University of Pittsburgh Drug Discovery Institute The Role of - - PowerPoint PPT Presentation

University of Pittsburgh Drug Discovery Institute

Fifth Annual Ri.MED Scientific Symposium

October 24, 2011

11/4/2011 1 Novel Chemistries and Systems Biology Power Discovery

• D. Lansing Taylor, Ph.D.

Director Professor of Computational and Systems Biology

The Role of Systems Biology in Drug Discovery

Agenda

• 1. Introduction to the Challenge/Opportunity in

Academic Drug Discovery

• 2. Overview of the UPDDI
• 3. Cellular Systems Biology Program (CSBP)
• 4. Platform for Protein-Protein Interactions

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 2

Target identification and validation Lead identification and optimization Lab and animal testing Clinical trials

Traditional Steps in Drug Discovery and Development at Pharma: Why is it Not Working Well?

Novel Chemistries and Systems Biology Power Discovery 3

Pre-discovery Discovery Preclinical studies Humans

Up to 15 years and $ 1 billion High attrition rates Very low success rates

Chemical libraries/Biologics

What is the State of the Drug Discovery Industry?

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 4

In the Last 15 Years

• Cost of discovering a drug up more than 270%
• Number of NME’s approved down more than 50%
• Major revenue generators (block busters) going off patent

How has the Pharmaceutical Industry Responded?

• Merging with other pharma’s to gain short-term pipeline
• Decreasing costs by laying off tens of thousands of researchers
• Stopping research and development (R&D) in some therapeutic areas
• Shifting more R&D to China and India
• Exploring Personalized Medicine-drug candidate with diagnostic
• Developing more collaborations with academia

Some Thoughts On How To Be Successful in Academic Drug Discovery

• 1. Understand what other academic discovery programs are doing
• 2. Identify and then Integrate strengths across the Institution and key

partners

• 3. Create a “marketing” program to educate industry on capabilities
• 4. Establish a pharmaceutical industry collaboration program

(e.g.Italian Institute of Technology’s D3)

• 5. Engage industry involved in personalized medicine
• 6. Select some initial therapeutic focus areas for internal discovery

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 5

Some Existing Pharmaceutical Company Collaborations at the University of Pittsburgh

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 6

Pharmaceutical Company

Therapeutic Area Johnson & Johnson Cancer Janssen Biotech (Centocor) Asthma Janssen Biotech (Centocor) Scleroderma Janssen Biotech (Centocor) COPD Novartis Alpha-1 antitrypsin deficiency Hawthorn Pharmaceuticals Cancer GE Healthcare Alzheimer's Disease Abbott Necrotizing enterocolitis Arno Therapeutics Cancer

Examples of Successful “Academic” Drugs: This Can Be Done!

• Remicade (Infleximab)-tumor necrosis factor α

Jan Vilcek and Junming Le New York University-Centocor, Inc.

• NYU has received $650M in royalties
• Paclitaxel (Taxol)-MT stabilizer (mitotic inhibitor)

Robert Holton, Chemist-total synthesis Florida State-Bristol Myers Squibb

• Florida State has received $350M in royalties

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 7

Example: Drug Discovered in Italy

Biphosphates-Osteoporosis Giorio Staibano and Sergio Rosini Instituto Gentili Research Laboratories in Pisa

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 8

Agenda

• 1. Introduction to the Challenge/Opportunity in

Academic Drug Discovery

• 2. Overview of the UPDDI
• 3. Cellular Systems Biology Program (CSBP)
• 4. Platform for Protein-Protein Interactions

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 9

Drug Discovery in BST-3 at the University of Pittsburgh

University of Pittsburgh Biomedical Science Tower 3

4 Novel Chemistries and Systems Biology Power Discovery

1 1

Leadership of the UPDDI

Director Lans Taylor Associate Director Chemistry Peter Wipf Associate Director Med-Chemistry Barry Gold Associate Director Comp & Sys Biol Ivet Bahar Associate Director

Cancer Institute

• Dept. Medicine

Edward, Chu, MD

Novel Chemistries and Systems Biology Power Discovery

Distributed Structure Of UPDDI

Novel Chemistries and Systems Biology Power Discovery 12

Technology Developments & Corporate Collaborations & Licensing Drug Discovery & Development Core Screening Support for Collaborative Development Pre-Clinical Studies

Formulations & Delivery Animal Tox & Pharmacokinetics Animal Efficacy

Educational Programs Focused Discovery and Development Portfolio In Vitro Safety & Metabolism Biologics Chemistry & Medicinal Chemistry Focused Discovery & Development Projects

Novel Chemistries and Systems Biology Power Discovery

We have the Capabilities: Prominent Centers, Institutes & Departments

Initial Focus Areas

• Selected two therapeutic areas for initial focus where we

have great strengths, while supporting any therapeutic area brought to us by faculty or industry

• Cancer
• Neurological Diseases
• Selected three technical areas for initial focus where we

have unique strengths, while harnessing all technologies

• Novel Chemistries including Biologics
• Computational Chemistry (Includes Structural Biology)
• Computational Biology and Systems Biology

Novel Chemistries and Systems Biology Power Discovery 14

Novel Chemistries from Pitt: Two Examples from Multiple Faculty in A&S, SOM & SOPharm

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 15

Combinatorial Chemistry Center Chemical Diversity Center

Peter Wipf, Ph. D. Distinguished University Professor of Chemistry Phase 2 Clinical trials (Glioblasoma multiforme and Prostate) -PI3 Kinase Edward Chu, M.D. Professor of Medicine Chief, Division of Hematology/Oncology Deputy Director of UPCI

Chinese Herbal Medicines PHY906

• Combination with cytotoxic chemotherapy

in metastatic colorectal cancer

• Phase 1-2

Computational Chemistry/ Biology & Systems Biology

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 16

Ivet Bahar, Ph. D. Chair, Department of Computational and Systems Biology

Mapping of complete set of FDA approved drugs and their targets

Extracted from DrugBank (Sept 2010).

Predicting protein interaction dynamics using elastic network models1 (1) Bahar et al (2010) Annual Rev Biophys 39, 23-42.; (2) Liu, Eyal & Bahar (2008) Bioinformatics 24, 1243-50.

Novel Chemistries and Systems Biology Power Discovery 1 7

To build an Institute that applies novel and traditional technologies to

• ptimize

the discovery and development of new molecular entities (NME’s) through integrated activities across departments, institutes and commercial partners, while advancing the science and technology of drug discovery

Vision

Agenda

• 1. Introduction to the Challenge/Opportunity in

Academic Drug Discovery

• 2. Overview of the UPDDI
• 3. Cellular Systems Biology Program (CSBP)
• 4. Platform for Protein-Protein Interactions

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 18

 -------- 25 Å ------ 

Bridge between Molecular and Cellular Dynamics and Physiology: Computational and Systems Biology

Lei Yang, Ph.D. Cecilia Lo, Ph.D Neil Hukreide, Ph.D. Andreas Vogt, Ph.D. Gary Silverman, M.D., Ph.D. David Perlmutter, M.D. Bert Gough, Ph.D., Tim Lezon, Ph.D. Stephen Thorne, Ph.D. Novel Chemistries and Systems Biology Power Discovery

Addressing the Biological Complexities Center for Biologic Imaging

Heterogeneity of Response to Therapeutics in Cell Populations: Treatment of Tumors

Novel Chemistries and Systems Biology Power Discovery 20

Bert Gough, Ph.D. Tim Lezon, Ph.D. Lans Taylor, Ph.D. & Jennifer Grandis, M.D., FACS Tobacco Grant Funding Cancer Cell Lines TME Patient Samples

Heterogeneity of Drug Responses in Tumors and Pathway Modulations

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 21

Heterogeneity

in a cell population

Extrinsic Non-genetic clonal population Genetic

Population noise Temporal noise

Intrinsic Macroheterogeneity Microheterogeneity

Adapted from Huang (2009) Development 136: 3853

cell parameter 1 Cell count Microheterogeneity Macroheterogeneity

21

Platform for Cellular Systems Biology

Novel Chemistries and Systems Biology Power Discovery 22

Cell Data Database Store Database Sample Database

Protocol Development Sample Prep Data Process Data Mining/Visualization Computational & Systems Biology InCell 6000 Data Acquisition 6 -384 Well Plates

Server Room Wireless Classroom Laptops (23) Miscellaneous Laptops (20+)

Workstation VLAN

Internet

802.11g Wireless Access Points (4)

PITTnet Network Printers, Copiers, Scanners Desktops / Workstations (90+) Gig-E Gig-E VMware ESXi hosts (3) Promise Vtrak (11TB) 4Gbps FC optical Misc Servers Each Cluster has login node and dedicated Gig-E private network

Cluster 1 864 CPU cores 1388 GB ram 75 nodes on Infiniband Cluster 2 46 CPU cores 124 GB ram Cluster 3 124 CPU cores 132 GB ram Cluster 4 40 CPU cores 40 GB ram DMZ VLAN

Novel Chemistries and Systems Biology Power Discovery 23

384 Well Plates Single Well 4x 10,000 cells Higher Magnification Multiple Biomarkers

High Content Analysis

High Throughput, Hyperplexed Imaging to Identify Heterogeneous Subpopulations

Novel Chemistries and Systems Biology Power Discovery 24

Cell 1 Cell 2 Cell N Dose Assay Plates Hyperplexed Images

TOR1 cAMP PKA RPP1A UTH1 BMH1 HFD1 CMD1 cytoskeleton ARP2/3 ARC15 FPR1 rapamycin Calcineurin Mitochondria Nutrients Nucleus

Profiles of Subpopulations Pathway Identification

IL-6

STAT3 Pathway

MLKs TYK2 JAK2 MKK1/2 Rac GTP P P P P BCL2

Anti-Apoptosis Cell Growth, Survival, Differentiation and Oncogenesis

P P WAF1 Growth Arrest and Progression G1 to S Cell Cycle Progression Gene Expression P P P STAT3 STAT3 P ISRE P MKKs Raf1 ERK1/2 P Src Pim1 p38 JNK SOCS STAT3 STAT3 c-Myc CDC25A STAT3 STAT3 P P P P P STAT1 P P STAT1 STAT1 P P

Anti-tumor Inflammatory Response

I F N g R 1 I F N g R 2

JAK2 STAT1 P NF-κB

STAT1 Pathway

IKK IkBs NF-κB IkBs P

G P 1 3

25

11 Cell Feature STAT 3 Heterogeneity & Pathway Analysis in Cal-33 Cancer Line (HNSCC)

Cell Features Cell Number Nuclear Size DNA Content/ Cell Cycle DNA Texture Mitochondrial Membrane Potential STAT3 Activation STAT1 Activation NFkB Activation ERK MT Stability Apoptosis Live Readout 1st panel 2nd panel

26 Novel Chemistries and Systems Biology Power Discovery

Novel Chemistries and Systems Biology Power Discovery 27

Heterogeneity in STAT3 and STAT1 Activation by IL-6 and IFNg

0.00E+00 5.00E+05 1.00E+06 1.50E+06 2.00E+06 2.50E+06 3.00E+06 50000 100000 150000 200000 250000 300000 total Nuclear Intensity pSTAT3 (TotalCircAvgInten)

Unstimulated Cells - 30 min

0.00E+00 5.00E+05 1.00E+06 1.50E+06 2.00E+06 2.50E+06 3.00E+06 50000 100000 150000 200000 250000 300000 total Nuclear Intensity pSTAT3 (TotalCircAvgInten)

Max Stim IL-6 - 30 min

0.00E+00 5.00E+05 1.00E+06 1.50E+06 2.00E+06 2.50E+06 3.00E+06 50000 100000 150000 200000 250000 300000 total Nuclear Intensity pSTAT1 (TotalCircAvgInten)

Unstimulated Cells - 60 min

0.00E+00 5.00E+05 1.00E+06 1.50E+06 2.00E+06 2.50E+06 3.00E+06 50000 100000 150000 200000 250000 300000 total Nuclear Intensity pSTAT1 (TotalCircAvgInten)

Max Stim IFN-g - 60 min

STAT3 Activation with IL-6 STAT1 Activation with IFNg DNA Content DNA Content DNA Content DNA Content

pSTAT3 Activity pSTAT3 Activity pSTAT1 Activity pSTAT1 Activity

pSTAT3 active In some cells pSTAT1 Active Apoptotic? pSTAT3 Active

• Proliferation

pSTAT1 Active No Response

Statistical Networks are Calculated from Feature Data

Novel Chemistries and Systems Biology Power Discovery 28 Density Energy

Probability density is used to determine an energy function Multivariate distribution for data defines a probability density Effective interactions between features define statistical network that accounts for system’s behavior

pH3. Microtubule Stability (MS) Nuclear Cond (NC) Nuclear Area DNA Content var(pH3) var(NC) var(MS)

Clinical Impact of Understanding Heterogeneity: Personalized Medicine

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 29

Patient Sample Genomic Tests Proteomics Tests Hyperplexed Pathology Optimal Treatments & Clinical Trails Bioinformatics/ Systems Biology

Agenda

• 1. Introduction to the Challenge/Opportunity in

Academic Drug Discovery

• 2. Overview of the UPDDI
• 3. Cellular Systems Biology Program (CSBP)
• 4. Platform for Protein-Protein Interactions

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 30

Computational Chemistry and Systems Biology Approach to Finding Inhibitors of P-P Interactions

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 31

David Koes Alexander Dömling Carlos Camacho

Protein-Protein Interactions as Targets

• PPIs are involved in all biochemical and cell biological events in cells
• PPIs comprise a rather novel target class for pharmaceutical interventions
• Orders of magnitude more PPIs are known than “druggable proteins”
• The complete interaction map of proteins during the lifetime of an
• rganism is called the “interactome”
• Compounds interacting with PPIs are currently discovered by HTS
• Jacoby (NOVARTIS) recently analyzed the HTS screening success of

Large Pharma libraries for PPIs (<<1%)

• lack of deep pockets, large contact surfaces, “unsuitable” compounds in libraries
• Atomic details of the binding site may provide clues for “druggability”:
• Pocket size and shape
• Hydrophobic - hydrophilic

Goal: Enable Rational Design, Synthesis and Testing of Novel P-P Interactions

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 33

• Expand chemical space
• Design suitable chemical libraries that can increase hit rates
• Implement a truly interactive virtual screening technology
• Utilize Cell-based biosensors for functional testing
• Short time between design, synthesis and testing of leads
• Synergy of chemists, biologists and experts on specific protein

interaction pairs

Key Points

• P-P Interactions often mediated by only a few key amino acids

“Hot Spots” or “Anchor” Key is how deeply buried!

• Anchor amino acid side chains might serve as a reasonable

starting point for the design of antagonists of the P-P interaction

• Particular amino acid side chain as an initial anchor for

screening virtual libraries of low-molecular weight scaffolds

• Multicomponent reactions (MCR) allow assembly of many

diverse and complex scaffolds

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 34

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 35

Selection of Disease Relevant P-P Interactions from PDB Select Optimal P-P Pairs Based on Ligand Chemistry

Build Biosensor and Profile Early Safety Assessment

Design and synthesis of Compounds: MCR Validation: Crystallographic SPR/FP Cell based Pre-Clinical Efficacy & Safety Virtual screening: MCR Biased to Anchor

I II III IV V

Steps to Identify Inhibitors of Protein-Protein Interactions: New Thinking

p53 p53 p53 HDM2

P P

HDM2

P

p53 p53

P

Xx

P

Therapeutic Target: Maintain Elevated p53 by Inhibiting p53-Mdm2 Interactions

Nutlin-3

Novel Chemistries and Systems Biology Power Discovery

p53-hDM2 Biosensor Design

p53 hdm2 NLS NES/NLS TagRFP Nuclear – Cytoplasm Shuttling Component Nuclear Anchored Component

Co-Express (Adenovirus)

Selective Disruption

Novel Chemistries and Systems Biology Power Discovery 37

Giuliano, Premkumar and Taylor, 2006

p53-hDM2 Biosensor: Nutlin-3 Control

Novel Chemistries and Systems Biology Power Discovery 38

Nutlin 3

M 1x10

• 7

1x10

• 5

Biosensor Activity 0.2 0.5 0.8 1.1 1.4 1.7

1.1 M

Nutlin-3 control

Plate 1 - Min/Max

• 50

50 100 150 200 250 300 350 400 24 48 72 96 120 144 168 192 216 240 264 288 312 336 360 384 Min1 Max1 Mean +3sdev

• 3sdev

Mean +3sdev

• 3sdev

Plate 2 - Min/Max

• 100

100 200 300 400 500 600

24 48 72 96 120 144 168 192 216 240 264 288 312 336 360 384 Min2 Max2 Mean +3sdev

• 3sdev

Mean +3sdev

• 3sdev

Z’=0.72 Z’=0.68 Validation plates showing consistency and reproducibility of the response

• f the adenovirus-delivered p53:hdm2 biosensor to Nutlin3 challenge

Quantitative Profiling

+ Nutlin 3

Novel Chemistries and Systems Biology Power Discovery 39

Cellular Systems Biology Profile Integrates Biosensor Activity with Pathway and Off-Target Effects

40 Novel Chemistries and Systems Biology Power Discovery

Three p53-Mdm2 Inhibitor Leads in Pre-Clinical Testing

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 41

Characteristic Pitt Leads Nutlin Derivative Amgen J&J Mich.

Ki <100 nM

++ ++

• +++

Biosensor Potency ++ ++ NA NA NA NCI 60 Growth Inhibition +++ ++ NA NA NA Ease of Synthesis +++

• cLogP

+++ ++ +

• +++

Water Solubility +++

• NA

NA +

Several patent families 3 leads from over 700 synthesized compounds based on 10 different small MW scaffolds

Novel Chemistries/ Computational Chemistry & Computational and Systems Biology

Novel Chemistries and Systems Biology Power Discovery 4 2

….

Computational Chemistry/Biology and Systems Biology

Target Molecules Cells Tissues/Organs Human Animals Novel Therapeutic Molecules

11/4/2011 Novel Chemistries and Systems Biology Power Discovery 43

Grazie molto!