Pharmacophore Guided

Fragment-Based Drug
Design

Tien Luu, PhD

Lead Scientific Specialist, Life Science
tluu@accelrys.com

Discovery Studio 2.1: New Science and Customized Workflows for

Drug Discovery Research
10th July 2008

This presentation and/or any related documents contains statements regarding our plans or expectations for future features, enhancements or functionalities of current or future products
(collectively "Enhancements"). Our plans or expectations are subject to change at any time at our discretion. Accordingly, Accelrys is making no representation, undertaking no commitment or
legal obligation to create, develop or license any product or Enhancements. The presentation, documents or any related statements are not intended to, nor shall, create any legal obligation
upon Accelrys, and shall not be relied upon in purchasing any product. Any such obligation shall only result from a written agreement executed by both parties. In addition, information
disclosed in this presentation and related documents, whether oral or written, is confidential or proprietary information of Accelrys. It shall be used only for the purpose of furthering our
business relationship, and shall not be disclosed to third parties.
Fragment-Based Design

• Combination of small fragments results in high diversity

• Addressing sub-site specificity and ligand efficiency
• Low molecular weight compounds represent suitable basis for
optimisation of ADME properties

S. Borman, Chemical & Engineering News, June 19, 2006, Volume 84, 25, pp.
56-78; November 28, 2005, Volume 83, 48, pp. 28–30

© 2008 Accelrys, Inc. 2

Why Pharmacophores?

• Smaller fragments may bind in various parts of the binding cavity

• From the way a fragment docks, it may be difficult to discover how
the ligand containing that fragment may dock
• The pose of the fragment in a ligand may not be a top hit when the
fragment is docked
• When two or more fragments are joined properly, the chances of
finding the ligand pose seems greater
• Pharmacophores can direct the fragment to the location occupied
when the fragments form a ligand

Babaogly, K, Shoichet, B, Deconstructing fragment-based drug discovery,

Nature Chem. Biol., 2 (12), 2006

© 2008 Accelrys, Inc. 3

Fragment-Based Workflow

Protein Target File

Fragment Queries
Reference Query Fragment
Database

Hit Lists

Library
Enumeration

De-novo Library
Reference Query
Screen / Focus Scoring &
Database Prioritisation

Focused Hit List

Hit Refinement

Synthetic Feasibility Filtering

Refined Hit List
© 2008 Accelrys, Inc. 4
Example: Thymidine Kinase

• Fragment based approach applied to thymidine kinase

1. Protein preparation
2. Pharmacophore construction and link definitions
3. Enumeration and hit refinement
4. Selection of candidate molecules

© 2008 Accelrys, Inc. 5

The Target

• Example: HSV Thymidine Kinase (1kim.pdb)

© 2008 Accelrys, Inc. 6

Protein Preparation

• Fully automated protocol dealing with protein preparation:

1. Cleans the protein/ligand complex
– All Hydrogen are added
– Alternate residue conformation are removed
– Names are standardized
– Missing residues completed
2. CHARMm pKa calculation to protonate the protein

http://forums.accelrys.org/eve/forums/a/tpc/f/2011007181/m/3601080603
© 2008 Accelrys, Inc. 7
Describing the Interactions

• 3HBD, 1HBA, 1Hydrophobe

© 2008 Accelrys, Inc. 8

Fragment Queries

• Two options explored

– Difference in the resulting molecules expected
– The link atom queries need to be formulated differently

© 2008 Accelrys, Inc. 9

Fragment Linking

• Link atom does not have to be of the same type

– Sometimes a different atom type in each fragment is required for
enumeration

H
O N O H2N O

NH OH
H 3C
H OH
O N O H

N O
H3C
H OH
H
OH O N O
O

N
H 3C CH 3 OH

H OH
© 2008 Accelrys, Inc. 10
Fragment Queries

• Link1 and Link2

Shape parameters
can be tuned

Strategy 1 Strategy 2
© 2008 Accelrys, Inc. 11
Fragment Database

• Commercial, corporate
• Customised

– 120000 compounds
– 20000 fragments
– Conversion into a multi-conf database

© 2008 Accelrys, Inc. 12

Synthetic Feasibility Score

• Molecular Networks (www.molecular-

networks.com)
• ISV partner (Pipeline Pilot Components)

• Fast estimation of synthetic accessibility of

organic compounds
– Scoring of compounds on a scale from 1 and 10
• 1 - Easy to synthesise
• 10 - Difficult to synthesise

• Prioritisation of chemical compounds

– Generated by de novo design experiments
– Generated by inverse QSAR/QSPR experiments
– Large virtual compound libraries

Boda, K. et al. Structure and reaction based evaluation of synthetic accessibility. J. Comput.-Aided Mol.
Des.
© 2008 2007,
Accelrys, Inc. 21, 311-325. 13
SYLVIA – Components of Accessibility Scoring

• Structure-based
– Molecular graph complexity 1

– Ring complexity 2
– Stereochemical complexity
• Starting material-based
– Similarity to starting materials
• Product bond-based
– Reaction fitness: based on presence of product reaction center substructures
(RCSS) extracted from reaction databases

1 Bertz, S.H. J. Am. Chem. Soc. 1981, 103, 3599-3601.

2 Gasteiger, J.; Jochum, C. J. Chem. Inf. Comput. Sci. 1979, 19, 43-48.
© 2008 Accelrys, Inc. 14
Fragment Screening

• Customised Database
– 20000 compounds
– Synthetic feasibility filtering
(threshold: 4)
– Multiple searches done simultaneously

Hit List From Fragment-library

Query1 Query2
Strategy1 29 31
Strategy2 116 61

© 2008 Accelrys, Inc. 15

Enumerating and Re-Focusing the de-novo Library

Energy or structural filters

are applied

Number of successful
molecules is not
the simple product of
number of fragments
Enumerated Library Focused Library
Strategy1 159 (899) 146
Strategy2 704 (7076) 388
© 2008 Accelrys, Inc. 16
Hit Refinement

• Focused lists are just a starting point for additional refinement

• Refinement is a Multiple Step procedure
• Is project dependent
• In this example, focused hit lists are:
– Docked into the active site of thymidine kinase (CDDOCKER)
• 2-deoxythymidine is used for comparison
– Filter the poses with the pharmacophore
– Evaluated against a library of pharmacophores built from kinase targets –
pharmacological profiling

Pharmacophore-based Pharmacological
CDOCKER
© 2008 Accelrys, Inc. pose filtering profiling 17
Hit Refinement – Results: Strategy 1

• DeoxyThymidine CDOCKER energy: 25.05 – 21.96

• For 57 compounds, best pose CDOCKER energy > best pose CDOCKER
energy of DeoxyThymidine

© 2008 Accelrys, Inc. 18

Hit Refinement – Results: Strategy 1 continued

• Filtering CDOCKER poses with the pharmacophore

• 24 compounds

© 2008 Accelrys, Inc. 19

Hit Refinement – Results: Strategy 1 continued

Profile of 163584

1
0.9
0.8
0.7
0.6
0.5
0.4
0.3 Scaled Fit
0.2
0.1
0
TK Bcr-Abl CDK2 CDK5 Hck
(HSV1)
Kinase target

Profile of 164538

1
0.9
0.8
0.7
0.6
0.5
0.4
0.3 Scaled Fit
0.2
0.1
0
TK PDK1 CDK2 CDK5 Pim1
(HSV1)
Kinase target

Profile of 224933

1
0.9
0.8
0.7
0.6
0.5
0.4 Scaled Fit
0.3
0.2
0.1
0
TK (HSV1) CDK2
Kinase target

Profile of 238894

1
0.9
0.8
0.7
0.6
0.5
0.4 Scaled Fit
0.3
0.2
0.1
0
TK (HSV1) CDK2 CDK5
Kinase Target

© 2008 Accelrys, Inc. 20

Hit Refinement – Results: Strategy 2

• DeoxyThymidine CDOCKER energy: 25.05 – 21.96

• For 180 compounds, best pose CDOCKER energy > best pose CDOCKER
energy of DeoxyThymidine

© 2008 Accelrys, Inc. 21

Hit Refinement – Results: Strategy 2 continued

• Filtering CDOCKER poses with the pharmacophore

• 131 compounds

© 2008 Accelrys, Inc. 22

Hit Refinement – Results: Strategy 2 continued

Profile of 109397-17414 Profile of 109397-148362

1 1
0.9 0.9
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 Scaled Fit
0.4 Scaled Fit
0.3
0.3
0.2
0.2
0.1
0.1
0
0 TK (HSV1) CDK2 c-Met
TK (HSV1) CDK2
Kinase Target
Kinase Target

Profile of 109397-75086
Profile of 109397-254669

1
1
0.9 0.9
0.8 0.8
0.7 0.7
0.6 0.6
0.5 0.5
0.4 Scaled Fit 0.4 Scaled Fit
0.3 0.3
0.2 0.2
0.1
0.1
0
0 TK (HSV1) CLK1
TK (HSV1) CLK1
Kinase Target
Kinase Target

© 2008 Accelrys, Inc. 23

Conclusion

• Easy to set and fast procedure

• Used here for lead hopping, but can also be used for Lead
optimization (scaffold + pharmacophore)
• Focused hit list is only the starting point for refinement
• Synthetic feasibility score used to filter the compounds during
enumeration phase

© 2008 Accelrys, Inc. 24

Aknowledgements

• Remy Hoffmann
• Konstantin Poptodorov
• Catalyst Development Team
– Jon Sutter
– Al Maynard
– Jiabo Li
– Roman Kuchkuda
– Christoph Koelmel

© 2008 Accelrys, Inc. 25

Thank You

• Thank You for attending today’s webinar. If you have any further
questions please e-mail me at tluu@accelrys.com

• You can also contact us using the form on our website:

http://accelrys.com/company/contact/

• We will be exhibiting at the following upcoming events:

– CHI Protein Kinase Targets (June 23 – 25, Boston, Booth #4)
– CHI Structure Based Design (June 25 – 27, Boston, Booth #7)
– Drug Discovery Technology and Development (August 4 – 7, Boston, Booth
#512)
– ACS Fall 2008 (August 17 – 21, Philadelphia, Booth #211)

• Reminder: The next webinar in this series will be:

– Advances in Pharmacophore Modeling and Parallel Screening (Ragi Raghavan)
– July 17, 2008 – 3pm GMT (7am PST) and 10am PST
© 2008 Accelrys, Inc. 26
Discovery Studio 2.1: New Science and Customized
Workflows for Drug Discovery Research

• Advances in Protein Modeling and Protein-Protein Interactions in Discovery

Studio
– Francisco Hernandez-Guzman – June 12, 2008

• Physics Based Protein Ionization and pK Estimation

– Francisco Hernandez-Guzman – June 19, 2008
• Towards Increased Accuracy in Computational Drug Discovery with QM/MM
– Dipesh Risal – June 26, 2008

• Pharmacophore Guided Fragment-Based Drug Design

– Tien Luu – July 10, 2008

• Advances in Pharmacophore Modeling and Parallel Screening

– Ragi Raghavan – July 17, 2008

• Workflow Customization with DS Developer Client

– Luke Fisher – July 24, 2008
• Tips and Tricks in using Discovery Studio
– Al Maynard – July 31, 2008

© 2008 Accelrys, Inc. 27