University of Southern California Dissertations and Theses

Inhibition of monoamine oxidase A and histone deacetylase inhibitors: computational prediction of ligand binding

(USC Thesis Other)

Inhibition of monoamine oxidase A and histone deacetylase inhibitors: computational prediction of ligand binding

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Inhibition of Monoamine Oxidase A and Histone Deacetylase Inhibitors: Computational
Prediction of Ligand Binding

by

Thomas Russell Asbell IV

A Thesis Presented to the
FACULTY OF THE USC SCHOOL OF PHARMACY
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
Master of Science
Molecular Pharmacology and Toxicology

August 2021

Copyright 2021 Thomas Russell Asbell IV
ii
Acknowledgements

I wish to thank the many individuals that made this work possible and have assisted
throughout my Master’s career at USC.
First, I wish to thank Dr. Haworth for all of his insight and guidance throughout this project.
I would also like to thank Dr. Shih for all of her mentorship and time with me. I aim to be a scientist
both of my professors can be proud of.
I would like to thank Kaichen Wang and Noam Morningstar-Kywi for their coding
expertise and suggestions throughout this project. They both have greatly lessened my workload
with their help and continue to be peers I can learn from.
Lastly, I want to thank my family for all of their love and support throughout this journey.
They continue to be my strongest support system.

iii
Table of Contents

Acknowledgements ......................................................................................................................... ii
List of Tables .................................................................................................................................. iv
List of Figures .................................................................................................................................. v
Abstract ........................................................................................................................................ viii
Chapter 1: Introduction .................................................................................................................... 1
Monoamine Oxidase ............................................................................................................ 1
Histone Deacetylase ............................................................................................................ 3
HDAC/MAO A Dual Inhibitors .......................................................................................... 4

Chapter 2: TMD Setup and Initial TMD Runs ................................................................................ 5
Inhibitor Preliminary Dockings; AutoDock Vina ............................................................... 5
TMD setup: Determination of the Covalent Adduct ........................................................... 8
.zmat Primer ........................................................................................................................ 9
TMD setup: Generation of Clorgyline-Based Inhibitor Skeleton .zmat files .................... 10
TMD Initial Molecular Docking ....................................................................................... 16
TMD Setup: Use of MODELLER to Reconstruct MAO A Flexible Loops ..................... 21
TMD setup: Generation of an Optimized Inhibitor Skeleton .zmat .................................. 23

Chapter 3: Results .......................................................................................................................... 31
Final TMD Molecular Docking ......................................................................................... 33
Solvation Analysis and Pose Selection .............................................................................. 42
Energetics Analysis and Discussion .................................................................................. 56

Chapter 4: Conclusions and Future Prospects ............................................................................... 58

References ..................................................................................................................................... 60
Appendix ....................................................................................................................................... 62

iv
List of Tables
Table 1 .......................................................................................................................................... 17
Table 2 .......................................................................................................................................... 17
Table 3 .......................................................................................................................................... 18
Table 4 .......................................................................................................................................... 18
Table 5 .......................................................................................................................................... 19
Table 6 .......................................................................................................................................... 19
Table 7 .......................................................................................................................................... 24
Table 8 .......................................................................................................................................... 24
Table 9 .......................................................................................................................................... 32
Table 10 ........................................................................................................................................ 34
Table 11 ........................................................................................................................................ 34
Table 12 ........................................................................................................................................ 35
Table 13 ........................................................................................................................................ 35
Table 14 ........................................................................................................................................ 36
Table 15 ........................................................................................................................................ 44

v
List of Figures
Figure 1 ........................................................................................................................................... 4
Figure 2 ........................................................................................................................................... 6
Figure 3 ........................................................................................................................................... 7
Figure 4 ........................................................................................................................................... 7
Figure 5 ........................................................................................................................................... 8
Figure 6 ......................................................................................................................................... 11
Figure 7 ......................................................................................................................................... 12
Figure 8 ......................................................................................................................................... 13
Figure 9 ......................................................................................................................................... 14
Figure 10 ....................................................................................................................................... 15
Figure 11 ....................................................................................................................................... 16
Figure 12 ....................................................................................................................................... 20
Figure 13 ....................................................................................................................................... 22
Figure 14 ....................................................................................................................................... 23
Figure 15 ....................................................................................................................................... 25
Figure 16 ....................................................................................................................................... 26
Figure 17 ....................................................................................................................................... 27
Figure 18 ....................................................................................................................................... 28
Figure 19 ....................................................................................................................................... 29
Figure 20 ....................................................................................................................................... 30
Figure 21 ....................................................................................................................................... 32
vi
Figure 22 ....................................................................................................................................... 33
Figure 23 ....................................................................................................................................... 37
Figure 24 ....................................................................................................................................... 38
Figure 25 ....................................................................................................................................... 39
Figure 26 ....................................................................................................................................... 40
Figure 27 ....................................................................................................................................... 45
Figure 28 ....................................................................................................................................... 45
Figure 29 ....................................................................................................................................... 46
Figure 30 ....................................................................................................................................... 46
Figure 31 ....................................................................................................................................... 47
Figure 32 ....................................................................................................................................... 47
Figure 33 ....................................................................................................................................... 48
Figure 34 ....................................................................................................................................... 49
Figure 35 ....................................................................................................................................... 49
Figure 36 ....................................................................................................................................... 50
Figure 37 ....................................................................................................................................... 50
Figure 38 ....................................................................................................................................... 51
Figure 39 ....................................................................................................................................... 52
Figure 40 ....................................................................................................................................... 53
Figure 41 ....................................................................................................................................... 53
Figure 42 ....................................................................................................................................... 54
Figure 43 ....................................................................................................................................... 54
vii
Figure 44 ....................................................................................................................................... 55
Figure 45 ....................................................................................................................................... 55
Figure 46 ....................................................................................................................................... 56

viii
Abstract
In silico molecular docking is a crucial step in the drug discovery process as it allows
informed decisions to be made on how a drug may be designed based on predicted binding
conformations leading to favorable drug-protein interactions. However, most computational
docking programs are only able to generate accurate models of reversible inhibitors. Methods to
model irreversible inhibitors are less common. This study aims to model a set of novel monoamine
oxidase A (MAO A) / histone deacetylase (HDAC) dual inhibitors that bind covalently to the co-
factor in the MAO A enzyme. In order to generate models of these inhibitors in the MAO A active
site, an in-house program, TMD (Tethered Molecular Docking), was used to survey the potential
binding poses of a small library of the inhibitors. In order to assess the energetic viability of the
poses generated by TMD, the poses were subjected to a solvation analysis using data from a second
in-house program, Watgen5. The results show that the optimal poses of the protein-bound dual
inhibitors require desolvation of the MAO A binding cavity to drive ligand binding. This project
demonstrates a method by which irreversible inhibitors can be modelled with reasonable accuracy.
In addition, this work provides an example of the importance of water desolvation in the ligand
binding process.

1
Chapter 1: Introduction
Monoamine Oxidase
Monoamine oxidase (MAO) is an important mitochondrial membrane bound enzyme
responsible for the degradation of monoamine neurotransmitters in the nervous system and other
tissue. MAO A and MAO B are two MAO isoenzymes that share nearly 70% sequence identity
and overlap in substrate specificity with some exceptions. For the purposes of this study, there will
be a heavy focus on MAO A function, inhibition, and its link to cancer. MAO substrates include
norepinephrine, epinephrine, serotonin, and other exogenous monoamines consumed in the diet.
These substrates modulate many important physiological roles including mood, memory, and
behavior
1
. MAOs catalyze a reaction in which the amine is first converted to an imine. This imine
is then hydrolyzed and results in the formation of ammonium, hydrogen peroxide, and aldehydes
as products. Aldehydes produced through this reaction are toxic and subsequently reduced by
reductases and dehydrogenases
1
.
The impact of reactive oxygen species (ROS) such as hydrogen peroxide generated as
byproducts of catalysis by MAO A has drawn attention recently. For example, studies suggest that
MAO A can promote tumorigenesis by mediating a hypoxic state in cells due to increased ROS
levels via the stabilization of hypoxia-inducible factor-1a (HIF-1a)
2
. HIF-1a is a widely
recognized anticancer target as it plays a role in regulating the cellular response to hypoxic
conditions -a common state of many tumor cells- to promote the many malignant characteristics
of tumor cells such as metastasis and survival. HIF-1a is constitutively expressed in normal cells,
though quickly degrades when under non-hypoxic conditions, having a short half-life of about 5
minutes
2
. Hypoxic conditions stabilize HIF-1a, leading to a downstream signal cascade resulting
in the transcription of several genes regulating metabolism, proliferation, and survival
2
.
2
Specifically, HIF-1a induces the transcription of GLUTs, VEGF, EGF, and other growth factors
2
.
As such, MAO A overexpression has been linked to certain cancers, since MAO A activity in
cancer cells facilitates hypoxic conditions. Past studies have found high expression levels of MAO
A in prostate cancer with reduced patient survival
3
. MAO A knockdown studies in highly
tumorigenic cell lines such as MPC3 cells both exhibited a considerable reduction in MAO A
activity and a decrease in tumor growth in vivo, while inhibition of MAO A has been shown to
slow cancer growth and reduce tumor size in the early stages of cancer in mouse models
3,4,5
. Thus,
MAO A inhibition is an attractive strategy to hinder the progression of certain cancers, such as
prostate, lung, and brain cancer.
The active site of MAO A is surrounded largely by hydrophobic and aromatic moieties and is
often described as an aromatic cage
6
. The volume of the substrate binding cavity is estimated to
be approximately 550 A
3
, and contains 5 aromatic residues and 11 aliphatic residues
7
. The flavone
flavin adenine dinucleotide (FAD) co-factor located in this cage is required for the oxidative
deamination of monoamine neurotransmitters
6
. The binding cavity extends from the cofactor to
the loop containing residues 210-216. This flavone is also the target of MAO A inhibitors such as
the potent suicide substrate clorgyline, which irreversibly binds to FAD via its alkyne and prevents
further catalysis
6
. The resulting resonance structure extends from the flavone to the nitrogen on
the inhibitor closest to the flavone. Clorgyline itself has extensively been studied for decades as
an antidepressor. Recently, it has been broadly explored as a potential anti-cancer agent due to its
ability to inhibit MAO A and alter hypoxic HIF-1alpha signals
3,8
.

3
Histone Deacetylase
Histone deacetylases (HDACs) are nuclear enzymes responsible for modulating the acetylation
state of histones, ultimately controlling gene expression. There are 18 HDAC enzymes separated
into 4 different classes. HDACs 1, 2, 3, and 8 comprise the class I Rpd3-like proteins. HDACs 4,
5, 6, 7, 9, and 10 comprise the class 2 Hda1-like proteins. SIRTs 1, 2, 3, 4, 5, 6, and 7 comprise
the class 3 SIR-like proteins. Finally, HDAC11 is the lone member of the class 4 protein. Classes
1, 2, and 4 utilize a Zn
2+
ion to facilitate enzyme catalysis while class 3 HDACs instead use an
NAD
+
cofactor for deacetylation. It is unsurprising, considering how many HDAC enzymes there
are, that there is much substrate overlap between classes. For example, HDAC1 has been
evidenced to deacetylate all core histones at different efficiencies, yet HDAC3 has been shown to
acetylate histone 4 more efficiently than HDAC1. The differences and nuances of the above HDAC
enzymes are beyond the scope of this work; however all contribute to altering gene expression,
working in concert to do so
9
.
Studies investigating the role of HDAC enzymes in cancer show HDACs are indeed necessary
for cell cycle progression. This is evidenced in studies showing many HDAC inhibitors such as
trichostatin induce cell cycle arrest. HDAC recruitment leads to tighter packing of chromatin, often
obscuring access to key promoter regions by transcription agents. HDACs, under normal
conditions, work in tandem with histone acetylases (HATs) to modulate access to key promoter
regions. This is normal in healthy cells, but HDACs are frequently overexpressed in several
cancers and therefore cause dysregulation in many cell cycle checkpoints. Thus, it is surmised that
HDACs may be an attractive therapeutic target against cancer, as their inhibition prevents the
suppression of important tumor-suppressor genes
10
.

4
HDAC/MAO A Dual Inhibitors
This present study aims to explore the interactions between a small set of novel MAO A and
HDAC dual inhibitors with the MAO A isoenzyme in silico using a variety of computational
techniques. Specifically, the inhibitors have been docked in the MAO A binding cavity to obtain
predicted structures of the inhibitors covalently bound to the enzyme. The generated models were
analyzed to provide insight on the binding characteristics of these novel inhibitors.
The aforementioned dual inhibitors have been synthesized in the likeness of the MAO A
inhibitor clorgyline as well as the hydroxamic acid-based HDAC inhibitors vorinostat and
panobinostat (Figure 1). Each dual inhibitor contains an HDAC targeting group, an MAO
targeting group, and a hydrophobic linker of varying length and rigidity. Preliminary in vitro data
of these dual inhibitors have been collected and their affinities for MAO A have been
experimentally determined (S, Mehndiratta personal communication). The generated structures of
these inhibitors in the MAO A active site will be able to guide further rational design of these
compounds and can justify the affinity of each inhibitor for MAO A.

Figure 1: General structure of the MAO A/HDAC dual inhibitors. The R group depicts the
MAO A targeting pharmacophore modelled after clorgyline. The R’ depicts the structure of a
hydroxamic acid; a necessary chemical moiety enabling HDAC inhibition (S, Mehndiratta
personal communication).

O N
Cl
HOHN
O
R= R’=
R
N
n
R’
HDAC targeting group
Hydrophobic linker of varying length and rigidity
MAO A
targeting
group
5
Chapter 2: TMD Setup and Initial TMD Runs
Inhibitor Preliminary Dockings; AutoDock Vina
Structures of dual inhibitors 357, 359, 440, and 441 (Figure 2) were converted to pdb files
using SMILES strings in Chimera. Using the MAO A crystal structure 2BXS chain B on the PDB
as a template protein, AutodockVina was used in an attempt to dock the molecules
11
. These
dockings initially failed, however, as there were a few complicating factors. First, for the purposes
of this study, it is reasonable to assume that the dual inhibitors behave much like the irreversible
inhibitor clorgyline. Clorgyline is a potent MAO A inhibitor that binds to an FAD cofactor in the
MAO A binding cavity. This effectively prevents any further catalysis by the enzyme, as the FAD
is vital to the catalysis (Figure 3). Clorgyline and the dual inhibitors have the same necessary
chemical structure to form a covalent bond with the FAD. As such, the dual inhibitors should be
docked in a such a way that the irreversible adduct is formed. Therefore, these first attempts to
dock the molecules failed as AutodockVina was unable to simulate the covalent linkage between
the inhibitor and the FAD.
A second problem that complicated this method of docking the inhibitors came in the PDB
file used. The residues between GLY110 and TRP116 are missing in the crystal structure, leaving
a broken loop in the model (Figure 4). This complicates creating an accurate model, since these
missing residues may adopt conformations that interact or interfere with ligand binding. In any
case, using the AutodockVina program alone does not address the issue that these potentially
important residues are missing.
Lastly, some of the inhibitors appeared to be simply too big to fit into the binding site. 440
and 441 for example, were too large to dock and generated no results using Autodock. These
bulkier inhibitors clashed with many of the residues surrounding the entrance of the binding cavity.
6
Because of these problems, it was deemed necessary to use a program called Tethered Molecular
Docking (TMD) to dock the inhibitors. TMD is in-house software from Dr. Ian Haworth’s lab that
allows simulation of the desired covalent adduct between the inhibitors and the enzyme while
simultaneously varying torsions and bond angles of the ligand to generate ligand-bound models.

Figure 2: Structures of the MAO /HDAC dual inhibitors analyzed in this study (S, Mehndiratta
personal communication).

O N
Cl
N
H
O
HOHN
O
O N
Cl
N
H
O
HOHN
O
-
OHN
N
H
O N
O
Cl
-
OHN
N
H
O N
Cl
O
357
359
440
441
7

Figure 3: Schematic of the reaction that is proposed to occur between the alkyne structure in
clorgyline and the FAD cofactor found in the MAO A binding cavity.

Figure 4: Crystal structure of MAO A 2BXS chain B. The black arrow shows the broken strand
of the protein flanked by GLY110 (green) and TRP116 (red). Clorgyline is colored blue and can
be seen just underneath the broken loop.

Cl
Cl
O N
Clorgyline
+
FAD
O N
Cl
Cl
FAD
8
TMD setup: Determination of the Covalent Adduct
In order to dock the molecules using TMD, it is first necessary to consider what the
covalent adduct would look like upon ligand binding. One of the main attractions in using TMD
is that the program has the ability to simulate the covalent linkage between the inhibitors and the
FAD cofactor. Pavlin et al. used quantum chemical calculations to predict the linkage between the
alkyne and FAD
12
. The calculations suggested that the triple bond is conserved and that the
nitrogen on the inhibitor will remain trigonal pyramidal (Figure 5, (1)). However, other reports
indicate that a resonance structure is formed (Figure 5, (2) and (3))
13
. Careful inspection of the
2BXS chain B PDB showed the amine on the inhibitor near the triple bond to be planar, suggesting
the nitrogen contains a partial positive charge. This partial positive charge is likely stabilized by
cation-pi interactions due to the numerous aromatic residues in its vicinity, including PHE352,
TYR407, and TYR444. Thus, for the purposes of docking the inhibitors, the dominant resonance
structure (2) was generated with torsions appropriate for each double and single bond.

Figure 5: Structures of the proposed covalent adduct formed between clorgyline and an FAD
cofactor. Pavlin et. al. used thermodynamic calculations to predict (1), a structure generated by
an anionic mechanism. Resonance structures (2) and (3) are found in the 2BXS crystal structure,
with (2) being the dominant resonance structure.
Cl
Cl
O N
N
N
N
H
N
Enz
O O
R
Cl
Cl
O N
N
N
N
H
N
Enz
O O
R
Cl
Cl
O N
N
N
N
H
N
Enz
O O
R
1
2
3
9
.zmat Primer
The following is a brief primer on how to read a .zmat file from left to right, top to bottom.
For the purposes of this section, please refer to Figure 6 when an example is given:
- The first column is a numeric identifier of an atom. .zmat files assign each atom within the file a
number. For example, the first atom detailed in the .zmat file shown in Figure 6 is labelled number
4. The atom underneath it is labeled number 5. Important information on each atom’s geometry
and connectivity are contained in each row.
- The second column details the elemental identity of the atom. For example, atom number 4 is a
carbon atom. Atom number 7 is a nitrogen atom. Atoms 33-62 are hydrogen atoms.
- The third column refers to the connectivity of the atom in question. For example, atom number
4 is bound to atom 1. Atom number 7 is bound to atom number 6. It should be noted that atoms 1,
2, and 3 are not detailed in all .zmat files in this project, as they refer to specific atoms on the FAD
cofactor in the MAO A structure (see Figures 10 and 11 for the identities of atom numbers 1, 2,
and 3 on the FAD cofactor).
- The fourth column describes which atom forms a bond angle with the atom in question. For
example, atom 4 forms a bond angle with atom number 2.
- The fifth column describes which atom forms a torsional angle with the atom in question. For
example, atom number 4 forms a torsional angle with atom number 3.
- The sixth column describes the bond length of the atom in question with another. For example,
atom number 4 has a bond length of 1.38 angstroms with atom number 1.
- The seventh column describes the bond angle of the atom in question with another. For example,
atom number 4 forms a bond angle with atom 2 that is 75.58 degrees.
10
- The eighth and final column describes the torsional angle of the designated atom with another.
For example, atom number 4 forms a torsional angle of -114.79 degrees with atom number 3.

TMD setup: Generation of Clorgyline-Based Inhibitor Skeleton .zmat files
With the structure of the covalent adduct determined, certain files must be prepared in order
to operate TMD. One of the most important files in running TMD are .zmat files for each inhibitor
of interest. These files contain detailed information on the connectivity of each atom to one another
as well bond lengths, angles and torsions of each bond. At the start of this project, .zmat files were
generated with the intention of making the inhibitors’ bond angles and torsions very similar to the
clorgyline moiety found the 2BXS chain B crystal structure. This was done because there is already
physical evidence showing how a clorgyline-like structure may bind to MAO A within the crystal
structure itself. The rest of the inhibitor structure was generated using correct molecular geometry
for each bond. All atom conformations not comprising the clorgyline-like skeleton were
purposefully given torsions and bond angles that form staggered conformations in order to keep
these parts of the structure in their theoretically lowest energetic state. Figures 6, 7, 8, and 9 show
the raw .zmat files used for inhibitors 357, 359, 440, and 441 with some minor adjustments for
clarity. Figures 10 and 11 show each inhibitor bound to the FAD cofactor with each atom
described in the .zmat files labelled appropriately.

11

Figure 6: Raw .zmat file of 357 using torsions and bond angles of clorgyline in the crystal
structure where appropriate.
12

Figure 7: Raw .zmat file of 359 using torsions and bond angles of clorgyline in the crystal
structure where appropriate.
13

Figure 8: Raw .zmat file of 440 using torsions and bond angles of clorgyline in the crystal
structure where appropriate.

14

Figure 9: Raw .zmat file of 441 using torsions and bond angles of clorgyline in the crystal
structure where appropriate.
15
While each .zmat file contains a few key differences in order to uniquely store the
geometries of the atoms in each inhibitor, atoms 4-14 in the .zmats depicted in Figures 6, 7, 8,
and 9 are particularly important. These specific atoms have the same bond and torsional angles
associated with the original clorgyline ligand in the 2BXS chain B crystal structure. Later, as
described on Page 23, these geometries will be changed manually to optimize the binding poses
for each inhibitor.

Figure 10: Structure of 357 (A) and 359 (B) covalently bound to the FAD cofactor in MAO A.
All heavy atoms are labelled in accordance with their respective .zmat files.

O
N
Cl
NH O
N
H
O
N
N
NH
H
N
R
O
O
Enz
1 2
3
4
5
6
7 8
9 10
11 12
13
14 15
16
17
18 19
20
21
22
23
24
25
26
27
28
29
30
31
32
O
H
O
N
Cl
NH O
N
N
NH
H
N
R
O
O
Enz
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
30
33
34
N
H
O
26
O
H
27
28
29
30
31
A
B
16

Figure 11: Structure of 440 (C) and 441 (D) covalently bound to the FAD cofactor in MAO A.
All heavy atoms are labelled in accordance with their respective .zmat files.

TMD Initial Molecular Docking
With the proper .zmat files in-hand, TMD was run using the 2BXS chain B protein by
varying several different torsions within each .zmat file to explore possible poses each inhibitor
may take while bound to the FAD. In order to keep the clorgyline-like skeleton on each inhibitor
constant, only torsions within the hydrophobic linker of each dual inhibitor were varied. The
torsions tested in each TMD run and the results of the run are shown in Tables 1 to 6.
N
H
NH
O
N
O
Cl
N
N
NH
H
N
R
O
O
Enz
O
H
4
1
2
3
5
6
7 8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26 27
28
29
30
31
32
33
NH
O
N
Cl
N
N
NH
H
N
R
O
O
Enz
4
1
2
3
5
6
8
9
10
11
12
13 14
15
16
17
18
19
20
21
22
23
24
25
30
31
N
H
O
O
H
7
26
27
28
29
C D
17
Table 1: Torsions tested using TMD exploring different possible conformations of 357.
Atom # Set torsion to: End with Torsion: In increments of:
19 0 180 180
21 60 300 120
22 60 300 120
23 60 300 120
24 60 300 120
25 60 300 120
26 60 300 120
27 60 300 120
Maximum Clash Distance 1.5 Angstroms
Maximum Tolerated Clashes 5

Total number of Poses 15

Table 2: Torsions tested using TMD exploring different possible conformations of 359.
Atom # Set torsion to: End with Torsion: In increments of:
19 0 180 180
21 60 300 120
22 60 300 120
23 60 300 120
24 60 300 120
25 60 300 120
26 60 300 120
27 60 300 120
28 60 300 120
29 60 300 120
Maximum Clash Distance 1.5 Angstroms
Maximum Tolerated Clashes 5

Total number of Poses 0

18
Table 3: Torsions tested using TMD exploring different possible conformations of 440.
Atom # Set torsion
to:
End with
Torsion:
In increments
of:
19 160 180 10
20 60 300 120
21 0 330 30
27 0 180 180
29 0 180 180
Maximum Clash Distance 1.5
Angstroms

Maximum Tolerated Clashes 5

Total number of Poses 0

Table 4: Torsions tested using TMD exploring different possible conformations of 440. Note the
torsions tested for atom 19 are different from the torsions tested for this same atom in Table 3.
Atom # Set torsion
to:
End with
Torsion:
In increments
of:
19 340 380 10
20 60 300 120
21 0 330 30
27 0 180 180
29 0 180 180
Maximum Clash Distance 1.5
Angstroms

Maximum Tolerated
Clashes
5

Total number of Poses 0

19
Table 5: Torsions tested using TMD exploring different possible conformations of 441.
Atom # Set torsion
to:
End with
Torsion:
In increments
of:
19 160 180 10
20 60 300 120
21 0 330 30
27 0 180 180
Maximum Clash Distance 1.5
Angstroms

Maximum Tolerated
Clashes
5

Total number of Poses 6

Table 6: Torsions tested using TMD exploring different possible conformations of 441. Note the
torsions tested for atom 19 are different from the torsions tested for this same atom in Table 5.
Atom # Set torsion
to:
End with
Torsion:
In increments
of:
19 340 380 10
20 60 300 120
21 0 330 30
27 0 180 180
Maximum Clash Distance 1.5
Angstroms

Maximum Tolerated
Clashes
5

Total number of Poses 0

Of the inhibitors and the conformations tested only 357 and 441 yielded any results in TMD
(Tables 1 and 5). 359 and 440 produced no results (Tables 2, 3, and 4). Many of these first TMD
docking attempts failed due to a number of reasons. Upon further inspection, 359 and 440 sterically
clashed with residues 108-110 at the entrance of the MAO A binding cavity (Figure 12). 440
additionally sterically clashed with residues 209-212. It has been reported by several groups that
the entrance of the binding pocket lies in three separate loops; residues 93-95, 109-112, and 208-
212. As is noted by Son et. al., the entrance to the binding cavity is far too small for an inhibitor
20
or substrate to enter
6
. Through careful mutagenesis studies, Son et. al. demonstrated the need for
these loops to be flexible and dynamic, adopting “open” conformations that would allow an
inhibitor or substrate to enter the active site. The residues that the inhibitors initially clashed with
are many of these same flexible loops
6
. This indicates that these flexible loops adopt different
conformations upon binding to a larger ligand, such as dual inhibitors 359 and 440. Another
proposed reason why the first initial docking attempts failed was that it may be the case the dual
inhibitors do not adopt the same clorgyline-like conformation that clorgyline itself adopts when
binding to the FAD cofactor. Small variations to the clorgyline backbone may better fit the
hydrophobic linker and HDAC targeting moiety of each inhibitor such that the inhibitors
themselves do not clash as often with the protein. The following sections will address both of these
issues in preparation for the final dockings using TMD.

Figure 12: Entrance to the binding cavity of MAO A. Of particular note are residues 93-95
(orange), 109-112 (blue), and 208-212 (green).
21
TMD Setup: Use of MODELLER to Reconstruct MAO A Flexible Loops
As previously discussed above, many of the loops the dual inhibitors clashed with are
flexible, likely allowing the protein to move and adjust these same loops to accommodate substrate
entry into the binding cavity. In order to explore alternative loop conformations of these residues,
and to reconstruct the residues missing in the crystal structure, MODELLER was used.
MODELLER is an application within Chimera that allows one to reconstruct and refine specific
residues within a protein as well as totally reconstruct poorly resolved portions of a protein
structure
14
. Two different sets of MODELLER generated MAO A proteins were produced.
Because 359 only clashed with the loop containing residues 108-110, the first set of reconstructed
proteins comprise only of protein structures in which MODELLER was instructed to solely refine
loop 108-116. The second set of reconstructed proteins were produced by refining loop 108-116
as well as 208-212. In total, MODELLER produced 5 protein structures using residues 108-116
and 3 structures using residues 108-116 and 208-212 (Figures 13 and 14). In order to determine
the viability of each model generated, each protein was overlayed with a rat MAO A crystal
structure to ensure each loop conformation was reasonably recreated. These final MAO A models
will be used as protein templates for docking the dual inhibitors in future TMD dockings.

22

Figure 13: Structures of protein models 1-5 with strands created using MODELLER. Models 1-
5 contain conformations of residues 108-116 (red).

23

Figure 14: Structures of protein models 6-8 with strands created using MODELLER. Models 6,
7, and 8 contain conformations of 108-116 (red) and 208-212 (blue).
TMD setup: Generation of an Optimized Inhibitor Skeleton .zmat
In order to investigate conformations alternative to the clorgyline-like backbone of each
inhibitor, several key torsions were varied. Torsions of backbone atoms 5, 10, 11, 12, 13, and 19
of inhibitor 357 were varied as specified in Table 7. Additional edits to certain torsions were also
made to ensure each atom exhibited correct geometry and torsional angles. In order to allow the
hydrophobic linker and HDAC targeting moiety more space to move while exploring alternative
conformations of the backbone, the loop consisting of residues 208-212 was deleted from the MAO
A crystal structure for this docking attempt. This docking attempt yielded 36 possible poses
utilizing six different combinations of torsional angles of atoms 5, 10, 11, 12, 13, and 19 (Table
8). .zmat files of inhibitor 357 using the six different backbone results were then created for future
inspection (Figures 15-20).
24
Table 7: Torsions tested using TMD exploring different conformations of the clorgyline
backbone of 357.
Atom # Set torsion
to:
End with
Torsion:
In increments
of:
5 235 355 5
10 0 350 10
11 60 300 120
12 60 300 120
13 60 300 120
19 0 180 180
Maximum Clash Distance 1.5
Angstroms

Maximum Tolerated
Clashes
3

Total number of Poses 36

Table 8: Acceptable torsions of atoms 5, 10, 11, 12, 13, and 19 of each pose generated from the
docking attempt described in Table 7.
Pose ID 5 10 11 12 13 19
A 235 70 300 180 60 0
B 235 40 60 180 300 0
C 240 40 60 180 300 0
D 245 30 60 180 300 0
E 245 70 60 60 60 0
F 250 60 60 60 60 0

25

Figure 15: .zmat file of inhibitor 357 with torsions consistent with backbone “A” generated by
the settings shown in Table 7.
26

Figure 16: .zmat file of inhibitor 357 with torsions consistent with backbone “B” generated by
the settings shown in Table 7.

27

Figure 17: .zmat file of inhibitor 357 with torsions consistent with backbone “C” generated by
the settings shown in Table 7.
28

Figure 18: .zmat file of inhibitor 357 with torsions consistent with backbone “D” generated by
the settings shown in Table 7.
29

Figure 19: .zmat file of inhibitor 357 with torsions consistent with backbone “E” generated by
the settings shown in Table 7.
30

Figure 20: .zmat file of inhibitor 357 with torsions consistent with backbone “F” generated by
the settings shown in Table 7.
31
Chapter 3: Results
With .zmat files made of backbones A-F of inhibitor 357, each backbone was docked in
the same MAO A protein mentioned at the start of this section (MAO A protein with residues 208-
212 deleted) while simultaneously varying the torsions of the hydrophobic linker (See results in
Table 9). While each backbone generated a respectable number of poses, it was deemed only
backbones E and F produced feasible inhibitor models. Backbones A-D generated inhibitor poses
that positioned the hydrophobic linker to pass through loops 68-72, 201-216, and 474-488,
eventually causing each inhibitor pose to go through the side of the protein in lieu of staying in the
hydrophobic pocket (Figure 21). Since little evidence exists suggesting the flexibility for these
loops to move upon ligand binding, and because it makes little sense to predict the largely
hydrophobic linker will be surrounded by bulk water instead of staying inside a hydrophobic
cavity, these backbones were not deemed as good alternative conformations to the clorgyline
backbone and were excluded from any further analysis. E and F, by comparison, largely keep the
hydrophobic linker inside the binding cavity (Figure 22). The hydroxamic acid in most poses of
the E and F skeletons seem to reach the entrance of the binding cavity accessing the bulk solvent
surrounding the protein, solvating the hydrophilic acid. Due to the high similarities in torsional
angles that exists between backbones E and F, F was chosen as a representative backbone
containing optimal torsional angles that more appropriately accommodates the dual inhibitors for
future analysis.

32
Table 9: Generated results for backbones A-F.
Pose ID Results
A 59
B 11
C 24
D 6
E 28
F 10

Figure 21: Dual inhibitor 357 with backbone A torsions (light blue) overlayed with the
clorgyline (green) bound MAO A crystal structure (purple).

33

Figure 22: Dual inhibitor 357 with backbone F torsions (light blue) bound to MAO A (tan). The
inhibitor is predicted to stay within the binding pocket.

Final TMD Molecular Docking
Using inhibitor backbone F generated in the previous section, .zmat files were made such
that 357, 359, 440, and 441 all had the same torsional angles for their clorgyline-like skeleton
(Figures 23-26). TMD was then run using these .zmat files to dock each inhibitor in the binding
cavities of protein models 1-8 using the settings described in Tables 10-13. As a control, each
inhibitor with the original torsional angles for the clorgyline-like moiety were also docked against
protein models 1-8. The results for all final runs can be seen in Table 14.

34
Table 10: Final torsions tested using TMD for various atoms in the hydrophobic linker of the
optimized and clorgylone backbones of inhibitor 357.
Atom # Set torsion
to:
End with
Torsion:
In increments
of:
19 340 380 10
21 60 300 120
22 60 300 120
23 60 300 120
24 60 300 120
25 60 300 120
26 60 300 120
27 60 300 120
Maximum Clash Distance 1.5
Angstroms

Maximum Tolerated
Clashes
5

Table 11: Final torsions tested using TMD for various atoms in the hydrophobic linker of the
optimized and clorgyline backbones of inhibitor 359.
Atom # Set torsion
to:
End with
Torsion:
In increments
of:
19 340 380 10
21 60 300 120
22 60 300 120
23 60 300 120
24 60 300 120
25 60 300 120
26 60 300 120
27 60 300 120
28 60 300 120
29 60 300 120
Maximum Clash Distance 1.5
Angstroms

Maximum Tolerated
Clashes
5

35
Table 12: Final torsions tested using TMD for various atoms in the hydrophobic linker of the
optimized and clorgyline backbones of inhibitor 440.
Atom # Set torsion
to:
End with
Torsion:
In increments
of:
19 160 200 10
19 340 380 10
20 60 300 120
21 0 330 30
27 0 180 180
29 0 180 180
Maximum Clash Distance 1.5
Angstroms

Maximum Tolerated
Clashes
5

Table 13: Final torsions tested using TMD for various atoms in the hydrophobic linker of the
optimized and clorgyline backbone of inhibitor 441.
Atom # Set torsion
to:
End with
Torsion:
In increments
of:
19 160 200 10
19 340 380 10
20 60 300 120
21 0 330 30
27 0 180 180
Maximum Clash Distance 1.5
Angstroms

Maximum Tolerated
Clashes
5

36
Table 14: Final results for the clorgyline (“C”) and optimized (“O”) backbone runs in TMD for
inhibitors 357, 359, 440, and 441.
Protein
ID
C-357
Results
O-357
Results
C-359
Results
O-359
Results
C-440
Results
O-440
Results
C-441
Results
O-441
Results
1 0 58 0 138 0 0 6 0
2 0 17 0 22 0 0 6 0
3 0 25 0 58 0 0 6 0
4 0 44 0 85 0 0 6 0
5 0 4 0 1 0 0 6 0
6 0 49 0 49 0 6 0 10
7 0 0 0 0 0 0 0 0
8 0 0 0 0 0 0 0 0

37

Figure 23: .zmat of inhibitor 357 with torsions consistent with backbone F.
38

Figure 24: .zmat of inhibitor 359 with torsions consistent with backbone F.
39

Figure 25: .zmat of inhibitor 440 with torsions consistent with backbone F.
40

Figure 26: .zmat of inhibitor 441 with torsions consistent with backbone F.
41
Across all TMD dockings with the exception of 441, the optimized backbone of each
inhibitor consistently generated far more results than its original clorgyline backbone counterpart.
This provides good evidence that larger and bulkier inhibitors have different clorgyline backbone
conformations when compared to the original clorgyline backbone found in the crystal structure.
Additionally, the docking also demonstrated the need for specific loops in the MAO A crystal
structure to adopt alternative conformations from what is seen in the 2BXS crystal structure. This
is because very few or no results were obtained when docking the inhibitors in the original MAO
A protein, yet several conformations were generated when these same inhibitors were docked
against MAO A proteins containing alternative loop conformations.
Though 441 was the only inhibitor to generate more total poses using the clorgyline
backbone when compared to the optimized backbone, docking using the TMD backbone gave
more poses compared across any single protein model tested. This demonstrates a final obstacle in
this project; picking a pose and protein model that accurately depicts how each inhibitor binds to
the MAO A protein. For instance, while 441 generated 30 poses using the clorgyline backbone, it
may be the case that the 10 poses generated using the optimized backbone are energetically more
favorable. In order to estimate how energetically favorable each inhibitor-protein pose is, a
solvation analysis was conducted.

42
Solvation Analysis and Pose Selection
The results of the docking yielded little information on how well each inhibitor would bind
to MAO A due to the number of results generated and because many poses are nearly identical.
However, in vitro IC50 MAO A inhibitory activities of each dual inhibitor in GL26 cells were
obtained (Table 15) (S, Mehndiratta personal communication). Certain dual inhibitors exhibited
very potent MAO A inhibition relative to other inhibitors. With such large differences in affinity
for MAO A, the generated inhibitor-protein poses should reflect this difference through variations
in the energetic feasibility of each pose. To determine which poses are energetically favorable, and
to understand the large differences in inhibitor affinities compared to one another, a solvation
analysis of each pose was conducted. This analysis yields data on the hydrogen bond network
differences between each individual generated pose with the ligand bound or unbound to the
protein, such as how many water molecules were displaced from the binding cavity and how many
protein-ligand hydrogen bonds were disrupted. An algorithm was then employed using this data
in order to compute an estimated free energy of binding value by comparing the different states of
the hydrogen bond network pre-ligand binding and post-ligand binding. In order to calculate the
free energy of binding from each pose, the algorithm employs the following equation: Free Energy
(Kcal/Mol) = ((-2*Protein-Ligand HB) + (2*(Broken Protein HBs + Broken Ligand HBs) + (1*
Like Charge Clash count)) - ((2*( Absolute Displaced Waters + Contact Displaced Hydrophobic
Waters + Contact Displaced Bulk Waters)) + (1* Rotatable Bonds of Bound Ligand) + (2* Trapped
Waters within the protein))) + % Unmatched Water Molecules (See raw data for all poses in
Appendix Tables 1-22).
Using these estimated free energy of binding values, specific inhibitor-protein poses can
be selected as models offering a good representation of what each inhibitor may look like upon
43
binding. However, it is beyond the scope of this project to determine the energetic favorability of
each protein model alone. Therefore, all protein models generated as well as the original clorgyline
protein structure will be deemed as having a net difference of 0 DG. This implies all protein models
are equally possible for the protein to adopt. So, energetic penalties associated with the movement
of certain loops within the protein will not be included in the solvation analysis and consequently
pose selection. The only exception to this comes when considering how inhibitors 357 and 359
may bind. Since it is likely each inhibitor forces certain loops to adopt conformations higher in
energy, only protein models 1-5 will be considered as appropriate models for 357 and 359 since
protein models 6-8 predict an additional strand will move and adopt a conformation different from
that in the original crystal structure.
Determination of a pose and protein model for each inhibitor was decided by two important
factors. The first factor taken into consideration was the calculated free energy of each pose, as
determined by the solvation analysis algorithm. Since poses with the lowest free energy will be
energetically preferred for the protein-inhibitor complex to adopt over poses with higher free
energy, only the lowest estimated free energy models were considered as the most favorable poses.
In order to ensure a precise selection of each pose, all poses were given a “Hydrogen Bond-Factor
Sum” score (HB-Factor Sum). This score for each pose was calculated as: total # of water
molecules displaced by the ligand + total # of hydrogen bonds between the protein and ligand –
total # of lost hydrogen bonds by the protein. This score was also used in pose selection to ensure
the most important determinants of ligand binding, such as intermolecular interactions and water
displacement, were considered alongside the estimated free energy. Interestingly, there was an
inverse correlation between HB-Factor sums and estimated free energy values, as can be seen in
figures such as Figure 27. It is also worth noting that though a complex may have a good
44
combination of HB-Factor sum and DG, certain results contained geometries of the inhibitor that
are highly improbable. Therefore, these inhibitor poses were excluded from consideration due to
the lack of geometrical feasibility of the predicted inhibitor conformation. As such, the protein
model-pose combination with the highest sum of HB-Factor sum + -(DG) and a feasible inhibitor
geometry were chosen amongst all poses for a given inhibitor as the final result.
Table 15: Unpublished experimental MAOA inhibitory IC50 data using GL26 cells (S,
Mehndiratta personal communication).
Compound MAOA Inhibitory IC50
(M)
357 2.64 × 10
-8

359 1.91 × 10
-12

440 4.03 × 10
-9

441 4.09 × 10
-10

TMD generated 148 total possible poses for 357 using both the optimized inhibitor skeleton
and protein models 1-5 (Table 14). No poses were generated using both the clorgyline-like
inhibitor skeleton and protein models 1-5. Figures 27-31 show the calculated free energy estimates
and HB-factor sum of each pose for each protein model. Of all protein models tested, model 4
generated some of the most impressive results with 2 poses with a HB-Factor score of 13 and 1
pose with an estimated -12.5 Kcal/Mol DG. The pose with the most optimal free energy estimate,
HB-Factor sum, and feasible geometry was selected as the likeliest inhibitor-protein conformation
of 357 in complex with MAO A (Figure 32).

45

Figure 27: Free energy estimate vs HB-factor sum for inhibitor 357 docked to protein model 1.

Figure 28: Free energy estimate vs HB-factor sum for inhibitor 357 docked to protein model 2.
0
2
4
6
8
10
12
14
-6.27
-3.11
-1.67
-0.74
-0.30
0.64
0.89
1.18
1.38
1.74
2.27
2.70
3.01
3.06
3.55
4.52
4.95
5.10
6.32
6.61
6.93
7.51
7.62
8.01
8.46
8.94
9.67
10.97
11.65
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
0
2
4
6
8
10
12
14
-9.70 -3.88 -3.59 -3.05 -2.56 -2.26 -1.94 -1.05 -0.13 0.95 2.17 2.55 3.03 3.67 6.53 6.85 8.04
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
46

Figure 29: Free energy estimate vs HB-factor sum for inhibitor 357 docked to protein model 3.

Figure 30: Free energy estimate vs HB-factor sum for inhibitor 357 docked to protein model 4.
0
2
4
6
8
10
12
14
-10.11
-8.84
-5.85
-5.15
-4.89
-4.56
-4.46
-4.08
-3.87
-3.51
-3.45
-3.37
-2.73
-0.92
-0.23
0.22
0.98
1.49
1.49
4.31
4.32
5.26
5.65
7.35
8.12
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate Vs HB-Factor Sum
0
2
4
6
8
10
12
14
-12.50
-7.67
-6.30
-5.56
-4.96
-4.67
-4.43
-3.70
-2.32
-1.77
-1.07
-0.99
-0.27
-0.03
0.24
1.02
1.54
2.10
3.60
3.73
4.07
5.31
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
47

Figure 31: Free energy estimate vs HB-factor sum for inhibitor 357 docked to protein model 5.

Figure 32: Most energetically favorable pose generated by TMD for 357 with MAO A.
0
5
10
15
-6 -3.593495935 -3.227091633 1.396226415
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
48
TMD generated 304 total possible poses for 359 using the optimized inhibitor skeleton and
protein models 1-5 (Table 14). No poses were generated using the clorgyline-like inhibitor
skeleton and protein models 1-5. Figures 33-36 show the calculated free energy estimates and HB-
factor sum of each pose for each protein model. The lone pose generated using protein model 5
had a HB-factor score of 4 and an estimated free energy of +7.23 Kcal/Mol. Though TMD
produced several poses for this dual inhibitor, many of the results contain poses that have
comparatively low DG estimates. The only clear result showing any promise is the lone pose in
protein model 3 with an HB-Factor sum of 12. Thus, this pose was selected as the likeliest
inhibitor-protein conformation of 359 in complex with MAO A (Figure 37).

Figure 33: Free energy estimate vs HB-factor sum of inhibitor 359 docked to protein model 1.

0
2
4
6
8
10
12
14
-6.56
-3.95
-3.30
-2.77
-1.81
-0.25
0.46
1.10
1.81
2.29
2.64
3.78
4.01
4.33
4.81
5.33
5.64
5.76
6.09
6.49
6.98
7.27
7.47
8.25
8.75
9.08
9.84
10.51
11.09
11.63
12.47
13.20
14.01
15.74
17.59
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
49

Figure 34: Free energy estimate vs HB-factor sum of inhibitor 359 docked to protein model 2.

Figure 35: Free energy estimate vs HB-factor sum of inhibitor 359 docked to protein model 3.
0
2
4
6
8
10
12
14
16
-4.37
-2.28
-0.17
0.21
1.78
1.94
2.63
3.16
3.54
3.67
4.37
5.64
6.00
6.16
6.27
6.80
8.07
8.15
9.11
9.35
12.12
14.39
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
0
2
4
6
8
10
12
14
-9.45
-6.92
-6.17
-6.00
-5.47
-4.58
-4.12
-3.65
-3.44
-3.05
-2.70
-2.50
-2.35
-1.86
-1.82
-1.50
-0.89
-0.52
-0.48
-0.40
-0.33
0.21
0.62
0.74
1.30
1.59
2.26
3.23
7.71
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimatevs HB-Factor Sum
50

Figure 36: Free energy estimate vs HB-factor sum of inhibitor 359 docked to protein model 4.

Figure 37: Most energetically favorable pose generated by TMD of 359 with MAO A.
-2
0
2
4
6
8
10
12
14
-4.48
-2.44
-2.06
-0.39
-0.02
1.34
1.60
1.99
2.59
2.89
3.13
3.32
4.06
4.43
4.67
4.91
5.44
5.59
6.60
7.06
7.41
8.14
8.61
9.27
9.49
10.37
10.98
11.91
18.00
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
51
TMD generated 6 total possible poses for 440 using the optimized inhibitor skeleton and
protein models 1-8 (Table 14). No poses were generated using the clorgyline-like inhibitor
skeleton and protein models 1-8. Figure 38 shows the calculated free energy estimates and HB-
factor sum of each pose for protein model 6. The pose with lowest free energy, highest HB-factor
sum, and most feasible geometry was selected as the likeliest inhibitor-protein conformation of
440 in complex with MAO A (Figure 39).

Figure 38: Free energy estimate vs HB-factor sum of inhibitor 440 docked to protein model 6.

0
2
4
6
8
10
-4.89 -1.73 -1.42 0.01 0.89 2.61
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
52

Figure 39: Most energetically favorable pose generated by TMD of 440 with MAO A.

TMD generated 10 total possible poses for 441 using the optimized inhibitor skeleton and
protein models 1-8 (Table 14). 30 poses were generated using the clorgyline-like inhibitor skeleton
and the original protein structure (Table 14). Figure 40 shows the calculated free energy estimates
and HB-factor sum of each pose for protein model 6 using the optimized inhibitor skeleton.
Figures 41-45 show the calculated free energy estimates and HB-factor sum of each pose for
protein models 1-5. This is the only set of TMD dockings that show the optimized backbone is not
as energetically favorable over the original clorgyline-like skeleton if one were to compare the
data in Figures 40 and 42. However, upon closer inspection of all clorgyline-skeleton poses, each
result yielded geometrically unreasonable conformations of the inhibitor. Therefore, these results
were excluded from consideration and only poses generated from the optimized backbone were
53
assessed. The pose with lowest free energy, highest HB-factor sum, and most feasible geometry
was selected among the optimized inhibitor skeleton as the likeliest inhibitor-protein conformation
of 441 in complex with MAO A (Figure 46).

Figure 40: Free energy estimate vs HB-factor sum of inhibitor 441 docked to protein model 6
using the optimized inhibitor skeleton.

Figure 41: Free energy estimate vs HB-factor sum of inhibitor 441 docked to protein model 1
using the clorgyline-like skeleton.
0
2
4
6
8
10
12
14
-10.22 -9.92 -7.15 -6.15 -5.39 -4.85 -4.74 -3.04 -2.80 -0.38
HB-Factor Sum
Estimnated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
0
2
4
6
8
10
12
14
16
-19.74 -19.74 -18.82 -17.88 -11.19 -10.87
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
54

Figure 42: Free energy estimate vs HB-factor sum of inhibitor 441 docked to protein model 2
using the clorgyline-like skeleton.

Figure 43: Free energy estimate vs HB-factor sum of inhibitor 441 docked to protein model 3
using the clorgyline-like skeleton.
0
2
4
6
8
10
12
14
16
18
-24.67 -24.35 -22.02 -19.38 -14.46 -14.14
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
13
13.5
14
14.5
15
15.5
16
16.5
-22.44 -21.46 -20.65 -18.75 -16.37 -14.37
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
55

Figure 44: Free energy estimate vs HB-factor sum of inhibitor 441 docked to protein model 4
using the clorgyline-like skeleton.

Figure 45: Free energy estimate vs HB-factor sum of inhibitor 441 docked to protein model 5
using the clorgyline-like skeleton.
11
11.5
12
12.5
13
13.5
14
14.5
-18.77 -18.44 -14.51 -13.43 -12.77 -11.86
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
13
14
15
16
17
-17.95 -16.99 -14.40 -13.31 -13.31 -13.13
HB-Factor Sum
Estimated Free Energy (Kcal/Mol)
Free Energy Estimate vs HB-Factor Sum
56

Figure 46: Most energetically favorable pose generated by TMD of 441 with MAO A.
Energetics Analysis and Discussion
The solvation analysis done was not quite in line with the experimental data obtained by
Samir. et. al., who predicted the following in order from highest affinity to least: 359, 440, 441,
357. The solvation analysis predicted 441, 357, 359, 440. It is likely that the solvation analysis
either underestimated or overestimated certain free energy values for each pose. However, the
solvation analysis does not take into account any free energy penalties associated with the
movement of the loops reconfigured in this project. It may be possible that even though a loop can
have different possible conformations due to its flexibility, those conformations may not entirely
be energetically favorable. This is one limitation of this study as it assumes most all alternative
57
protein models are equally possible. The only exception to this comes with the binding of 357 and
359 in which only protein models 1-5 were considered. It is also worth noting that 359 in particular
generated an extremely large number of results and was also predicted to be the best inhibitor by
Samir et. al. Inhibitor 359 may be so potent because it has so many different relatively low energy
conformations it can adopt once bound to the protein. Energetically it may be favorable for an
inhibitor to remain in a low energy state with several different possible conformations to sample
versus remaining in an even lower energetically favorable state with very few poses to sample.
Lastly, all 4 dual inhibitors are subtly different in their hydrophobicity. There is likely a free energy
gain for a hydrophobic molecule such as 359 to partition into the hydrophobic binding cavity of
MAO A. This free energy gain would be much greater than the energy a more hydrophilic molecule
such as 441 would get in partitioning in the same hydrophobic cavity. This study does not calculate
any such free energy gain through hydrophobic interactions.

58
Chapter 4: Conclusions and Future Prospects

HDAC and MAO A dual inhibitors were docked using TMD in this project. Each result
from the TMD docking was then subject to a solvation analysis in which the differences in the
hydrogen bond network pre- and post-ligand binding were assessed. The solvation analysis and
the HB-factor sum score yielded enough information to choose both a singular pose and protein
model that adequately represents what each inhibitor-protein complex may look like.
From the data and past docking attempts, it is clear that MAO A itself will not adopt the
same conformation when accepting relatively smaller inhibitors into its binding cavity when
compared to bigger ligands. This is clearly seen in early docking attempts in which the inhibitors
were docked to the original crystal structure protein generating little or no results. As soon as
certain loops were moved into different conformations, more room was generated in the binding
cavity that allowed TMD to produce results for each inhibitor. This expands upon the research of
Son et. al., implying that not only are loops containing residues 93-95, 109-112, and 208-212
crucial for substrate specificity, these loops may also move to accommodate larger ligands within
the binding cavity itself
6
. This is critical for future development of inhibitors of MAO A, since
the current 2BXS crystal structure alone may be insufficient to produce in silico models of larger
and bulkier MAO A inhibitors due to the likelihood of certain loops being in different
conformations. Further still, this study shows applications such as MODELLER can be an
excellent way to explore alternative loop conformations in cases in which a protein may change
its loop configuration depending on the size of a ligand attempting to bind to the protein.
TMD itself was quite crucial in the docking process. The program provided a novel way to
model irreversible inhibitors docked to a specific cofactor in the protein, a feat unachievable using
conventional and publicly available software due to the difficulties in modelling the protein-
59
inhibitor covalent adduct and issues recognizing cofactors within the program itself. While TMD
cannot provide information on the most energetically favorable poses in each docking attempt,
TMD coupled with the solvation analysis data can be used to do so. This study therefore provides
a method by which future attempts to dock an irreversible inhibitor to a protein can prove to be
successful.
The results from this project yielded very interesting insights as to how 357, 359, 440, and
441 bind to MAO A. The MAO A binding pocket is very hydrophobic, containing several aliphatic
and aromatic residues. The design of each dual inhibitor generally allowed for a minimal amount
of hydrogen bond disruption in both the protein and inhibitor. This is because the placement and
length of the hydrophobic linker largely allowed for hydrophobic interactions to occur while
disrupting few if any hydrogen bonds. Additionally, the hydrophilic HDAC targeting moiety
seemed to be solvated towards the mouth of the binding cavity in each energetically favorable pose
generated. This suggests that the bulk water at the entrance of the binding cavity plays an important
role in interacting with the dual inhibitors, allowing the hydroxamic acid to keep itself solvated
despite being in close vicinity to hydrophobic protein residues. Future attempts to design MAO A
inhibitors (or dual inhibitors) should take note of this and consider the length of the inhibitor as
well as the placement of hydrophilic residues to take advantage of this tendency of hydrophilic
residues that interact with the bulk water at the entrance of the binding cavity.

60
References
1. Mirowska-Guzel D, Balkowiec-Iskra E. The Role of Monoamine Oxidase in Humans and Its
Metabolism. Psychiatr Ann. 2014;44(11):495-501.
2. Masoud GN, Li W. HIF-1α pathway: role, regulation and intervention for cancer therapy. Acta
Pharm Sin B. 2015;5(5):378-389.
3. Wu JB, Shao C, Li X, et al. Monoamine oxidase A mediates prostate tumorigenesis and cancer
metastasis. J Clin Invest. 2014;124(7):2891-2908.
4. Kushal S, Wang W, Vaikari VP, et al. Monoamine oxidase A (MAO A) inhibitors decrease
glioma progression. Oncotarget. 2016;7(12):13842-13853.
5. Wu JB, Lin TP, Gallagher JD, et al. Monoamine oxidase A inhibitor-near-infrared dye conjugate
reduces prostate tumor growth. J Am Chem Soc. 2015;137(6):2366-2374.
6. Son SY, Ma J, Kondou Y, Yoshimura M, Yamashita E, Tsukihara T. Structure of human
monoamine oxidase A at 2.2-A resolution: the control of opening the entry for
substrates/inhibitors. Proc Natl Acad Sci U S A. 2008;105(15):5739-5744.
7. De Colibus L, Li M, Binda C, Lustig A, Edmondson DE, Mattevi A. Three-dimensional
structure of human monoamine oxidase A (MAO A): relation to the structures of rat MAO A and
human MAO B. Proc Natl Acad Sci U S A. 2005;102(36):12684-12689.
8. Wang K, Luo J, Yeh S, et al. The MAO inhibitors phenelzine and clorgyline revert enzalutamide
resistance in castration resistant prostate cancer. Nat Commun. 2020;11(1):2689.
9. Seto E, Yoshida M. Erasers of histone acetylation: the histone deacetylase enzymes. Cold Spring
Harb Perspect Biol. 2014;6(4):a018713.
10. Glozak M, Seto E. Histone deacetylases and cancer. Oncogene. 2007;26(37):5420-5432.
61
11. Sanner MF, Olson AJ, Spehner JC. Reduced surface: an efficient way to compute molecular
surfaces. Biopolymers. 1996;38(3):305-320.
12. Pavlin M, Mavri J, Repič M, Vianello R. Quantum-chemical approach to determining the high
potency of clorgyline as an irreversible acetylenic monoamine oxidase inhibitor. J Neural Transm
(Vienna). 2013;120(6):875-882.
13. Wright MH, Sieber SA. Chemical proteomics approaches for identifying the cellular targets of
natural products. Nat Prod Rep. 2016;33(5):681-708.
14. Sali A, Blundell TL. Comparative protein modelling by satisfaction of spatial restraints. J Mol
Biol. 1993;234(3):779-815.

62
Appendix
A. Table 1: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 357 run using protein model 1.
Pose
ID
Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 4 3 1 2 2 5.52 4.52
2 5 3 1 2 1 4.26 -0.74
3 5 4 1 2 1 9.55 3.55
4 4 3 2 2 1 6.71 0.71
5 5 5 1 2 1 11.18 1.18
6 3 4 1 2 1 12.01 8.01
7 2 4 1 2 1 2.64 0.64
8 2 5 1 2 1 5.88 1.88
9 4 4 1 2 1 5.25 -0.75
10 2 4 1 2 1 10.00 8.00
11 3 5 1 2 1 5.47 -0.53
12 2 4 1 2 1 3.38 1.38
13 8 4 1 2 1 6.73 -6.27
14 5 4 1 2 1 8.89 0.89
15 6 3 1 2 5 3.19 7.19
16 5 3 1 2 1 12.61 6.61
17 4 4 1 2 1 7.47 1.47
18 3 3 1 2 1 4.84 3.84
19 3 3 1 2 1 3.32 2.32
20 4 4 1 2 2 8.67 4.67
21 5 3 1 2 1 9.06 3.06
22 4 2 2 2 5 0.32 6.32
23 4 4 2 3 1 8.33 -1.67
24 4 5 1 2 1 11.01 3.01
25 3 4 1 2 1 7.55 3.55
26 2 3 1 2 1 7.54 7.54
27 3 5 1 2 1 7.74 1.74
28 6 4 1 2 1 10.90 0.90
29 8 4 1 2 2 6.11 -5.89
30 3 1 1 2 2 8.61 13.61
31 5 2 1 2 2 10.46 8.46
32 7 2 1 2 2 8.70 2.70
63
33 5 3 1 2 2 8.95 4.95
34 7 1 1 2 5 2.94 8.94
35 4 2 1 2 2 6.34 6.34
36 3 2 1 2 2 8.97 10.97
37 2 2 1 2 3 5.65 11.65
38 4 1 2 2 2 6.77 6.77
39 5 3 1 2 2 0.70 -0.30
40 2 4 1 2 2 5.10 5.10
41 8 3 1 3 2 7.89 -3.11
42 2 2 1 2 1 6.06 8.06
43 4 1 2 2 1 8.14 6.14
44 3 2 1 2 3 6.71 10.71
45 4 2 1 2 2 9.67 9.67
46 6 2 2 2 4 0.33 1.33
47 6 1 2 2 1 5.50 0.50
48 4 4 3 2 2 5.06 -2.94
49 4 4 1 3 1 10.27 2.27
50 3 3 1 3 2 8.93 6.93
51 2 3 1 2 2 2.99 4.99
52 2 3 1 2 2 7.57 9.57
53 4 3 1 2 2 9.62 7.62
54 5 2 1 2 2 9.51 7.51
55 3 1 1 2 1 8.00 11.00
56 7 2 1 2 3 3.04 3.04
57 7 2 1 2 2 6.74 2.74
58 3 2 1 2 1 8.93 8.93

64
A. Table 2: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 357 run using protein model 2.

ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot
HB
%Mat
ch
Free
Energ
y
1 4 7 0 1 1 8.95 0.95
2 5 7 0 1 1 8.06 -1.94
3 6 6 0 1 1 7.44 -2.56
4 2 6 0 1 1 8.53 6.53
5 4 6 1 1 0 7.74 -2.26
6 5 5 1 1 1 7.87 -0.13
7 8 5 1 1 1 4.30 -9.70
8 2 6 0 1 1 4.17 2.17
9 8 4 1 1 2 5.12 -3.88
10 4 5 1 2 2 9.03 3.03
11 6 5 2 2 1 12.95 -1.05
12 3 5 2 1 2 12.04 8.04
13 3 5 0 1 2 3.67 3.67
14 5 4 1 1 3 8.85 6.85
15 7 4 1 1 4 -1.59 -3.59
16 7 4 0 1 0 6.95 -3.05
17 3 5 0 1 0 6.55 2.55

65
A. Table 3: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 357 run using protein model 3.
ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 5 6 0 2 0 9.27 -2.73
2 4 6 0 2 0 5.11 -4.89
3 5 6 1 2 0 10.13 -3.87
4 3 6 1 2 0 5.92 -4.08
5 3 6 1 2 0 4.85 -5.15
6 2 5 0 2 3 6.12 8.12
7 9 5 0 2 3 1.16 -8.84
8 3 6 0 2 3 1.49 1.49
9 4 5 0 2 0 4.55 -3.45
10 3 7 0 3 0 8.63 -3.37
11 3 6 0 3 0 5.54 -4.46
12 2 6 0 2 1 4.22 0.22
13 2 4 1 2 1 7.26 5.26
14 3 5 1 2 0 7.77 -0.23
15 8 4 0 2 0 3.89 -
10.11
16 7 3 0 2 4 2.32 4.32
17 3 3 0 2 2 3.65 5.65
18 7 3 0 2 0 6.49 -3.51
19 5 4 0 2 0 6.08 -0.92
20 6 4 0 2 0 4.15 -5.85
21 3 3 0 2 0 9.35 7.35
22 4 5 0 2 1 7.49 1.49
23 3 6 1 2 0 5.44 -4.56
24 3 5 1 2 0 8.98 0.98
25 2 5 1 2 2 6.31 4.31

66
A. Table 4: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 357 run using protein model 4.

ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 4 4 1 3 0 11.66 3.66
2 4 6 2 2 0 7.52 -6.48
3 4 6 2 3 0 11.38 -4.62
4 3 7 2 2 0 12.65 -1.35
5 3 5 1 2 0 7.97 -0.03
6 2 5 1 2 0 7.54 1.54
7 4 6 1 2 0 10.06 -1.94
8 3 7 1 2 0 9.68 -2.32
9 2 5 1 2 0 9.90 3.90
10 5 6 1 2 0 8.44 -5.56
11 6 4 1 2 5 3.31 5.31
12 4 5 2 2 0 10.94 -0.06
13 5 4 1 2 1 3.30 -3.70
14 5 4 1 2 0 4.04 -4.96
15 4 5 2 3 0 8.90 -4.10
16 4 4 1 2 3 -2.01 -0.01
17 4 4 2 3 0 6.95 -5.05
18 4 7 2 2 0 11.14 -4.86
19 3 5 1 3 0 13.60 3.60
20 2 4 1 2 0 7.57 3.57
21 3 6 1 2 0 7.49 -2.51
22 8 4 1 2 0 3.50 -
12.50
23 7 2 2 2 1 10.24 0.24
24 5 5 1 2 1 2.33 -7.67
25 7 3 1 2 5 -1.68 0.32
26 4 5 1 2 1 10.09 2.09
27 2 3 1 2 1 4.07 4.07
28 5 3 1 2 0 3.93 -1.07
29 2 5 1 2 1 3.18 -0.82
30 8 3 2 2 1 5.08 -7.92
31 2 3 1 2 1 3.73 3.73
32 4 5 2 2 1 8.23 -1.77
33 6 4 1 2 4 -2.01 -1.01
67
34 6 3 1 2 0 4.33 -4.67
35 4 4 2 3 1 5.57 -4.43
36 3 6 2 3 1 11.01 -0.99
37 3 4 2 3 1 9.48 1.48
38 2 4 1 2 1 7.46 5.46
39 2 3 2 2 0 5.02 1.02
40 3 4 2 2 0 10.10 2.10
41 4 3 2 2 1 10.10 4.10
42 7 4 2 2 3 2.33 -5.67
43 7 3 1 2 0 5.70 -6.30
44 3 4 1 2 0 5.73 -0.27

A. Table 5: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 357 run using protein model 5.

ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 8 6 0 0 2 4.00 -6.00
2 7 6 0 0 3 0.41 -3.59
3 7 5 0 0 0 6.77 -3.23
4 3 5 0 0 0 3.40 1.40

68
A. Table 6: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 357 run using protein model 6.
ID Prot-Lig
HB
Absolute
Displace
d Waters
Contact
Displaced
Hydropho
bic
Contact
Displace
d Bulk
Broken
Prot HB
%Match Free
Energy
1 1.00 5.00 1.00 1.00 5.00 1.32 11.32
2 1.00 5.00 0.00 2.00 1.00 5.28 5.28
3 1.00 5.00 2.00 1.00 1.00 4.46 2.46
4 1.00 6.00 1.00 1.00 2.00 4.82 4.82
5 2.00 4.00 1.00 2.00 2.00 5.47 5.47
6 2.00 4.00 1.00 1.00 4.00 1.31 11.31
7 1.00 5.00 1.00 1.00 4.00 0.99 10.99
8 1.00 3.00 2.00 2.00 1.00 4.44 4.44
9 1.00 5.00 1.00 1.00 1.00 6.86 6.86
10 1.00 3.00 0.00 1.00 2.00 5.73 13.73
11 4.00 6.00 0.00 1.00 2.00 6.82 2.82
12 2.00 5.00 2.00 1.00 1.00 3.19 1.19
13 1.00 4.00 0.00 1.00 1.00 6.51 10.51
14 2.00 3.00 2.00 3.00 2.00 6.65 4.65
15 2.00 4.00 0.00 2.00 2.00 4.89 6.89
16 1.00 4.00 1.00 1.00 1.00 4.21 6.21
17 1.00 4.00 0.00 1.00 0.00 4.81 6.81
18 1.00 4.00 0.00 1.00 2.00 5.81 11.81
19 2.00 3.00 1.00 3.00 2.00 4.46 4.46
20 2.00 4.00 0.00 1.00 2.00 6.17 10.17
21 1.00 3.00 0.00 1.00 0.00 2.78 6.78

69
A. Table 7: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 359 run using protein model 1 (poses 1-50).
ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 2.00 4.00 1.00 2.00 1.00 5.33 5.33
2 2.00 5.00 1.00 2.00 1.00 6.81 4.81
3 2.00 5.00 2.00 2.00 1.00 6.29 2.29
4 2.00 4.00 1.00 3.00 1.00 5.15 3.15
5 2.00 4.00 2.00 3.00 1.00 4.59 0.59
6 2.00 5.00 1.00 3.00 1.00 11.47 7.47
7 3.00 4.00 1.00 3.00 1.00 11.21 7.21
8 3.00 6.00 0.00 2.00 1.00 11.30 7.30
9 2.00 5.00 1.00 2.00 1.00 7.62 5.62
10 5.00 6.00 0.00 2.00 1.00 12.64 5.64
11 5.00 7.00 0.00 2.00 1.00 10.36 0.36
12 7.00 6.00 1.00 2.00 4.00 -0.29 -4.29
13 3.00 5.00 0.00 2.00 1.00 9.65 7.65
14 3.00 6.00 0.00 2.00 1.00 6.51 2.51
15 2.00 5.00 1.00 3.00 1.00 8.68 4.68
16 4.00 6.00 1.00 3.00 1.00 10.92 0.92
17 2.00 6.00 0.00 2.00 2.00 14.01 14.01
18 2.00 5.00 0.00 3.00 1.00 10.92 8.92
19 2.00 8.00 0.00 2.00 1.00 3.05 -2.95
20 2.00 6.00 1.00 2.00 2.00 6.29 4.29
21 2.00 5.00 1.00 3.00 1.00 9.50 5.50
22 2.00 6.00 1.00 2.00 1.00 8.33 4.33
23 2.00 6.00 1.00 2.00 1.00 9.97 5.97
24 6.00 5.00 3.00 2.00 4.00 -1.56 -6.56
25 5.00 5.00 2.00 2.00 5.00 1.28 4.28
26 6.00 5.00 1.00 2.00 2.00 8.18 1.18
27 2.00 6.00 0.00 2.00 1.00 6.01 4.01
28 2.00 6.00 0.00 2.00 1.00 7.67 5.67
29 2.00 5.00 0.00 3.00 1.00 7.76 5.76
30 8.00 6.00 2.00 2.00 5.00 1.87 -6.13
31 3.00 6.00 0.00 2.00 1.00 14.37 10.37
32 5.00 7.00 2.00 2.00 1.00 11.70 -2.30
33 3.00 5.00 1.00 2.00 1.00 9.29 5.29
34 5.00 6.00 2.00 2.00 1.00 11.57 1.57
70
35 4.00 6.00 0.00 2.00 1.00 6.11 0.11
36 2.00 6.00 0.00 2.00 1.00 6.13 4.13
37 2.00 6.00 1.00 2.00 4.00 0.32 6.32
38 2.00 3.00 0.00 2.00 1.00 7.10 11.10
39 4.00 3.00 2.00 3.00 1.00 11.68 5.68
40 5.00 5.00 1.00 2.00 1.00 11.82 3.82
41 3.00 5.00 0.00 2.00 1.00 11.56 9.56
42 2.00 4.00 2.00 2.00 4.00 -0.32 7.68
43 2.00 4.00 1.00 2.00 1.00 8.04 8.04
44 3.00 4.00 2.00 2.00 5.00 0.98 8.98
45 4.00 4.00 1.00 2.00 1.00 9.17 5.17
46 2.00 6.00 0.00 2.00 1.00 8.54 6.54
47 2.00 4.00 0.00 2.00 1.00 8.95 10.95
48 4.00 5.00 1.00 2.00 4.00 0.00 4.00
49 2.00 7.00 1.00 2.00 1.00 6.96 0.96
50 2.00 5.00 0.00 2.00 1.00 7.07 7.07

71
A. Table 8: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 359 run using protein model 1 (poses 51-100).
51 5.00 4.00 2.00 2.00 2.00 3.29 -2.71
52 4.00 6.00 0.00 2.00 1.00 7.81 1.81
53 6.00 6.00 2.00 3.00 1.00 11.24 -3.76
54 5.00 7.00 0.00 3.00 1.00 9.09 -1.91
55 7.00 6.00 2.00 2.00 5.00 2.05 -3.95
56 3.00 5.00 1.00 3.00 1.00 7.65 1.65
57 3.00 7.00 1.00 2.00 1.00 6.19 -1.81
58 2.00 5.00 1.00 3.00 1.00 6.55 2.55
59 4.00 5.00 0.00 4.00 1.00 6.61 -1.39
60 2.00 6.00 0.00 3.00 1.00 5.10 1.10
61 2.00 5.00 0.00 4.00 1.00 10.32 6.32
62 2.00 5.00 0.00 2.00 1.00 5.38 5.38
63 3.00 6.00 1.00 2.00 1.00 6.36 0.36
64 2.00 7.00 1.00 2.00 1.00 2.70 -3.30
65 3.00 6.00 0.00 3.00 1.00 8.64 2.64
66 2.00 6.00 1.00 2.00 1.00 9.06 5.06
67 5.00 6.00 1.00 2.00 1.00 6.05 -2.95
68 5.00 5.00 2.00 2.00 4.00 -2.54 -3.54
69 5.00 6.00 2.00 2.00 5.00 -1.29 -0.29
70 5.00 5.00 1.00 2.00 1.00 6.75 -0.25
71 3.00 7.00 0.00 2.00 1.00 9.79 3.79
72 4.00 5.00 2.00 3.00 1.00 8.86 -1.14
73 5.00 4.00 2.00 3.00 1.00 5.21 -4.79
74 2.00 4.00 2.00 2.00 1.00 8.09 6.09
75 5.00 5.00 3.00 2.00 1.00 12.46 0.46
76 4.00 6.00 0.00 2.00 1.00 2.90 -3.10
77 2.00 6.00 1.00 2.00 4.00 0.00 6.00
78 4.00 2.00 1.00 2.00 1.00 8.25 8.25
79 2.00 3.00 0.00 2.00 2.00 6.47 12.47
80 2.00 1.00 0.00 2.00 2.00 6.82 16.82
81 3.00 2.00 2.00 2.00 2.00 8.51 10.51
82 4.00 1.00 3.00 2.00 2.00 13.55 13.55
83 7.00 2.00 3.00 2.00 4.00 1.91 1.91
84 2.00 3.00 0.00 2.00 2.00 4.11 10.11
85 2.00 3.00 0.00 2.00 1.00 4.75 8.75
86 4.00 3.00 1.00 3.00 2.00 5.78 3.78
87 6.00 4.00 1.00 2.00 2.00 9.20 3.20
72
88 4.00 3.00 0.00 2.00 2.00 9.09 11.09
89 3.00 4.00 1.00 4.00 1.00 8.55 2.55
90 2.00 3.00 2.00 2.00 6.00 -2.92 9.08
91 4.00 2.00 0.00 2.00 2.00 9.73 13.73
92 5.00 3.00 0.00 2.00 2.00 7.14 7.14
93 4.00 3.00 0.00 3.00 2.00 11.78 11.78
94 3.00 2.00 0.00 2.00 2.00 9.74 15.74
95 2.00 2.00 0.00 2.00 2.00 5.20 13.20
96 3.00 1.00 1.00 2.00 2.00 5.45 11.45
97 3.00 2.00 0.00 2.00 2.00 9.34 15.34
98 4.00 2.00 0.00 2.00 2.00 9.60 13.60
99 2.00 3.00 0.00 2.00 2.00 5.64 11.64
100 2.00 1.00 0.00 2.00 2.00 7.59 17.59

73
A. Table 9: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 359 run using protein model 1 (poses 101-138).
ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
101 3.00 2.00 1.00 2.00 2.00 11.35 15.35
102 3.00 3.00 1.00 2.00 2.00 13.15 15.15
103 4.00 3.00 1.00 2.00 4.00 4.72 10.72
104 2.00 4.00 1.00 2.00 1.00 9.84 9.84
105 2.00 2.00 0.00 2.00 1.00 10.29 16.29
106 4.00 2.00 3.00 2.00 5.00 -2.27 5.73
107 4.00 3.00 0.00 2.00 2.00 14.24 16.24
108 5.00 4.00 1.00 2.00 4.00 4.49 6.49
109 3.00 4.00 0.00 3.00 2.00 11.49 11.49
110 3.00 3.00 0.00 3.00 2.00 11.17 13.17
111 2.00 4.00 0.00 2.00 1.00 7.67 9.67
112 3.00 4.00 0.00 3.00 1.00 10.48 8.48
113 3.00 5.00 0.00 3.00 2.00 14.49 12.49
114 6.00 3.00 2.00 2.00 6.00 2.71 10.71
115 4.00 5.00 2.00 4.00 1.00 8.53 -3.47
116 3.00 6.00 1.00 2.00 2.00 11.28 7.28
117 3.00 4.00 2.00 4.00 2.00 8.96 2.96
118 2.00 6.00 0.00 2.00 2.00 8.31 8.31
119 2.00 4.00 1.00 4.00 2.00 10.76 8.76
120 2.00 4.00 1.00 4.00 2.00 10.57 8.57
121 2.00 3.00 1.00 2.00 1.00 4.15 6.15
122 4.00 5.00 0.00 2.00 1.00 5.18 2.18
123 4.00 3.00 2.00 2.00 5.00 0.98 6.98
124 3.00 4.00 3.00 2.00 2.00 10.57 6.57
125 3.00 3.00 3.00 2.00 2.00 12.05 10.05
126 4.00 4.00 2.00 2.00 3.00 3.59 6.59
127 4.00 2.00 2.00 2.00 1.00 6.56 4.56
128 3.00 5.00 1.00 4.00 2.00 10.68 4.68
129 4.00 4.00 1.00 2.00 2.00 7.99 5.99
130 4.00 5.00 2.00 2.00 2.00 7.84 1.84
131 6.00 2.00 2.00 2.00 2.00 10.28 7.28
132 6.00 3.00 5.00 2.00 4.00 2.23 -2.77
133 4.00 2.00 1.00 2.00 1.00 12.05 12.05
134 4.00 3.00 2.00 2.00 2.00 9.27 7.27
74
135 2.00 4.00 1.00 2.00 2.00 7.14 9.14
136 2.00 2.00 1.00 2.00 4.00 2.94 18.94
137 3.00 3.00 2.00 2.00 2.00 13.07 13.07
138 2.00 2.00 0.00 2.00 3.00 -0.37 11.63

A. Table 10: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 359 run using protein model 2.

ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 3.00 6.00 1.00 0.00 0.00 7.67 3.67
2 2.00 6.00 1.00 0.00 0.00 8.00 6.00
3 2.00 7.00 0.00 0.00 1.00 8.07 8.07
4 2.00 6.00 1.00 0.00 0.00 8.15 8.15
5 2.00 6.00 0.00 1.00 0.00 11.11 9.11
6 3.00 6.00 0.00 2.00 0.00 7.78 1.78
7 2.00 7.00 0.00 0.00 0.00 5.16 3.16
8 2.00 6.00 0.00 1.00 0.00 5.54 3.54
9 4.00 7.00 1.00 0.00 0.00 8.21 0.21
10 3.00 7.00 1.00 0.00 0.00 5.83 -0.17
11 5.00 5.00 2.00 1.00 0.00 3.63 -4.37
12 2.00 5.00 2.00 0.00 0.00 7.64 5.64
13 4.00 5.00 1.00 1.00 1.00 10.27 6.27
14 3.00 7.00 1.00 3.00 1.00 7.72 -2.28
15 3.00 5.00 1.00 1.00 1.00 11.35 9.35
16 2.00 6.00 1.00 1.00 1.00 14.12 12.12
17 2.00 6.00 1.00 0.00 1.00 6.80 6.80
18 2.00 6.00 1.00 1.00 1.00 8.16 6.16
19 3.00 6.00 2.00 1.00 1.00 10.37 4.37
20 6.00 6.00 3.00 0.00 1.00 14.63 2.63
21 4.00 6.00 2.00 1.00 1.00 9.94 1.94
22 2.00 4.00 0.00 0.00 3.00 0.39 14.39

75
A. Table 11: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 359 run using protein model 3.

ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 2.00 7.00 0.00 2.00 0.00 5.11 -0.89
2 3.00 7.00 0.00 2.00 0.00 1.83 -6.17
3 2.00 6.00 0.00 2.00 0.00 3.69 -0.31
4 4.00 7.00 0.00 3.00 0.00 6.53 -5.47
5 2.00 7.00 0.00 2.00 0.00 -0.35 -6.35
6 2.00 7.00 0.00 2.00 0.00 1.42 -4.58
7 2.00 7.00 0.00 2.00 0.00 2.17 -3.83
8 2.00 9.00 0.00 2.00 0.00 5.46 -4.54
9 3.00 7.00 0.00 4.00 0.00 5.08 -6.92
10 2.00 8.00 0.00 2.00 0.00 4.66 -3.34
11 3.00 5.00 1.00 3.00 1.00 4.26 -1.74
12 4.00 5.00 1.00 2.00 0.00 4.56 -3.44
13 3.00 5.00 0.00 2.00 0.00 2.82 -1.18
14 2.00 6.00 0.00 2.00 0.00 1.46 -2.54
15 2.00 6.00 1.00 2.00 0.00 3.30 -2.70
16 2.00 5.00 1.00 3.00 0.00 6.21 0.21
17 2.00 6.00 1.00 2.00 0.00 5.48 -0.52
18 4.00 7.00 1.00 2.00 0.00 7.88 -4.12
19 4.00 6.00 2.00 3.00 0.00 8.25 -5.75
20 3.00 5.00 1.00 2.00 0.00 2.95 -3.05
21 4.00 6.00 0.00 2.00 1.00 0.00 -6.00
22 4.00 5.00 0.00 3.00 1.00 5.54 -0.46
23 2.00 5.00 0.00 3.00 2.00 2.12 2.12
24 2.00 5.00 0.00 3.00 0.00 5.59 1.59
25 2.00 5.00 0.00 3.00 0.00 5.30 1.30
26 3.00 6.00 0.00 3.00 0.00 5.69 -2.31
27 3.00 5.00 0.00 4.00 0.00 5.50 -2.50
28 3.00 7.00 0.00 3.00 0.00 2.08 -7.92
29 2.00 6.00 0.00 4.00 0.00 5.57 -2.43
30 2.00 5.00 0.00 3.00 0.00 4.62 0.62
31 2.00 5.00 0.00 3.00 1.00 2.95 0.95
32 2.00 5.00 0.00 3.00 0.00 4.64 0.64
33 2.00 5.00 0.00 2.00 2.00 2.83 4.83
34 2.00 6.00 0.00 2.00 0.00 0.35 -3.65
76
35 3.00 5.00 0.00 3.00 1.00 3.51 -0.49
36 2.00 4.00 0.00 2.00 1.00 5.71 7.71
37 2.00 5.00 0.00 2.00 0.00 4.26 2.26
38 3.00 4.00 0.00 3.00 1.00 4.98 2.98
39 3.00 6.00 0.00 3.00 0.00 4.56 -3.44
40 3.00 6.00 0.00 2.00 0.00 4.14 -1.86
41 4.00 5.00 0.00 2.00 0.00 0.00 -6.00
42 5.00 4.00 0.00 2.00 0.00 3.30 -2.70
43 2.00 4.00 0.00 2.00 0.00 0.74 0.74
44 8.00 4.00 0.00 2.00 0.00 2.55 -9.45
45 2.00 6.00 1.00 2.00 0.00 3.65 -2.35
46 2.00 5.00 1.00 2.00 0.00 -0.67 -4.67
47 2.00 5.00 1.00 2.00 0.00 3.67 -0.33
48 4.00 4.00 0.00 2.00 0.00 3.48 -0.52
49 3.00 5.00 0.00 3.00 0.00 4.18 -1.82
50 3.00 4.00 0.00 3.00 0.00 3.52 -0.48
51 2.00 4.00 1.00 2.00 0.00 5.23 3.23
52 2.00 4.00 1.00 2.00 0.00 3.34 1.34
53 2.00 4.00 1.00 4.00 0.00 5.63 -0.37
54 2.00 5.00 0.00 2.00 1.00 0.37 0.37
55 2.00 6.00 0.00 2.00 0.00 3.60 -0.40
56 3.00 5.00 0.00 2.00 0.00 2.50 -1.50
57 3.00 5.00 0.00 2.00 0.00 2.17 -1.83
58 2.00 3.00 0.00 2.00 2.00 1.28 9.28

77
A. Table 12: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 359 run using protein model 4 (poses 1-50).

ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 2.00 5.00 0.00 2.00 2.00 5.28 7.28
2 2.00 4.00 0.00 2.00 1.00 7.06 9.06
3 2.00 5.00 0.00 2.00 1.00 7.88 7.88
4 2.00 4.00 0.00 3.00 0.00 6.06 4.06
5 2.00 4.00 0.00 3.00 1.00 5.54 5.54
6 2.00 5.00 1.00 2.00 1.00 10.14 8.14
7 3.00 4.00 1.00 3.00 1.00 5.36 1.36
8 3.00 4.00 1.00 4.00 1.00 7.60 1.60
9 2.00 6.00 0.00 2.00 1.00 5.13 3.13
10 3.00 5.00 0.00 5.00 0.00 12.81 2.81
11 3.00 5.00 0.00 2.00 1.00 9.64 7.64
12 7.00 4.00 0.00 3.00 5.00 0.59 2.59
13 3.00 6.00 0.00 3.00 1.00 11.01 5.01
14 3.00 6.00 0.00 2.00 1.00 7.04 3.04
15 2.00 5.00 0.00 4.00 1.00 5.34 1.34
16 2.00 6.00 0.00 3.00 0.00 11.44 5.44
17 2.00 5.00 0.00 4.00 2.00 12.85 10.85
18 2.00 6.00 0.00 3.00 0.00 13.06 7.06
19 2.00 7.00 0.00 2.00 1.00 6.89 2.89
20 2.00 6.00 1.00 2.00 1.00 10.00 7.00
21 2.00 5.00 0.00 3.00 1.00 6.52 4.52
22 2.00 5.00 0.00 2.00 1.00 5.63 5.63
23 2.00 4.00 1.00 3.00 1.00 7.59 5.59
24 2.00 4.00 1.00 2.00 1.00 7.32 7.32
25 2.00 5.00 4.00 2.00 1.00 10.32 3.32
26 4.00 4.00 3.00 2.00 1.00 15.41 7.41
27 3.00 4.00 2.00 3.00 1.00 5.40 -0.60
28 2.00 4.00 0.00 2.00 1.00 4.62 6.62
29 4.00 4.00 0.00 3.00 1.00 7.62 3.62
30 4.00 5.00 0.00 3.00 1.00 15.38 9.38
31 3.00 4.00 0.00 3.00 1.00 12.17 10.17
32 2.00 5.00 0.00 2.00 5.00 -1.63 10.37
33 3.00 4.00 0.00 6.00 1.00 11.26 3.26
34 3.00 4.00 0.00 4.00 0.00 10.41 4.41
78
35 7.00 4.00 0.00 3.00 5.00 -0.29 1.71
36 3.00 4.00 0.00 4.00 1.00 6.43 2.43
37 3.00 6.00 0.00 5.00 1.00 7.94 -2.06
38 2.00 5.00 0.00 4.00 0.00 7.37 1.37
39 2.00 4.00 0.00 5.00 1.00 8.91 4.91
40 2.00 4.00 0.00 4.00 1.00 10.61 8.61
41 2.00 5.00 0.00 4.00 1.00 8.43 4.43
42 2.00 5.00 0.00 2.00 1.00 4.50 4.50
43 4.00 4.00 3.00 2.00 2.00 7.99 1.99
44 2.00 4.00 1.00 3.00 0.00 10.50 8.50
45 3.00 5.00 1.00 3.00 0.00 5.88 -2.12
46 2.00 5.00 0.00 2.00 1.00 4.67 4.67
47 3.00 4.00 4.00 4.00 0.00 12.46 -1.54
48 4.00 5.00 4.00 3.00 1.00 11.66 -2.34
49 4.00 4.00 2.00 3.00 1.00 7.72 -0.28
50 5.00 5.00 3.00 2.00 1.00 7.52 -4.48

79
A. Table 13: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 359 run using protein model 4 (poses 51-85).
ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot
HB
%Mat
ch
Free
Energ
y
51 2.00 5.00 3.00 2.00 0.00 5.56 -2.44
52 2.00 4.00 0.00 2.00 2.00 6.98 10.98
53 2.00 3.00 0.00 2.00 2.00 3.49 9.49
54 4.00 4.00 0.00 3.00 2.00 2.47 0.47
55 6.00 5.00 0.00 2.00 2.00 5.61 -0.39
56 3.00 4.00 0.00 4.00 2.00 11.18 9.18
57 3.00 3.00 0.00 4.00 2.00 4.75 4.75
58 2.00 4.00 0.00 2.00 7.00 -3.68 14.32
59 4.00 2.00 0.00 3.00 2.00 7.40 9.40
60 5.00 4.00 0.00 3.00 2.00 7.12 3.12
61 3.00 3.00 0.00 4.00 2.00 12.15 12.15
62 3.00 2.00 0.00 4.00 2.00 8.68 10.68
63 2.00 2.00 0.00 5.00 2.00 9.24 11.24
64 2.00 4.00 0.00 3.00 2.00 6.69 9.69
65 2.00 3.00 0.00 5.00 2.00 11.91 11.91
66 3.00 4.00 0.00 4.00 3.00 11.91 11.91
67 3.00 3.00 0.00 5.00 2.00 10.53 8.53
68 7.00 4.00 0.00 3.00 6.00 1.79 3.79
69 4.00 3.00 0.00 4.00 2.00 6.89 4.89
70 3.00 4.00 0.00 4.00 2.00 3.66 1.66
71 3.00 4.00 0.00 4.00 2.00 7.58 5.58
72 2.00 3.00 0.00 6.00 2.00 11.27 9.27
73 2.00 4.00 1.00 4.00 2.00 8.60 6.60
74 2.00 4.00 0.00 5.00 1.00 8.22 4.22
75 2.00 6.00 0.00 2.00 2.00 6.43 6.43
76 4.00 5.00 1.00 2.00 1.00 5.86 -0.14
77 5.00 4.00 3.00 2.00 6.00 -1.64 0.36
78 2.00 4.00 2.00 4.00 2.00 8.38 5.38
79 5.00 4.00 3.00 3.00 2.00 7.51 -2.49
80 2.00 4.00 3.00 4.00 1.00 9.28 2.28
81 5.00 4.00 2.00 2.00 2.00 5.98 -0.02
82 3.00 4.00 2.00 4.00 1.00 10.62 2.62
83 3.00 5.00 2.00 5.00 2.00 13.24 3.24
84 4.00 5.00 2.00 3.00 2.00 4.13 -3.87
80
85 2.00 2.00 0.00 2.00 4.00 0.00 18.00

A. Table 14: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 359 run using protein model 5.
ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 2.00 6.00 0.00 1.00 4.00 -0.77 7.23

81
A. Table 15: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 359 run using protein model 6.
ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot
HB
%Mat
ch
Free
Energ
y
1 1.00 5.00 2.00 2.00 0.00 7.69 3.69
2 6.00 5.00 4.00 2.00 1.00 5.42 -9.58
3 0.00 0.00 0.00 0.00 0.00 #DIV
/0!
#DIV
/0!
4 2.00 6.00 1.00 2.00 4.00 -1.68 4.32
5 2.00 5.00 0.00 2.00 2.00 10.26 13.26
6 3.00 5.00 0.00 1.00 0.00 7.29 6.29
7 2.00 6.00 0.00 2.00 2.00 0.32 0.32
8 5.00 6.00 0.00 2.00 0.00 4.23 -5.77
9 1.00 6.00 0.00 2.00 2.00 7.83 9.83
10 1.00 7.00 0.00 2.00 1.00 8.63 6.63
11 1.00 6.00 1.00 2.00 1.00 6.01 4.01
12 2.00 6.00 1.00 2.00 2.00 3.44 1.44
13 1.00 5.00 1.00 2.00 1.00 4.13 4.13
14 2.00 5.00 1.00 2.00 2.00 3.31 3.31
15 2.00 4.00 0.00 2.00 0.00 10.00 11.00
16 3.00 4.00 0.00 1.00 1.00 9.68 11.68
17 2.00 5.00 0.00 1.00 3.00 3.50 9.50
18 5.00 5.00 0.00 1.00 2.00 6.08 4.08
19 1.00 5.00 0.00 1.00 1.00 8.46 13.46
20 1.00 6.00 0.00 1.00 1.00 8.08 10.08
21 1.00 5.00 0.00 1.00 0.00 4.78 6.78
22 2.00 5.00 1.00 1.00 0.00 5.66 3.66
23 1.00 4.00 0.00 1.00 1.00 5.75 11.75
24 2.00 4.00 1.00 2.00 1.00 6.96 6.96
25 2.00 4.00 0.00 1.00 1.00 2.81 6.81
26 3.00 6.00 0.00 1.00 1.00 8.06 6.06
27 1.00 4.00 0.00 3.00 1.00 8.96 10.96
28 1.00 5.00 0.00 1.00 1.00 10.95 14.95
29 1.00 4.00 0.00 3.00 0.00 5.34 5.34
30 1.00 4.00 0.00 2.00 1.00 7.44 11.44
31 1.00 6.00 0.00 1.00 0.00 6.34 6.34
32 2.00 4.00 0.00 1.00 2.00 2.77 8.77
33 1.00 6.00 1.00 2.00 3.00 6.73 8.73
34 3.00 5.00 0.00 3.00 0.00 6.75 0.75
82
35 1.00 5.00 1.00 2.00 1.00 6.83 6.83
36 4.00 5.00 1.00 1.00 1.00 6.29 2.29
37 2.00 4.00 1.00 1.00 1.00 4.14 6.14
38 2.00 4.00 0.00 1.00 3.00 3.46 11.46
39 3.00 6.00 0.00 1.00 1.00 9.31 7.31
40 1.00 4.00 0.00 2.00 0.00 4.80 6.80
41 1.00 5.00 0.00 1.00 0.00 10.71 12.71
42 1.00 4.00 0.00 2.00 1.00 5.67 9.67
43 1.00 4.00 0.00 1.00 1.00 8.96 14.96
44 1.00 6.00 0.00 1.00 2.00 8.21 12.21
45 2.00 4.00 0.00 1.00 0.00 3.08 5.08
46 1.00 5.00 1.00 1.00 1.00 3.98 5.98
47 3.00 5.00 0.00 1.00 1.00 7.67 7.67
48 1.00 5.00 1.00 1.00 1.00 7.50 9.50
49 4.00 5.00 0.00 1.00 0.00 3.16 -0.84

A. Table 16: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 440 run using protein model 6.
ID Prot-
Lig
HB
Contact Displaced
Hydrophobic
Contact
Displaced
Bulk
Contact
SWB
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 2.00 1.00 3.00 9.00 1.00 9.01 0.01
2 2.00 1.00 2.00 8.00 2.00 7.61 2.61
3 2.00 2.00 4.00 9.00 0.00 10.11 -4.89
4 2.00 2.00 2.00 9.00 0.00 9.27 -1.73
5 2.00 2.00 2.00 7.00 2.00 7.89 0.89
6 2.00 2.00 3.00 8.00 1.00 9.58 -1.42

83
A. Table 17: Raw data of the solvation analysis of TMD generated poses of the optimized
backbone 441 run using protein model 6.
ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 3.00 5.00 3.00 3.00 1.00 9.85 -7.15
2 3.00 5.00 2.00 2.00 0.00 10.15 -4.85
3 2.00 5.00 2.00 2.00 1.00 7.96 -3.04
4 2.00 5.00 2.00 2.00 1.00 10.62 -0.38
5 3.00 5.00 3.00 3.00 1.00 11.61 -5.39
6 3.00 5.00 3.00 2.00 1.00 12.20 -2.80
7 3.00 5.00 4.00 2.00 1.00 10.85 -6.15
8 3.00 5.00 4.00 3.00 0.00 11.08 -9.92
9 3.00 5.00 3.00 2.00 1.00 10.26 -4.74
10 3.00 5.00 4.00 3.00 0.00 10.78 -
10.22

84
A. Table 18: Raw data of the solvation analysis of TMD generated poses of the clorgyline-based
backbone 441 run using protein model 1.
ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 8 5 3 3 1 7.255
5205
-
19.74
4479
2 8 5 3 3 1 7.255
5205
-
19.74
4479
3 8 5 3 3 1 8.176
10063
-
18.82
3899
4 8 5 3 3 1 9.119
49686
-
17.88
0503
5 8 3 3 3 2 9.810
12658
-
11.18
9873
6 8 3 3 3 2 10.12
65823
-
10.87
3418

85
A. Table 19: Raw data of the solvation analysis of TMD generated poses of the clorgyline-based
backbone 441 run using protein model 2.
ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot
HB
%Mat
ch
Free
Energ
y
1 8.00 7.00 3.00 3.00 1.00 6.65 -
24.35
2 8.00 7.00 3.00 3.00 1.00 6.33 -
24.67
3 8.00 6.00 3.00 3.00 2.00 7.62 -
19.38
4 8.00 6.00 3.00 3.00 1.00 6.98 -
22.02
5 8.00 5.00 3.00 3.00 2.00 10.86 -
14.14
6 8.00 5.00 3.00 3.00 2.00 10.54 -
14.46

A. Table 20: Raw data of the solvation analysis of TMD generated poses of the clorgyline-based
backbone 441 run using protein model 3.
ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 8.00 8.00 0.00 2.00 0.00 8.25 -
18.75
2 8.00 8.00 0.00 2.00 0.00 6.35 -
20.65
3 8.00 7.00 0.00 2.00 1.00 8.63 -
14.37
4 8.00 7.00 0.00 2.00 0.00 8.63 -
16.37
5 8.00 7.00 0.00 3.00 0.00 4.56 -
22.44
6 8.00 7.00 0.00 3.00 0.00 5.54 -
21.46

86
A. Table 21: Raw data of the solvation analysis of TMD generated poses of the clorgyline-based
backbone 441 run using protein model 4.
ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot
HB
%Mat
ch
Free
Energ
y
1 8.00 5.00 2.00 2.00 1.00 4.56 -
18.44
2 8.00 5.00 2.00 2.00 1.00 4.23 -
18.77
3 8.00 5.00 2.00 2.00 2.00 6.49 -
14.51
4 8.00 5.00 1.00 2.00 2.00 7.14 -
11.86
5 8.00 4.00 1.00 2.00 1.00 5.57 -
13.43
6 8.00 4.00 2.00 2.00 2.00 6.23 -
12.77

A. Table 22: Raw data of the solvation analysis of TMD generated poses of the clorgyline-based
backbone 441 run using protein model 5.

ID Prot-
Lig
HB
Absolute
Displaced
Waters
Contact
Displaced
Hydrophobic
Contact
Displaced
Bulk
Broken
Prot HB
%Mat
ch
Free
Energ
y
1 8.00 7.00 2.00 1.00 1.00 8.01 -
16.99
2 8.00 7.00 2.00 1.00 1.00 7.05 -
17.95
3 8.00 6.00 2.00 1.00 1.00 9.87 -
13.13
4 8.00 6.00 2.00 1.00 1.00 8.60 -
14.40
5 8.00 6.00 2.00 1.00 2.00 7.69 -
13.31
6 8.00 6.00 2.00 1.00 2.00 7.69 -
13.31

Abstract (if available)

Abstract In silico molecular docking is a crucial step in the drug discovery process as it allows informed decisions to be made on how a drug may be designed based on predicted binding conformations leading to favorable drug-protein interactions. However, most computational docking programs are only able to generate accurate models of reversible inhibitors. Methods to model irreversible inhibitors are less common. This study aims to model a set of novel monoamine oxidase A (MAO A) / histone deacetylase (HDAC) dual inhibitors that bind covalently to the cofactor in the MAO A enzyme. In order to generate models of these inhibitors in the MAO A active site, an in-house program, TMD (Tethered Molecular Docking), was used to survey the potential binding poses of a small library of the inhibitors. In order to assess the energetic viability of the poses generated by TMD, the poses were subjected to a solvation analysis using data from a second in-house program, Watgen5. The results show that the optimal poses of the protein-bound dual inhibitors require desolvation of the MAO A binding cavity to drive ligand binding. This project demonstrates a method by which irreversible inhibitors can be modelled with reasonable accuracy. In addition, this work provides an example of the importance of water desolvation in the ligand binding process.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Inhibition of MAO-A by Dual MAO-A/HDAC inhibitors: in silico approach for ligand binding and affinity prediction

PDF

Integration of KNIME and molecular docking for evaluation of tau fibril inhibitors

PDF

Potential therapeutic effect of monoamine oxidase (MAO) inhibitor on human neuroblastoma

PDF

Monoamine oxidase and cancer

PDF

NMI (near-infrared dye conjugate MAO A inhibitor) outperformed FDA-approved prostate cancer drugs with a unique mechanism based on bioinformatic analysis of NCI60 screening data

PDF

Monoamine oxidase deficiency and emotional reactivity: neurochemical and developmental studies

PDF

Molecular docking of sulfonylureas to the SUR1 receptor

PDF

Computational model for predicting ionic solubility

PDF

Modeling of water molecules in protein-ligand binding

PDF

NMI: a near infrared conjugated MAO-A inhibitor as a novel targeted therapy for colorectal and other cancers

PDF

Differential regulation of monoamine oxidase A and B genes

PDF

The role of monoamine oxidase in behavioral plasticity

PDF

MAO a deficient mice exhibit an altered immune system in the brain and prostate

PDF

Solvation as a driving force for peptide docking to the major histocompatibility complex (MHC) class II molecules

PDF

Monoamine oxidase inhibitors regulate tumorigenesis and mitochondrial function in a prostate cancer mouse model

PDF

Computer modeling of protein-peptide interface solvation

PDF

Monoamine oxidase A inhibitors and androgen receptor antagonists regulate mitochondrial function in prostate cancer cells

PDF

Structure-based computational analysis and prediction of TCR CDR3 loops in the TCR-peptide-MHC complex using solvation parameters and peptide molecular dynamics.

PDF

Artificial intelligence in medicinal chemistry and drug discovery

PDF

Computational modeling of solvation and docking of peptide-MHC class I

Asset Metadata

Creator Asbell, Thomas Russell, IV (author)

Core Title Inhibition of monoamine oxidase A and histone deacetylase inhibitors: computational prediction of ligand binding

School School of Pharmacy

Degree Master of Science

Degree Program Molecular Pharmacology and Toxicology

Degree Conferral Date 2021-08

Publication Date 07/18/2021

Defense Date 07/15/2021

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Chimera,HDAC,histone deacetylase,MAO A,molecular docking,monoamine oxidase,OAI-PMH Harvest,solvation,TMD,WATGEN

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Haworth, Ian (committee chair), Romero, Rebecca (committee chair), Shih, Jean (committee chair)

Creator Email tasbell@usc.edu,trasbell4@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC15610959

Unique identifier UC15610959

Legacy Identifier etd-AsbellThom-9776

Document Type Thesis

Format application/pdf (imt)

Rights Asbell, Thomas Russell, IV

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu