Part I. Design of a novel class of reversible non-covalent small molecule inhibitors for human Granzyme B (hGrB) Part II. Curcumin and mimics as proteasome inhibitors Part III. Design of novel coactivator binding inhibitors (CBIs) for the estrogen receptor α : break the 1µM barrier 公开

Kim, Mi-Sun (2012)

Permanent URL: https://etd.library.emory.edu/concern/etds/8p58pd936?locale=zh
Published

Abstract

Grazyme B, lymphocyte serine protease, plays a critical role in controlling graft
versus leukemia and graft-versus-host diseases. A key aim of this study is to design non-
covalent small molecule inhibitors using a computational model and screening approach:
1) computational solvent mapping to identify "hot spots" in the active site; 2) virtual
screening with three constraints based on the solvent mapping results; 3) measure enzyme
activity and selectivity; 4) validate by modeling known covalent blockers. As a result,
novel classes of hGrB inhibitors have been identified. To extend the pool of scaffolds,
'scaffold hopping' has been carried out to perform shape-based searching.



The proteasome is an important target of curcumin, and several groups have
reported that inhibition of proteasome activity by curcumin and chalcone-based
derivatives cause apoptosis in human cancer cell lines. In those studies, DFT calculations
and in silico docking studies were performed to suggest binding poses in the CT-like
subunit. However, these studies introduce critical and most-likely incorrect assumptions:
1) the LUMOs were calculated for the unstable diketo form; 2) the LUMO characteristics
were interpreted to suggest that the carbonyl carbons should be the site of the terminal
Thr1 nucleophilic attack instead of the β-carbon of the enone; 3) an incorrect cis-
structure for the chalcone analogs was employed for the docking. Thus, the specific aim
of this study is to identify the correct binding poses of curcumin and chalcone-based
derivatives and to explore the binding pockets of the three active β subunits to explain
why curcumin potently inhibits the CT-like activity.


Novel classes of coactivator binding inhibitors against estrogen receptor α have
been developed. However, they fail to deliver IC50 values below 1μM in the reporter gene
assay. To understand these observations and break the 1μM barrier, we performed MD
simulations to examine solvent-based entropic contributions to the free energy of ligand
binding. The coactivator is far more effective at expelling water molecules from the
binding site than the CBIs. These observations strongly suggest that the next generation
of small molecule CBIs should span more of the peptide space, particularly on the shelf
adjacent to the binding groove.

Table of Contents

Part 1. Design of a novel class of reversible non-covalent small molecule
inhibitors for human Granzyme B (hGrB)


Chapter 1. Introduction...1

1.1 Graft-versus-leukemia (GVL) vs graft-versus-host disease (GVHD)...1
1.2 Specific aims...2
Chapter 2. Background...3
2.1 Granzymes (lymphocyte serine proteases)...3
2.2 Granzyme B...4
2.3 Substrate for granzyme B...4
2.4 Covalent serine protease inhibitors...5
2.5 Disadvantages of covalent serine protease inhibitors...6
2.6 Strategy for non-covalent inhibitor design...6
Chapter 3. Preliminary study...7
3.1 X-ray structure of apo-hGrB (PDB ID: 1FQ3)...7
3.1.1 Overall Structure...7
3.1.2 Active Site...8
3.2 X-ray structure of hGrB complex with tetrapeptide aldehyde inhibitor
(PDB ID: 1IAU)...9
3.3 Comparison between apo- and inhibitor-bound GrB...10
3.3.1 S1 subsite / P1-Asp...10
3.3.2 Oxyanion Hole...11
3.3.3 S2 subsite / P2-Pro...12
3.3.4 S3 subsite / P3-Glu...13
3.3.5 S4 subsite / P4-Ile...14
3.3.6 S1' subsite / P1'-Trp...15
3.3.7 S2' subsite / P2'-Gly...16
Chapter 4. Computational solvent mapping...17
4.1 FTMAP algorithm...17

4.2 Computational solvent mapping results comparison before and after
inhibitor binding...19
4.2.1 S1 subsite / P1-Asp...19
4.2.2 S2 subsite / P2-Pro...21
4.2.3 S3 subsite / P3-Glu...22
4.2.4 S4 subsite / P4-Ile...23
4.2.5 S1' subsite / P1'-Trp...24
4.2.6 S2' subsite / P2'-Gly...25
4.3 Summary of GrB mapping results...26
4.4 Distribution analysis of non-bonded interactions and hydrogen bonds
between probes and hGrB residues...27
4.5 Three constraints...29
Chapter 5. Virtual screening...30
5.1 Libraries...30
5.1.1 Maybridge Screening Collection (MSC)...30
5.1.2 TimTec ActiTarg-P (Protease targeted library)...31
5.1.3 ChemDiv...31
5.2 Glide Docking...31
Chapter 6. Bioassay results...32
6.1 Hits from TimTec ActiTarg-P...33
6.1.1 Thiotetrazole series...33
6.1.2 Thiotriazole series...35
6.1.3 Thiazolidinedione series...37
6.1.4 Diazolidinedione series...41
6.1.5 Others...44
6.2 Hits from ChemDiv...46
6.2.1 Sulfonamide series...46
6.2.2 Tricyclic series...48
6.2.3 Other...50
6. 3 Summary of the characteristics of the various classes of compounds...50
Chapter 7. Selectivity...51

7.1 Caspase-3 vs Caspase-8...51
7.2 GrB vs Caspase-3 and 8...52
Chapter 8. Scaffold hopping...54
8.1 Merck compounds modeling...54
8.1.1 Binding pose of Merck 15...55
8.1.2 Binding pose of Merck 19...56
8.1.3 Binding pose of Merck 20...57
8.2 Search Queries...60
8.3 Library: protease-targeted libraries...62
8.4 Computational procedures...63
8.4.1 Create a query...63
8.4.2 Perform a ROCS run...63
8.4.3 Docking using Glide...64
8.4.4 Induced fit docking using Glide and Prime...65
8.5 Results...67
8.5.1 Example: Enamine Casp-3 Target Library Results...68
8.6 Molecular property calculations using QikProp...70
Chapter 9 Synthesis, Purchase, and Biological tes...72
Chapter 10. Mid-Micromolar Activities; What Could be Done?...72
Chapter 11. Conclusions...75

Part II. Curcumin and Mimics as Proteasome Inhibitors


Chapter 1. Introduction...77
1.1 Ubiquitin-proteasome pathway (UPP) and 26S proteasome...77
1.2 Proteasome inhibitors...79
1.2.1 Covalent reversible inhibitors...79
1.2.2 Covalent irreversible inhibitors...81
1.2.3 Non-covalent reversible inhibitors...84
1.3 Specific aims...85
Chapter 2. Background...86
2.1 26S proteasome and 20S proteasome...86

2.2 Curcumin and chalcone-based derivatives as proteasome inhibitors...88

2.3 Critical issues common to both papers...89

2.3.1 Issue 1: Misuse of the diketo form of curcumin...90

2.3.2 Issue 2: Implausible binding mode of RA-1 with the cis-trans configuration...91

2.3.3 Issue 3: N-terminal Thr1-OH nucleophilic attack on C=O...91

Chapter 3. Structural examination and reactivity prediction...92 3.1 Structural examination of curcuminoids in the Cambridge Structural Database (CSD)...92 3. 2 Optimization and single-point calculations...94 3. 3 Quantification of the anticipated reactivity...95 3. 4 X-ray structure of the peptide inhibitor possessing the vinyl sulfone moiety...102 Chapter 4. Modeling the subunit targets...104 4.1 Active sites analysis...104 4.2 Non-covalent docking studies...106 4.3 Docking of curcumin...107 4.3.1 The predicted binding pose of curcumin at β1 and β2...108 4.3.2 The predicted binding pose of curcumin at β5...110 4.3.3 Summary of the predicted binding poses of curcumin at all three catalytic subunits...113 4.4 Docking of RA-1 analogs...114 4.4.1 The predicted binding pose of RA-1...116 4.4.2 The predicted binding pose of RA-2...121 4.4.3 The predicted binding pose of RA-3...124 4.5 Summary of the docking results for curcumin and RA-1, 2, and 3...126 4.6 Extending the curcumin analog SAR...128 Chapter 5. Conclusions...132 Part III. Design of Novel Coactivator Binding Inhibitors (CBIs) for the Estrogen Receptor α: Break the 1μM barrier Chapter 1. Introduction...133 1.1 Estrogen receptor (ER)...133 1.2 Blockers of ER agonist signaling...134 1.3 Specific aims...134 Chapter 2. Preliminary study...135 2.1 Structure of the ERα...135 2.2 High-throughput screening (HTS) and optimization...137 2.2.1 Modification of the A and C...138 2.2.2 Modification of phenyl piperazine sectors A and B...139 2.2.3 Modification of phenyl quinazoline region D...140 2.2.4 Modification of benzothiozole...141 Chapter 3. Molecular modeling...142 3.1 Molecular modeling...142 3.2 Induced fit docking...143 3.3 Molecular volume and qikprop calculations...144 3.4 MM-GBSA energy evaluation...145 3.5 Desmond molecular dynamics simulations and determination of numbers of displaced waters... 145 Chapter 4. Results...146 4.1 Possible competition at the ligand binding pocket...146 4.2 Phenyl piperazine scaffold...148 4.3 Benzothiazole scaffold...154 Chapter 5. What the current CBIs offer and what they lack...156 5.1 Entropic effects originating from water displacement...157 Chapter 6. Conclusion...159

References...161

About this Dissertation

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
Department
Degree
Submission
Language
  • English
Research field
关键词
Committee Chair / Thesis Advisor
Committee Members
最新修改