Ras Initiative Update Frank McCormick and Levi Garraway

DEPARTMENT OF HEALTH AND HUMAN SERVICES • National Institutes of Health • National Cancer Institute The Frederick National Laboratory is a Federally Funded Research and Development Center operated by Leidos Biomedical Research, Inc., for the National Cancer Institute RAS Initiative Accomplishments: Evaluating Ras dependency

2 SiREN assay for Ras dependency

SiREN assay for Ras dependency

70% KD

siEGFP

0-10% KD

50% KD

100% KD

SiREN assay for Ras dependency RAS Initiative Accomplishments: Biophysical and structural analysis

7 Ras proteins

1 -166 167-185,6

Raf, PI 3’ kinase RalGDS, GAPs

Raf, PI 3’ kinase RalGDS, GAPs

Raf, PI 3’ kinase RalGDS, GAPs

Raf, PI 3’ kinase RalGDS, GAPs

Palmitoyl Farnesyl Fully processed KRAS4b

A. Gorfe, U-Texas Houston Engineering baculovirus for improved production of processed KRAS • recombineering used to insert FNTA/FNTB genes into the baculovirus genome • eliminated issues with coinfection of multiple viruses • maltose-binding protein (MBP) fusion for greater yield and solubility • Trichoplusia ni (Hi5) insect cells for increased yield

Carissa Grose, Dom Esposito, Bill Gillette Processed KRAS4b characterization

• Extensive protein characterization – Purified to homogeneity; yield >7mg/L – Intact mass – Predominantly monomeric – Secondary structure equivalent to non-processed

KRAS4b KRAS4b- FME – Lower thermal stability

Intact mass analysis Analytical ultracentrifugation Secondary structure by CD

2.14 S 25.7 kDa

10

Bill Gillette, Zhaojing Meng, Shelley Perkins, Peter Frank, Pat Alexander, Rodolfo Ghirlando Processed KRAS4b binds to Nanodiscs in a phosphotidylserine-dependent manner

Nanodiscs containing DMPS

1,2-dimyristoyl-sn-glycero-3- phospho-L-serine

11 Lakshman Bindu, Que Van FNL; Steve Sligar and Mike Gregory UI-UC Processed KRAS enables assays, screens and structural analysis in the context of membrane

Lipid Nanodiscs Liposomes and Processed KRAS tethered bilayers Structures of Characterize Assays and Structural KRAS and reagents Screens analysis effectors on membranes

Protein QC Intrinsic GTP hydrolysis NMR (NMRFAM) Biophysical properties GAP-stimulated hydrolysis Hi-res Cryo-EM (NCI) Membrane interactions Effector binding Crystallography (FNL) RBD-KRAS Alpha assays Sligar lab (U-Illinois) Groves lab (UC-Berkeley) Heinrich Lab (NIST) Processed KRAS4b workshop – collaborative opportunities

• Frank Heinrich (Neutron reflectivity of protein/membranes; Carnegie Mellon University) – Determine orientation of KRAS4b-FME GDP/GMP-PNP on membrane – Impact of Calmodulin on HVR • Alemayehu Gorfe (Modeling of RAS on membranes; University of Texas Health Science Center) – Interest in testing predicted dimer interactions – Molecular dynamics analysis of FNL extended switch 1 KRAS4b-GDP structure • Mitsuhiko Ikura (NMR of KRAS on membranes; University of Toronto) – Analysis of processed KRAS4b by NMR • Jeff Perry (Small angle X-ray scattering; University of California Riverside) – Screening of crystallography conditions of challenging targets • Vadim Cherezov (Crystallography of membrane proteins; University of Southern California) – Attachment of transmembrane helix for anchoring of processed KRAS4b for crystallography • Jay Groves (RAS on tethered bilayers; University of California at Berkley) – Effector interactions – Development of screenable assays on tethered bilayers

13 Full-length KRAS in complex with GDP

Full-length Wild-type KRAS-GDP Wild-type KRAS(1-166)-GMPPNP complex at 1.6 Ang complex at 1.35 Ang

N

C N

C

Switch-I Switch-II P-loop Extended switch-I conformation in KRAS

• Validate presence of extended switch-I conformation in solution by NMR. Electrostatic surface – Dynamic studies in collaboration with National Magnetic Resonance Facility at Madison. Que Van at FNLCR

– High-pressure NMR studies in collaboration with Dr. Kalbitzer, University of Regensburg, Germany.

• Virtual compound screening to target the groove present at the base of switch-I region – in collaboration with Dr. Brian Shoichet’s group at Red - negative charge White - neutral UCSF. Blue - positive charge 15 Processed KRAS in complex with PDEδ

N

C

KRAS PDEδ

N

Resolution: 2.0 Ang 16 RAS Initiative Accomplishments: Inhibitor Screens and Assays

17 Isogenic Screen for RAS Selective Inhibitors

-/- -/- lox/lox HRAS NRAS KRAS MEFs Untreated MEFs G1 arrest (day 19*)

Drosten M, Dhawahir A, Sum EY, Urosevic J, +4-OHT Lechuga CG, Esteban LM, Castellano E, Guerra C, Santos E, Barbacid M. EMBO J. 2010

+ drugs

+KRAS G12D Re-enter cell cycle

+ drugs

+HRAS

18 HRASWT vs KRASG12D Pilot Screen

HRAS-WT “selective”

KRAS-G12D “selective”

• Compound library was provided by NCATS (National Center for the Advancement of Translational Sciences)

• The library is enriched for “tool” compounds, but also contains FDA approved drugs 19 Kanika Sharma (FNLCR), Kyle Brimacombe (NCATS) KRAS-effector Inhibitor Screen Daiichi-Sankyo Protein-Protein Interaction Library

Z-factor = 0.65

20 Reproducibility of KRAS-RBD screening assay

Assay is highly reproducible at 50 μM The “hit” rate at 50 μM is approximately 4% 13/320 compounds inhibit the alpha signal >25%

21 KRAS4b-FME binds to CRAF-RBD on Nanodiscs

22 Matt Holderfield, Maria Abreu Blanco RAS Localization Assay Overview

~125 Hit 60 Wells Cells/location identification

8 Locations/ Segmentation well and Analysis

Nucleus GFP Membrane

GFP-KRAS4bG12V Alla Brafman

NCI Developmental Therapeutics Program screening set

Reconfirmed hits Primary assay: 150 GFP-KRASG12V 2-C 5 Plate 1 (Z' =0.84) 2-F6 4-B3 Plate 2 (Z' =0.77) 100 5-F7 250 Plate 3 (Z' =0.89) 7-B8 Plate 4 (Z' =0.80) 10-B2 % A ctivity 50 200 Plate 5 (Z' =0.79) 11-C5 Plate 6 (Z' =0.68) FOVMean PM GFP 11-E3 150 Plate 7 (Z' =0.64) 0 3 2 3 2 3 2 3 2 3 2 3 2 3 2 3 2 10 1 0 10 1 0 1 0 10 10 1 0 100 Plate 8 (Z' =0.84) 1.25 1.25 1.25 1.25 1.25 1.25 1.25 1.25 % A ctivity Plate 9 (Z' =0.74) [ µ M] 50 FOVMean PM GFP Plate 10 (Z' =0.73) 150 Plate 11 (Z' =0.83) 11-B5 0 Plate 12 (Z' =0.74) 12-D2 13-B7 Plate 13 (Z' =0.83) 100 PlatePlate 1P late2Plate 3 Plate 4 Plate 5Plate 6Plate 7 Plate 8 9 PlatePlate 10Plate 11Plate 12Plate 13Plate 14Plate 15Plate 16 17 13-C7 Plate 14 (Z' =0.79) 14-D6 Plate 15 (Z' =0.88) 14-F7 % A ctivity 50 ~800 small molecules with Plate 16 (Z' =0.74) 15-D9 biological activity Plate 17 (Z' =0.88) FOVMean PM GFP 16-B9 0

3 2 3 2 3 2 3 2 3 2 3 2 3 2 3 2 10 10 10 1 0 1 0 10 10 1 0 1.25 1.25 1.25 1.25 1.25 1.25 1.25 1.25

[ µ M]

Alla Brafman HaloTag-KRASWT driven-MEFs Proliferate

-/- -/- lox/lox HRAS NRAS KRAS MEFs Untreated MEFs G1 arrest (day 19*)

Drosten M, Dhawahir A, Sum EY, Urosevic J, Lechuga CG, Esteban LM, Castellano E, +4-OHT Guerra C, Santos E, Barbacid M. EMBO J. 2010

HaloTag-KRAS4b can be HaloTag-KRAS4b rescues imaged in live cells. RASless MEF proliferation.

+HaloTagKRAS Cell permeant, super bright, fluorescent Halo ligand from Janelia Farms Scale bar 20 µm

TIRF Image: membrane Transmitted light image Nikki Fer and De Chen Characterization of RAS molecules in live cell membranes

WT Jump squared displacement analysis HaloTag-KRAS driven-MEFs

Three components

Fraction Diffusion Mean Const. Rad. 2 Model (um /s) (SDev) Rc (nm) 0.505 0.73 - 1  Normal (0.0193) 0.233 0.1805 44.2 2  Constrained (0.021) 0.2624 0.0178 1.2 3  Constrained (0.026)

De Chen and Prabhakar Gudla Single molecule tracking analysis suggests three RAS states in live cell membranes.

Information extracted Hypothesis: states represent from individual different complexes in trajectories membrane.

P is the probability of the transition from one state to another per second

is the dwell time D is the diffusion rate Fτ is the fraction

RASless-MEFs, HaloTag-wtKRAS4b [JF646]=50pM, Serum Starved, 37°C, 22,325 trajectories and average trajectory length 12 frames. De Chen and Prabhakar Gudla RAS Initiative Accomplishments: Towards a “RAS Interactome”

28 Cancer.gov/ras

29 RAS Lab (Basecamp)

• Private, by invitation only • All posts and comments publish immediately • Supports uploads of documents, 1:1 interactions RAS Reagents Pathway 2.0 clone collection

• Generate stop and nostop Gateway Entry clones for all genes • Isoforms chosen by bioinformatic analysis (R. Stephens) - most common transcript observed across different cell lines - many are NOT “isoform 1” or “longest isoform” • 180 genes - 360 total constructs (stop/nostop) - 17 not available commercially in correct isoform - 32 additional not available without mutations • 98% completion Vanessa Wall Jen Mehalko Collaboration with the RAS Community

RAS events

Synthetic Lethality Workshop, January 6-7 2014 RAS Pathways Workshop, June 11, 2014 Cell Surfaces Workshop, July 23, 2014 AACR Annual Meeting, April 21, 2015 Seminars at FNLCR

RAS Structures Workshop, July 21-22, 2015 Channing Der, UNC RAS Immunotherapy Workshop, November 3, 2015 Ken Westover, UTSW RAS Immunotherapy Workshop SpeakersCarla Mattos Include:, Northeastern RAS Symposium, December 15-16, 2015 Mark Philips, NYU Vadim Gaponenko, U-Chicago RAS Symposium Confirmed Speakers Include:Jay Berzofsky , NCI Josh Salafsky, Biodesy, Inc. Jamie Chaft, Sloan-Kettering Calvin Kuo, Stanford Harold Varmus, Weill Cornell Medical College Kris Wood, Duke Lisa Coussens, Oregon Health SciencesMariano Barbacid, CNIO, Madrid Kevan Shokat, University of California, San Francisco Tom Dubensky, Aduro Biotech Cyril Benes, Mass General Allan Balmain, University of California, San Francisco Victor Engelhard, University of VirginiaCarolyn Buser, GlaxoSmithKline Mariano Barbacid, Spanish National Cancer Research Center (CNIO), Madrid Jay Groves, UC-Berkeley , Oslo University Hospital James Bradner, Dana Farber, Harvard UniversityGustav Gaudernack Stephen Sligar, UI-Champagne Urbana Karen Cichowski, Brigham and Women's Hospital,Liz Jaffe Harvard, Johns HopkinsUniversity Raffit Hassan, NCI Renata Grifantini, Externautics Spa, Siena Channing Der, University of North Carolina,Ira Chapel Mellman Hill , Genentech Renata Pasqualini, U-New Mexico Stephen Fesik, Vanderbilt University Isabelle Riviere, Sloan-Kettering Andrew Bradbury, Los Alamos Jay Groves, University of California, BerkleySteve Rosenberg, NCI Kent Rossman, UNC John Hancock, University of Texas, HoustonUgur Sahin, Translational Oncology, Mainz,Shiva Malek Germany, Genentech Frank McCormick, University of California,Bob San SchreiberFrancisco ,and Washington the RAS InitiativeUniversity Deborah Morrison, National Cancer InstituteBob Vonderheide, University of Pennsylvania Mark Philips, New York University David Sabatini, Whitehead Institute, Massachusetts Institute of Technology Kevin Shannon, University of California, San Francisco David Tuveson, Cold Spring Harbor Laboratory Michael White, University of Texas, Southwestern Matthew Vander Heiden, Koch Institute, Massachusetts Institute of Technology Collaboration with the RAS Community

Outside collaborators RAS Reference Reagents Steve Almo, Einstein Chris Kemp, Fred Hutch Jim Wells, USCF Eric Chang, Baylor Channing Der, UNC Silvia Thöne, Munich Ken Westover, UTSW Peter Jackson, Stanford University Carla Mattos, Northeastern Tyler Jacks, MIT Steve Sligar, U- Ill Calvin Kuo, Stanford Jay Groves, Berkeley Bill Hahn, Broad / Dana Farber Hirsch Nanda, Susan Kreuger, NIST Karla Satchell, Northwestern John Markley, NMRFAM, UW-Madison Julian Downward, Cancer Research UK Paul Cohen, DARPA Daniel Abankwas, University of Turku Kris Wood, Duke Said Sebti, Moffitt Cancer Center David Weber, U-Maryland Ian Prior, Liverpool Tina Yuan, Broad Muller Fabbri, Children’s Hospital LA Cameron Pitt, UCSF Faraz Bishehsari, Rush Krishna Kota, USAMRIID Amy Lee, USC Sotirios Koutsopoulos, MIT Yosef Yarden, Wiezmann Fred Wittinghofer, Dortmund University Richard Klemke, UCSD Lynn McGregor, UCSF (PanCan postdoc) Saidul Chowdhury, U-Texas Arlington John Hunter, UTSW (PanCan postdoc) Christian Gocke, JHMI Saori Sato, Daiichi-Sankyo Tobias Baumgart, U-Penn Walter Englaro, Sanofi-Aventis Emil Lou, U-Minnesota NIH collaborators Kirk Staschke, Lilly Ron Bose, Wash U Ji Luo, NCI Gad Getz, Mass Gen /Broad Neil Kelleher, Northwestern Anton Simeonov, NCATS Matt Meyerson, Dana Farber Sourav Bandyopadhyay, UCSF Debbie Morrison, NCI Immuno-MRM of RAS pathway Robert Chapkin, Texas A&M Rajat Varma, NIAID • Amanda Paulovich, Fred Hutch Udo Rudloff, NCI • Steve Carr, Broad Institute Sriram Subramaniam, NCI • John Koomen, Moffit Cancer Center Andreas Gosberg, Lilly 33 FNLAC RAS Working Group Recommendations

34 Science • Pursue a detailed understanding of Processed KRAS4b – Unique reagent that shows promise as a new tool compound and thus illustrates the potential of the RAS Initiative – Biophysical and structural analysis of RAS on membranes – Dedicate staff to make processed KRAS4b for research community – In-house workshops for reagent generation

• Biochemistry and structural biology (e.g., Cryo-EM) of RAS complexes and RAS:membrane interactions is a priority – Collaborative effort with CCR and Sriram Subramaniam

• The overall set of structures and reagents should represent a concrete and useful set of deliverables from the RAS effort • Other efforts were more exploratory and open-ended; some are being scaled back or phased out 35

Strategy

• Develop a plan to augment industry partnership (see next slide) • Implement a framework for tool/lead compound development – Need path to validation and optimization not dependent on pharma – Define and plan for medicinal chemistry needs • De-convolution assays might emerge as an area of specialty – Synergize biophysics/biochemistry with assays and screens – Develop an assay cascade for validation of hits/leads • Step-up awareness and dissemination efforts – Package reagents, assays and capabilities for presentation to academia and pharma • Publicity/marketing – Develop additional next-generation assays • Develop plan for renewal phase 36 – Present to FNLAC as part of renewal

Interactions with Pharma

• Pharma participation is a big plus – Pharma brings credibility and resources – Roadshows and marketing to increase participation

• Be creative when thinking about partnering possibilities – Preferred partner(s) – Separate company that holds IP? – Venture philanthropy? – Develop strategy for prosecuting IP

• Blueprint for bringing all projects to successful completion – Define metrics for success up-front – Framework for division of labor during follow-up phase 37

Outreach & Resources • Websites – “Interactome” that engages the community may have valueProvide additional information • Protocols ; cell line mutation and RAS dependence

• Compound collection and reagent distribution – Resource for internal efforts and extended RAS community – Level of effort and source of compounds – Validate compounds with “assay cascade” – Track distribution and the experience/coaching needs of the recipients

• Possibilities – Manuscript on test compounds with assay cascade? – De-bunk inaccurate claims? – RAS pathway proteomic studies? 38