4

His2787.43 (NE2)

O3

2.8

His2787.43 (NE2)

O2

3.1

Ser2777.42 (OG)

O2

3.0

His2506.52 (NE2)

O4

3.1

Thr883.36 (OG1)

N2

3.0

Non-polar interactions A2AAR

UK-432097

Distance (Å)

Met2707.35 (CE)

C32

3.7

Met2707.35 (CE)

C33

4.0

9

Met2707.35 (CB)

C25

3.7

Met2707.35 (CB)

C24

3.6

Met2707.35 (CE)

C24

3.8

Met2707.35 (CE)

C23

3.4

Met2707.35 (CE)

C22

3.4

Leu2677.32 (CD2)

C32

3.8

Leu2677.32 (CD2)

C33

3.7

His264 (ECL3) (CB)

C37

3.9

His264 (ECL3) (CG)

C37

3.7

His264 (ECL3) (CD2)

C37

3.4

His264 (ECL3) (CD2)

C38

3.8

Leu167 (ECL2) (CD2)

C29

3.8

Tyr2717.36 (CZ)

C28

3.6

Ile2747.39 (CD1)

C27

3.6

Ile2747.39 (CD1)

C12

3.8

Ile2747.39 (CD1)

C25

3.7

Ile2747.39 (CG1)

C25

3.9

Ile2747.39 (CD1)

C24

3.6

Phe168 (ECL2) (CB)

C12

3.3

Phe168 (ECL2) (CG)

C12

3.5

Phe168 (ECL2) (CD2)

C12

3.7

Phe168 (ECL2) (CG)

C11

4.0

Phe168 (ECL2) (CD2)

C11

3.3

Phe168 (ECL2) (CE2)

C11

3.7

Phe168 (ECL2) (CG)

C9

3.9

Phe168 (ECL2) (CZ)

C9

3.8

Phe168 (ECL2) (CE2)

C9

3.3

Phe168 (ECL2) (CD2)

C9

3.3

Phe168 (ECL2) (CG)

C10

3.6

10

Phe168 (ECL2) (CD1)

C10

3.7

Phe168 (ECL2) (CE1)

C10

3.8

Phe168 (ECL2) (CZ)

C10

3.7

Phe168 (ECL2) (CE2)

C10

3.6

Phe168 (ECL2) (CD2)

C10

3.6

Phe168 (ECL2) (CZ)

C8

3.6

Phe168 (ECL2) (CE2)

C8

3.8

Phe168 (ECL2) (CE1)

C4

3.8

Leu2496.51 (CD1)

C25

4.0

Leu2496.51 (CD2)

C2

3.7

Thr2566.58 (CG2)

C19

3.8

Thr2566.58 (CG2)

C18

3.4

Met1775.38 (CE)

C8

3.3

Trp2466.48 (CZ3)

C2

4.0

Trp2466.48 (CZ3)

C5

3.5

Ile923.40 (CD1)

C7

3.3

Leu853.33 (CD2)

C5

3.9

11

Table S3. Effect of different ligands on the thermalstability of purified A2AAR-T4L-C receptor.

a. b. c. d.

Measured in this study (Supplementary fig. 1). Measurement in rat striatal membranes (16, 17). Measurement in membranes of HEK293 cells expressing human A2AAR-T4L-∆C (1). Measurement in HEK293 membranes (18).

12

Supplementary Figures:

Figure S1. Displacement of [3H]ZM241385 with three agonists and one antagonist in membranes prepared from sf9 cells expressing A2AAR-T4L-∆C in the presence and absence of an allosteric modulator NaCl. Three agonists: (A) UK-432097; (C) NECA; (D) CI-936; Antagonist: (B) SCH58261. [3H]ZM241385 (1 nM) was incubated with membranes (2 g/mL of total proteins) in a final assay volume of 200 L Tris–HCl buffer (50 mM, pH 7.4) containing 10 mM MgCl2, 1 mM EDTA for 60 min at 25 ºC in the presence and absence of 1M NaCl. The binding affinities (Ki values) of UK-432097 for A2AAR-T4L-∆C construct in the absence and presence of NaCl were 4.75 ± 0.73 and 39.3 ± 6.47 nM. Data represent three to five separate experiments performed in duplicate.

13

Figure S2. Characterization of the agonist property of UK-432097 using CHO cells expressing human WT A2AAR. The efficacy and potency of UK-432097 was compared to three other known agonists (NECA, CGS21680 and CI-936) using a cylic AMP accumulation assay in intact CHO cells. The EC50 values of UK-432097, NECA, CGS21680 and CI-936 are 0.66 ± 0.19, 5.99 ± 1.86, 3.25 ± 1.22 and 14.5 ± 5.81 nM, respectively. Results are from three independent experiments performed in triplicate.

14

Figure S3. Examples of the electron density maps calculated from the refined model for A 2AAR/UK432097. (A) |Fo-Fc| omit map for the ligand UK-432097 (green stick), contoured at 3.0. (B) |2Fo-Fc| map for part of helices V and VI (orange tube and stick) contoured at 1.4. Residues Tyr197, Phe201, Lys227

15

and Glu228 are labeled. The images were created with PyMOL (The PyMOL Molecular Graphics System, Version 1.2r3pre, Schrödinger, LLC (2002)).

Figure S4. Conformational plasticity of TM helices III, V, VI and VII in A2AAR transition from antagonist bound (yellow) to agonist-bound (orange) state. For each helix, the left panel shows the overall movement of this helix in the context of the whole TM bundle superimposition. The right panel shows 16

deformations in each helix by individually superimposing two conformations in the top portion of the helix (superimposed Cα atoms marked with green dots). (A) Comparison of the left and right panels for helix III suggests that the axial translation is accompanied by a substantially reduced bend in the middle of the helix. (B), (C) and (D) For helices V, VI and VII, lateral movements in the intracellular region are accompanied by smaller “compensating” bending of each helix in the opposite direction as shown by black arrows. References:

1.

V. P. Jaakola et al., Science 322, 1211 (2008).

2.

K. L. Heckman, L. R. Pease, Nat Protoc 2, 924 (2007).

3.

D. M. Rosenbaum et al., Science 318, 1266 (2007).

4.

M. A. Hanson et al., Protein Expr Purif 56, 85 (2007).

5.

M. Caffrey, V. Cherezov, Nat Protoc 4, 706 (2009).

6.

V. Cherezov, A. Peddi, L. Muthusubramaniam, Y. F. Zheng, M. Caffrey, Acta Crystallogr D Biol Crystallogr 60, 1795 (2004).

7.

V. Cherezov et al., J R Soc Interface 6 Suppl 5, S587 (2009).

8.

Z. Otwinowski, W. Minor, Methods Enzymol 276, 307 (Academic Press, New York, 1997).

9.

A. J. McCoy et al., J Appl Crystallogr 40, 658 (2007).

10.

G. N. Murshudov, A. A. Vagin, E. J. Dodson, Acta Crystallogr D Biol Crystallogr 53, 240 (1997).

11.

G. Bricogne et al. (Global Phasing Ltd., Cambridge, United Kingdom, 2009).

12.

P. Emsley, B. Lohkamp, W. G. Scott, K. Cowtan, Acta Crystallogr D Biol Crystallogr 66, 486 (2010).

13.

A. I. Alexandrov, M. Mileni, E. Y. Chien, M. A. Hanson, R. C. Stevens, Structure 16, 351 (2008).

14.

Z. G. Gao, A. P. IJzerman, Biochem Pharmacol 60, 669 (2000).

17

15.

R. A. Laskowski, M. W. Macarthur, D. S. Moss, J. M. Thornton, J Appl Crystallogr 26, 283 (1993).

16.

K. A. Jacobson, P. J. van Galen, M. Williams, J Med Chem 35, 407 (1992).

17.

G. Pastorin et al., Bioorg Med Chem 18, 2524 (2010).

18.

M. W. Beukers et al., J Med Chem 47, 3707 (2004).

18

Movie S1. Structural transition from antagonist-bound to agonist-bound A2AAR. The movie starts with an overall view of the inactive-state A2AAR (yellow ribbon) bound to antagonist ZM241385 (gray stick) that transforms to the active-state (orange ribbon) upon binding of agonist UK-432097 (green stick), followed by three zoom-in views of this transition at ligand binding pocket, extracellular side and intracellular side, sequentially. The conformational morphing trajectory is repeated three times for each view. Key residues involved in agonist binding and large side-chain movements during this transition are highlighted. The movie is prepared with ICM molecular modeling suite (Molsoft LLC), the interactive 3D version is also available from http://gpcr.scripps.edu/A2a_ActivationMovie.html