FROM THE MAKERS OF AND

REPRINT FROM DECEMBER 4, 2014

GPCRs’ grand plans By Stephen Parmley, Senior Writer

In a move to expand tenfold the number of known 3D structures of the He said that industry members will provide the consortium with highly druggable class of GPCRs, Amgen Inc., Ono Pharmaceutical libraries of their chemical compounds, many of which have fallen by Co. Ltd. and Sanofi have teamed up with three academic organizations the wayside for pharmaceutical or safety reasons. He added that the to create the GPCR Consortium—a precompetitive alliance to build compounds are great tools for binding receptors and stabilizing them an open-source repository of GPCR structures. The consortium could for crystallization and might be useful for bootstrapping structures fill a hole left by the termination of the NIH-backed Protein Structure to develop new drugs. Initiative that until March constituted the main public effort to char- “Having their help in accessing and generating compounds that acterize GPCR structures. are going to bind to the receptors and analyzing the data associated Raymond Stevens—who started the consortium—told SciBX in with that binding event is probably the most important aspect of what late November that Novo Nordisk A/S will also join the group. He pharma is bringing to the collaboration,” he said. expects to sign up another pharma before year end, and he said that the consortium hopes to reach a total of eight industry members. Lessons learned The academic centers involved—the iHuman Institute at Shang- Stevens told SciBX that a major stimulus for the new collaboration was haiTech University, the Shanghai Institute of Materia Medica and the termination of funding for the Protein Structure Initiative (PSI) the University of Southern California—will conduct the research on by the NIH’s National Institute of General Medical Sciences earlier GPCR structures and make the results and supporting data available this year. According to an NIH press release, the initiative was discon- in the public domain. Financial terms for the consortium were not tinued after an external review committee concluded that, despite the disclosed. gains made since PSI was founded in 2000, the resources, products Stevens is founding director of the iHuman Institute and provost and results were “underutilized by the broader scientific community.” professor of biological sciences and chemistry at the University of PSI was originally formed to develop and use high throughput Southern California. He is also founder of Receptos Inc. and RuiYi screening systems to solve 3D atomic-level structures of proteins and Inc. make them easily obtainable by the scientific community. The pro- The goal is to elucidate the 3D structures of a large number of gram involved multicenter collaborative studies and produced more GPCRs and generate high-resolution pictures that can be used to than 6,300 protein structures and 400 technologies and methods to explore how the receptors work and aid the design of new compounds. streamline the process of structure determination. The consortium’s initial focus will be on diabetes, cancer and mental Stevens was the principal investigator from The Scripps Research disorders based on the industry members’ input. But, according to Institute who formed the GPCR Network, a collaborative program Stevens, there is no limit on therapeutic areas, and new consortium funded by PSI to understand GPCR structure and function. He told members may have different interests. SciBX that the GPCR structures obtained in the PSI program were Stevens said that with 8 companies on board, the consortium viewed as useful but the PSI program was controversial because it was believes it will be able to study at least 200 GPCRs. less hypothesis-driven research than the National Institute of General Michael Hanson, president of the GPCR Consortium, noted Medical Sciences is currently funding. that GPCRs constitute the largest family of proteins in the human He added that for some of the leading academic groups in the body and represent therapeutic targets for about 40% of marketed field of GPCR structural biology research, the only solution to the cut drugs. “What is surprising is that these developed drugs really only in PSI funding was to work more closely with industry. By allowing target a handful of the known family of GPCRs. So there is a vast the pharmas to select the targets and collaborate on the science, the untapped potential out there,” he said. But “at the moment, we only consortium hopes to generate data that is more therapeutically useful, have structures for 26 of the 826 known human GPCRs. There is a he told SciBX. lot that we do not know about this family.” (See Figure 1, “Solving a In putting together the GPCR Consortium, Stevens used the setup family problem.”) of the Structural Genomics Consortium (SGC) as a template with Hanson is the former director of structural biology at Receptos. assistance from Aled Edwards, director and CEO of the SGC.

1 ANALYSIS COVER STORY

SECRETIN GRM77 GRM8G GLUTAMATE (15) Figure 1. Solving a family problem. The GRM2 FZD7 TASAS1R3 (15) FZDF 2 GRM4 GRM3 TATAS1R1S FZDD1 GPCR Consortium aims to solve at least GGLLPPP2P2R22RR GGIIIPIPRPPRR GRM6GRGR GGRRPRPCPC6C6A /TAS2R ADHESION GLPGLGLP1RGLLPPP11R GCGGCCGR GRMG MM55 FZDF 3 200 unknown structures of GPCRs. PTHR2PPTHRR2R2 GRM1GRGRMMM11 (24) LEC1 VIPRPR22 PTHR1 (24) TAAS1SS111RR2 FZD6 The GPCR superfamily contains 826 LEC2 TAS2R13 CELSR2PAP CAP FZD8 CRHR2 FZD5 TAS2R16 TAS2R14 LEC3 CALCRLC members, identified based on sequence CELSR3 VIPRPR1 CRHR1CRHCRRHR1HR1R1 CASA R GABBR2 FZD10 TAS2R11 TAS2R10 EMR2 TTAAS2R5R5 TTAAASAS2R3AS2S R BAI2 SCTRR CALCC R FZFZDZDD44 FZDF 9 TATAAS2ASS2S2R2R9 ETL BAI3 GPR6G 0 similarity of their hallmark 7 transmembrane EMR3 CELSR1LSR11 GHRHRRH GABBRR1 GPR5G R59 TTAAS2AASS2S2RR8R TTAAASSS22R2R4 TATAASSS22R2R7R BAIA 1 SMSMOHOH CXCCXCRXXCCCRRR33 EMR11 CCXXXCCCRCR5R5R5 CCCCCRCR1RR1111 domains. Structures of 16 GPCRs have SSSTR3SSTSSSSTR3STTTRR3R3 CCCCR1C 10 CXCR2CXCXCR2 CD977 SSTS RR11 CCCCCRCRCR6R6 CXXCR1CR1CCRR11 SSTSSSTR5SSSSTR5STTTRRR55 GPR1111 CXCCXCR4CXXXCCCRR4R44 CCXXCXXCR6CCRCR6RR66 already been solved by the GPCR Network SSSSSTTRTR2R2 CCCRCCR9CCCRCR9CRR99 CCCRCCRCR7R7R7 GPR115 GPR8GPGPPRRR88 SSTSSTR4SSSSTR4SSTTTRRR44 GPR1166 GPR1122 GGPPPRRR77 CCRLC 2 (black labels), a division of the NIH-backed GPR1P 13 OPRLOOPPRP 1LL11OPRKO 1 CXC3CCXCCC3RR1CCRCCCR8CCR8CRRR88 GPR110 NTSR1NNTNTST OOPPPRRKKK11 CCCRCCR4CCCR4CCRRR4 HE6 NTSR2NNTSNTTSSSRR OPRMOOPPRPR 1 CCCCCRRR11 TM7XN11 GPR1114 GPGPPR54PRR5R5544 NNMU1RNMUNMU1MMU1MUMU1RUU11R GHSG R GALRGGAAALLR1R1GALR2 CCBP2C P22 Protein Structure Initiative that was termi- NPY1P R OPRD1 RDRRDCRDC1DDCCC11 XCR1X CCCCCRCR3RR33 GPR9977 PPYR1 NNMMMUU2RUU222RR GALRG 3 AADDMDMMRR MTLR MCHR1 AGTRA 1 TACR33 NPY2R CCR5 nated in March. Another 10 structures have PrRPP P UR2U R MCHR2 AAGTRGTRLGTGTTRL1 AGTRA 2 CCRR55 TAC3RLRL γ BDKRB2 GRP72 GPR26 CCR2C TACR11 TACR2 OR1A1 GPR15 SALPR NPFF1 OLFACTORY been solved by other groups, 9 of which are NPY5R OR1D2 CRC TH2 BDKBR1 β RECEPTORS(33388388888) NPFF22 HCRTRR22 CCKBRRCCKARR OR1G1 GPR32 GNRHRII AADDOOORRARA1 OR3A1 BLLTTRTRR22BLTR depicted in the dendrogram (gray labels). HCRTR1 MC3RC3RC3C3 HR FPR1 GNRHR ADORA2A AADORADDODDORAOORRRAA3 δ TRHRT GPR11 MC5RMC ADORA2AA GPR78 RE C5R2 FPRL2 α EB12 2 C5R1 ADORA2A B B12 GPR101 CMKLR1 GPCRs are divided into five major AVPR1AA MC4RR GPGGPR1PRPRR111119 GPR26 GPR62P LGR8 GPR1GGPC3ARC P2R 1 FPRLF 1 GPR33 RG LGR7L MC1RR GPR103PTP Y1 AVPR188 GPR6GPGPRPR622 AFR 1 FSHR AAVVPVPRVPPRRR22 GPR66 MC2RR R 8 MRGDM D families: , secretin, adhesion, OXTRR BRBRSR 3 CNR2 GGPR6GPPPRPR61RR661 LHCGR GPR1G 2 SSRRREEBEB1EBB11 EDG3EDDGG3 CNR11 PPTTGGEERR4R4 MRGFF LGR4 TSHRT NNMMMBBRBR GPR50 HRHHRHRHH1H11 MRG frizzled type 2 EDG1EDG1E 1 MTNR1BB SRSSREB2SREBRREREB2EEB2EBBB22 MAS MRGX2 glutamate and / GRGRPGRPRPRPR EDEDG5GG55 H9H963H9969636633 LGR6L MTNR1AA SRSREBSREREB3REEBBB33 FFFFAFA1A R FKSG80 EDG8 EDG6DGG6 HRH2 PTGERP PTGDDRDR ETBRLPLP1P1 EEDDNRADNN A PTPPTGITTGIRGIG R LGR5 (TAS2R; T2R). EDGG22 GPR5GPG 2 OPN4OPO N4 PAR1 HM74 EDNRB P2Y12P2Y12 MRGX1 HRH3 1 P GP ETBRLP2 GGPPR212211 T T F2F RRLLL11 G EEDDDGGG77 HRH44 BXA2 GE FKSG77FKSGGG777 PRP G R EDEEDG4DDGGG44 RRHRRRRRHH 2 P PR35 55 MRGXM 3 OOPPPNNN33 PTP G L2 2 9 Receptors with solved structures in HHTTTRR4R R F Y 2 PNR 2 KSK PTGR 2RL2 P2P 101 MRGX4 F 3 2Y OOPPPNN1N SW FR L3 Y GG79 87 1 OPN1NN11LLW DRD5DRDDDRRRDRD5D5D5 ER RLR HHTHTRTTRR6 TAR1 F2 2 1 P2Y9 the dendrogram include: adenosine A2A DRD1DDRDRDRD1RRDDD11 GPR87 T P2 G RHO TAT R3 3 LT P OPN1MWW 7 P Y6 2Y ADRB2AADRBB22 2 4 G R9 ADRBADR 3 GGPPR58 CYS R4 TAT R5 GPR106CYSL PR17 PR 5 1 GPR5GGPPR57PPRRR55757 G P2Y4 P receptor (ADORA2A), adrenergic recep- GP 80 ADRB1 TAR4 2Y HTR2BHTR2BR2BB 5 A GR1 HTHHTRHTR2HTR2CTR2TTRTR2CRR2R2C22CC 65 2A O 2 DRD4 R6 G tor b1 (ADRB1), ADRB2, CC chemokine HTR2ATR2A2A2A HTRHHTTRTR5R ADADDRRRAA1A1D HTR7HTRHHT 7 DRD3 ADADDRDRARARA11B1B DRDDRD receptor 5 (CCR5; CD195), muscarinic AADDRDDRARARAA11A HTHTRHTRTR1AR1R A HTTTRRR111EE ADRA2AA A DRD2 CHRM1 M2 (CHRM2; HM2), HHTTTRRR111FF RHODOPSIN HHTHTRTRTR1TR1DTRR1D1D AADDDRRRAA2C CHRMCHRRMM33 (701) HTR1BHTR1BTR1BR1 ADRA2BA CHRM3 (HM3), corticotropin-releasing CHRM5 CHRM4 factor receptor 1 (CRHR1; CRFR1), CXC CHRMCCHRRM222 1 (CXCR1), CXCR4 (NPY3R), dopamine D3 receptor (DRD3),

(GCGR), metabotropic Institute. Katya Kadyshevskaya, The Scripps Research subtype 1 (mGluR1; GRM1), (HRH1), serotonin (5-HT1B) receptor (HTR1B), HTR2B, 1 (NTSR1), d1 (OPRD1; DOR), k-opioid receptor (OPRK1; KOR), opiate receptor-like 1 (OPRL1), m-opioid receptor (OPRM1; MOR), protease-activated receptor 1 (PAR1), P2Y G protein–coupled 12 (P2RY12; ), rhodopsin (RHO; OPN2), (SMO), sphingosine 1-phosphate receptor 1 (S1PR1; S1P1; EDG1). Not shown: 1 (FFAR1; GPR40). (Figure adapted from Figure 1 in ref. 2.)

“That model worked really well, particularly for kinases and epi- takes to generate structural data. genetics,” said Stevens. “Before, labs were solving structures of a few Hanson added that acquiring the data is just the first step. Indus- kinases here and there, but the SGC did it in an organized fashion try funding would also be used for organizing and sharing the data and did an incredible job of opening up the kinase knowledge base.” so that it can be used more effectively by the pharma partners and The SGC is a public-private partnership founded in 2004 to deter- the larger biomedical community, he said. mine protein structures that operates out of the University of Oxford The importance of data management was a lesson learned from and University of Toronto. Results are placed in the Protein Data the genomics field, which also dealt with vast amounts of data, he Bank, which is the primary public source of protein structures.1 The said. “It’s very analogous to the Human Genome Project, where GPCR Consortium will also deposit its data in the Protein Data Bank. academia had their effort and industry had their effort,” which led Like the SGC, the GPCR Consortium will rely on international to a lot of expense, inefficiency and a cottage industry of small labs collaboration between academic centers that have different strengths engaged in redundant efforts. and skill sets. “What we are trying to do is take it a step further and evolve, so Stevens said, “In terms of novel technologies, in China we will be instead of competing we are working together to collect this informa- doing a lot of the in vitro stability screening and signaling assays, and tion,” Stevens said. in the U.S. we have access to a new technology called the free electron “There are a lot of GPCR data being generated, including but not laser that is able to use much smaller crystals.” He also noted that the limited to structural data, novel signaling pathways, allosteric modu- technology to obtain protein structures has advanced significantly in lation and polypharmacology,” he said. “We are developing solutions the last decade, which has lowered the cost and shortened the time it to integrate access and ultimately utilize all of this data to accelerate

SciBX: Science–Business eXchange REPRINT FROM DECEMBER 4, 2014 2 ANALYSIS COVER STORY

the process of drug discovery.” Parmley, S. SciBX 7(46); doi:10.1038/scibx.2014.1337 Published online Dec. 4, 2014

REFERENCES 1. Cain, C. SciBX 4(20); doi:10.1038/scibx.2011.562 2. Katritch, V. et al. Annu. Rev. Pharmacol. Toxicol. 53, 531–556 (2013)

COMPANIES AND INSTITUTIONS MENTIONED EDITORIAL Amgen Inc. (NASDAQ:AMGN), Thousand Oaks, Calif. Editor-in-Chief: Karen Bernstein, Ph.D. GPCR Consortium, Los Angeles, Calif. Managing Editor: Gaspar Taroncher-Oldenburg, Ph.D. National Institute of General Medical Sciences, Bethesda, Md. Executive Editor: Steve Edelson National Institutes of Health, Bethesda, Md. Senior Editors: Tracey Baas, Ph.D.; Amy Donner, Ph.D.; Novo Nordisk A/S (CSE:NVO; NYSE:NVO), Bagsvaerd, Denmark C. Simone Fishburn, Ph.D. Ono Pharmaceutical Co. Ltd. (Tokyo:4528), Osaka, Japan Associate Editor: Benjamin Boettner, Ph.D. Receptos Inc. (NASDAQ:RCPT), San Diego, Calif. Writers: Chris Cain, Ph.D.; Michael J. Haas; Kai-Jye Lou; Lauren Martz; RuiYi Inc., La Jolla, Calif. Lev Osherovich, Ph.D. Sanofi (Euronext:SAN; NYSE:SNY), Paris, France Research Director: Walter Yang , La Jolla, Calif. Research Manager: Kevin Lehnbeuter The Scripps Research Institute , Shanghai, China Production Editors: Brandy Cafarella; Carol Evangelista; Ivelisse Robles Shanghai Institute of Materia Medica Copy Editor: Nicole DeGennaro ShanghaiTech University, Shanghai, China Editorial Assistant: Mark Zipkin Structural Genomics Consortium, Oxford, U.K. Design: Claudia Bentley; Miles Davies University of Oxford, Oxford, U.K. For inquiries, contact [email protected] University of Southern California, Los Angeles, Calif. University of Toronto, Toronto, Ontario, Canada PUBLISHING Publisher: Peter Collins, Ph.D. Associate Publishers: Gaspar Taroncher-Oldenburg, Ph.D.; Eric Pierce Marketing: Sara Girard; Greg Monteforte Technology: Anthony Barrera; Julia Kulikova Sales: Ron Rabinowitz; Dean Sanderson; Tim Tulloch

OFFICES BioCentury Publications, Inc. Nature Publishing Group San Francisco New York PO Box 1246 75 Varick Street, 9th Floor San Carlos, CA 94070-1246 New York, NY 10013-1917 T: +1 650 595 5333 T: +1 212 726 9200 Chicago London 20 N. Wacker Drive, Suite 1465 The Macmillan Building Chicago, IL 60606-2902 4 Crinan Street T: +1 312 755 0798 London N1 9XW United Kingdom United Kingdom T: +44 (0)20 7833 4000 T: +44 (0)18 6551 2184 Tokyo Washington, DC Chiyoda Building 6F 2008 Q Street, NW, Suite 100 2-37 Ichigayatamachi Washington, DC 20009 Shinjuku-ku, Tokyo 162-0843 T: +1 202 462 9582 Japan T: +81 3 3267 8751

SciBX is produced by BioCentury Publications, Inc. and Nature Publishing Group Joint Steering Committee: Karen Bernstein, Ph.D., Chairman & Editor-in-Chief, BioCentury; David Flores, President & CEO, BioCentury; Bennet Weintraub, Finance Director, BioCentury; Steven Inchcoombe, Managing Director, Nature Publishing Group; Peter Collins, Ph.D., Publishing Director, NPG; Christoph Hesselmann, Ph.D., Chief Financial Officer, NPG. Copyright © 2015 Nature Publishing Group ALL RIGHTS RESERVED. No part of the SciBX publication or website may be copied, reproduced, retransmitted, disseminated, sold, distributed, published, broadcast, circulated, commercially exploited or used to create derivative works without the written consent of the Publishers. Information provided by the SciBX publication and website is gathered from sources that the Publishers believe are reliable; however, the Publishers do not guarantee the accuracy, completeness, or timeliness of the information, nor do the Publishers make any warranties of any kind regarding the information. The contents of the SciBX publication and website are not intended as investment, business, tax or legal advice, and the Publishers are not responsible for any investment, business, tax or legal opinions cited therein.

SciBX: Science–Business eXchange REPRINT FROM DECEMBER 4, 2014 3