Design, Development and Application of a Novel In-Silico High-throughput Screening Approach 公开

Shi, Qi (2012)

Permanent URL: https://etd.library.emory.edu/concern/etds/7d278t87f?locale=zh

Published

Abstract

Computer-assisted molecular design is currently a routine component of medicinal chemistry and drug discovery. In this thesis, a novel in-silico high-throughput screening approach is developed and applied to two drug discovery projects. The approach, termed FRESH (Fragment-based Exploitation of modular Synthesis by vHTS), is designed to balance multiple factors in addition to potency, namely synthetic feasibility, octane-water partition coefficient, cell-permeability, blood-brain-barrier penetration, metabolism and toxicity. The method combines rapid vHTS (virtual High Throughput Screening), pharmacological property prioritizing and 3D QSAR (Quantitative Structure-activity Relationship) construction. In the first applied project, the approach has provided an initial list of potentially selective NET/SERT (Norepinephrine Transporter/Serotonin Transporter) antagonists with different potency ratios based on existing homology models. In the second application project, the approach has provided a group of novel KCN1 analogs with predicted improvements in both potency and pharmacological properties by employing a two-site binding model for the p300/HIF-1α interaction. The third project on curcumin analogs is still attempting to build a suitable QSAR model. However, once the latter is developed, FRESH can be applied to assist identifying novel curcumin analogs as well.

Design, Development and Application of a Novel In-Silico High-throughput Screening Approach

B.S., Peking University, 2008
Advisor:
Dr. James P. Snyder
Dr. Dennis C. Liotta
A thesis submitted to the faculty of the
James T. Laney School of Graduate Studies of Emory University
in partial fulfillment of the requirements for the degree of
Master of Science in Chemistry
2012

Table of Contents
Chapter 1: Introduction ............................................................... 1

1.1. Requirements for a molecule to qualify as a drug ...............................1
1.1.1 Chemical space .................................................................................................. 1
1.1.2. Drug likeness .................................................................................................... 2
1.1.3. CNS drug likeness............................................................................................. 3
1.2. Computer assisted drug discovery .......................................................4
1.2.1. In-silico estimation of the ligand-protein interaction........................................ 5
1.2.2. In-silico estimation of physical/ADMET properties......................................... 7
1.3. Challenges in the lead optimization step .............................................8
1.3.1. Multi-target/site therapies ................................................................................. 8
1.3.2. Balancing multiple factors for multi-target drugs........................................... 10

Chapter 2: Design and development of a novel high-
throughput virtual screening approach for MTL.................... 11
2.1. Overall design strategy ......................................................................11
2.2. Screening software program selection...............................................11
2.3. Algorithm design and optimization ...................................................12

Chapter 3: Application I: Composing novel SNRIs................. 16
3.1. Project background............................................................................16
3.2. Challenges in the lead optimization step ...........................................17
3.3. Homology models of NET and SERT ...............................................19
3.4. Application of the FRESH approach .................................................22
3.4.1. Protocol design for the FRESH approach....................................................... 22
3.4.2. Resulting structures......................................................................................... 25

Chapter 4: Application II: Identification of novel KCN1
analogs to block the p300/KCN1 interaction............................ 26
4.1. Project background............................................................................26
4.2. Challenge in the lead optimization step .............................................27
4.3. Two-site modeling of KCN1 for the p300 transcription factor..........27
4.3.1. Binding receptor selection .............................................................................. 27
4.3.2. Binding site selection...................................................................................... 30
4.3.3. Linear regression attempts for KCN1 analogs................................................ 33
4.3.4. Evaluation of calculated energy values by ROC ............................................ 36
4.4. Application of the FRESH approach .................................................37
4.4.1. Protocol design for the FRESH approach....................................................... 38
4.4.2. Resulting structures......................................................................................... 40

Chapter 5: Curcumin analogs as Pleiotropic Kinase Blockers43

5.1. Project background............................................................................43
5.2. Modeling of curcumin analog against several kinases.......................43
5.2.1. Modeling of AKT-1 and AKT-2..................................................................... 44
5.2.2. Modeling of VEGFR2..................................................................................... 51
5.3. Anticipated FRESH analysis .............................................................52

Chapter 6: Conclusions and Future Work ............................... 54

Appendix I: A series of KCN1 analog with experimental IC50
values and predicted MM-GBSA values................................... 56

References.................................................................................... 64

List of Figures
Figure 1.

Ladostigil is derived from the framework combination of Rivastigmine and
Rasagiline...................................................................................................................... 10
Figure 2. A simple demo protocol for the FRESH approach........................................... 13
Figure 3. Milnacipran and arylcyclopropylamine compounds ........................................ 17
Figure 4. Correlations of estimated binding NET affinity with experimental Ki for six
ligands. Docking receptor was generated from Compound 5 induced fitting. ............. 21
Figure 5. Correlations of estimated binding NET affinity with experimental Ki for six
ligands. Docking receptor was generated from Compound 6 induced fitting. ............. 21
Figure 6. Correlations of estimated binding SERT affinity with experimental Ki for six
ligands. Docking receptor was generated from Compound 5 induced fitting. ............. 22
Figure 7. Correlations of estimated binding SERT affinity with experimental Ki for six
ligands. Docking receptor was generated from Compound 6 induced fitting. ............. 22
Figure 8. Synthetic route for the arylcyclopropylamine analogs ..................................... 23
Figure 9. The FRESH protocol (main interface) for prioritizing arylcyclopropylamine
analogs. Some sub-protocol components are not shown. ............................................. 24
Figure 10. Structure of KCN1 with three highlighted substituent groups ....................... 26
Figure 11. Affinity pull down analysis of p300 and HIF-1α proteins using KCN1-
coupled agarose beads................................................................................................... 29
Figure 12. 14C-KCN1 binding experiment result. ............................................................ 29
Figure 13. KCN1 attached to the gold surface................................................................. 30
Figure 14. SPR sensorgrams for KCN1 binding to p300................................................. 30
Figure 15. p300-CH1 extracted from the complex. ......................................................... 31

Figure 16.

Crucial residues Leu344, Leu345, Cys388 and Cys393 on p300 CH1. Leu344
is hidden under the helix behind Leu345...................................................................... 32
Figure 17. Four clefts chosen for the docking sites(left) The top two sites with docked
KCN1 (right)................................................................................................................. 32
Figure 18. Two helices on HIF-1α (purple) superimpose with docked KCN1 at Site 1 and
Site 2 ............................................................................................................................. 33
Figure 19. Crucial residues on HIF-1α, namely Leu795, Cys800, Leu818 and Leu822 . 33
Figure 20. Linear regression attempt at Site 1. ................................................................ 35
Figure 21. Linear regression attempt at Site 2. ................................................................ 35
Figure 22. ROC at Site 1. AUC = 0.68 ............................................................................ 36
Figure 23. ROC at Site 2. AUC = 0.70 ............................................................................ 37
Figure 24. Synthetic route for KCN1 and its analogs ...................................................... 38
Figure 25. The FRESH protocol (main interface) for prioritizing KCN1 analogs. Some
sub-protocol components are not shown....................................................................... 39
Figure 26. The initial bio-test results for Compound 95.................................................. 42
Figure 27. Structures of curcumin analogs: EF24, EF31, UBS109 and SEF31 .............. 44
Figure 28. Sequences of aligned AKT-1(lower row) and AKT-2 (upper row with residue
numbers). The residues around the ATP pocket are squared in black.......................... 45
Figure 29. Top pose of EF31 on AKT-2.......................................................................... 46
Figure 30. Top pose of EF31 and EF24 on AKT-2. ........................................................ 47
Figure 31. Top pose of UBS109 (unprotonated) on AKT-2 ............................................ 48
Figure 32. Top pose of SEF31 on AKT-2........................................................................ 48

About this Master's Thesis

Rights statement

Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.

School	Laney Graduate School
Department	Chemistry
Degree	MS
Submission	Master's Thesis
Language	English
Research Field	Chemistry, Pharmaceutical Chemistry, Biochemistry
关键词	Lead optimization vHTS ADMET FRESH
Committee Chair / Thesis Advisor	Liotta, Dennis C, Emory University
Committee Members	Davies, Huw, Emory University Conticello, Vincent, Emory University Snyder, James P, Emory University

Primary PDF

Thumbnail	Title	Date Uploaded	Actions
	Design, Development and Application of a Novel In-Silico High-throughput Screening Approach ()	2018-08-28 16:31:14 -0400	Download

Abstract

Table of Contents

About this Master's Thesis

Primary PDF

Supplemental Files