University of Southern California Dissertations and Theses

Targeting human base excision repair as a novel strategy in cancer therapeutics

(USC Thesis Other)

Targeting human base excision repair as a novel strategy in cancer therapeutics

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content TARGETING HUMAN BASE EXCISION REPAIR AS A NOVEL STRATEGY IN
CANCER THERAPEUTICS

by

Zahrah Zawahir

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PHARMACEUTICAL SCIENCES)

May 2009

Copyright 2009 Zahrah Zawahir

ii
Dedication

This work is dedicated to my grandparents. My grandfather, role model, and best
friend, M.A.S. Mohideen – his faith in me knew no bounds. His wisdom, his love,
and his support – even though he is no longer here to witness this milestone – will
inspire me in all future endeavors. My grandmother, teacher, and mentor, R.
Mohideen – her wisdom and insight continue to be a source of comfort to me to this
day.

iii
Acknowledgements

To my parents, Muaaz and Roshan Mubarak, for their love and kindness, and my
uncle, Sabry Mohideen, for his patience, and kind advice.

To Tariq Mohamed Ali, for his kindness and support.

To my committee members: Dr. Ian Haworth, Dr. Robert Ladner, Dr. Clay Wang,
and Dr. Nouri Neamati for their time and guidance.

To my friends/colleagues: Dr. Laith Q. Al-Mawsawi, Dr. Raveendra Dayam, Dr.
Jinxia Deng, Dr. Huabiao Zhou, Dr. Peter Wilson, Dr. William Fazzone, Dr. Wu
Zhang, Dr. Dong Yun Yang, Tino W. Sanchez, Melissa LaBonte, Laleh Taheri,
Yumna Shabaik, and all of my numerous labmates and colleagues for their friendship
and for making the lab a wonderful place to work.

To Dr. Nouri Neamati, my advisor, for giving me this opportunity when I was an
undergraduate student, and for helping me see it through to the end.

iv
Table of Contents

Dedication ii
Acknowledgements iii
List of Tables vi
List of Figures viii

Abstract xi

Chapter 1: Introduction
1.1 Current treatment strategy for colorectal cancer in the United States 1
1.2 DNA repair proteins as molecular targets in cancer therapy 2
1.3 Base Excision Repair in Mammalian Cells 3
1.4 Biological consequences of abasic sites 5
1.5 Substrate based interruption of BER 6
1.6 APE1/ Ref-1: Catalytic processes 7
1.7 Functional and structural aspects of APE1 9
1.8 Basis for targeting APE1/ Ref-1 for cancer therapeutics 11
1.9 Interactions of APE1 with other proteins in BER 13
1.10 Current progress in design of small molecule inhibitors to APE1 14
1.11 Computational approaches in development of small-molecule inhibitors 18

Chapter 2: Design and optimization of a Catalytic Assay for APE1
2.1 Tetrahydrofuran as a synthetic analog of an abasic site 22
2.2 Optimization of Escherichia coli expression system and
protein purification protocol for APE1 24
2.3 Optimization of experimental conditions for inhibitor screening 27
2.4 Behavior of 7-nitro-indole carboxylic acid with APE1 30
2.5 Conclusions and implications 32

Chapter 3:
3.1 Design and generation of pharmacophore models based on the 35
catalytic active site of APE1
3.2 Database search and compound selection 39
3.3 Selective inhibition of APE1 catalytic activity by
structurally diverse small molecules 41
3.4 Selective inhibition of APE1 catalysis versus inhibition of its
bacterial homolog and other DNA-binding enzymes 57

v
3.5 Conclusions and implications 62

Chapter 4: Discovery of a novel class of inhibitors to APE1
4.1 Bioisosteres of carboxylic acids 65
4.2 Optimization of tetrazole-containing compounds as
small molecule inhibitors of APE1 68
4.3 Conclusions and implications 72

Chapter 5: Biochemical validation of APE1 as a therapeutic target
5.1 APE1 polymorphisms in cancer disease studies 74
5.2 The APE1 D148E in colorectal cancer survival 76
5.3 Biochemical models to characterize potential inhibitor
behavior in clinically contingent pharmacological backgrounds 79
5.4 Identification of a role for BER in oxaliplatin toxicity 85
5.5 Conclusion and implications 88

Chapter 6: Materials and methods
6.1 Methods used for APE1 catalytic assay development 90
6.2 Computational methods 94
6.3 Molecular Biology protocols 96

References 105

Appendices 123
Appendix A: Moderately Active and Inactive Compounds from Chapter 3 125
Appendix B: Colony Formation Assays for APE1 Inhibitory Molecules 144
Appendix C: Peptide Inhibition of HIV-1 Integrase 150

vi
List of Tables

Table 1: Small-molecule inhibitors of DNA repair 17

Table 2: APE1 inhibition by pharmacophore-retrieved molecules 44

Table 3: APE1 inhibitory activity of analogs of compounds 1 and 2 47

Table 4: APE1 inhibitory activity of analogs of compound 5 50

Table 5 APE1 inhibitory activity of analogs of compounds 10 and 11 52

Table 6: APE1 inhibitory activity of analogs of compound 17 55

Table 7: Relative inhibitory activity of APE1 inhibitors with other proteins 58

Table 8: Docking scores of the most potent molecules for APE1 and ExoIII 59

Table 9: In silico docking of compounds for APE1 inhibition 70

Table 10: Tetrazole compounds yielded by substructure search 71

Table 11: Summary of APE1 polymorphisms and their impact 75

Table 12: Summary of APE1 localization patterns 76

Table 13: Distribution of APE1 genotypes by gender 77

Table A1: Moderately active compounds retrieved by pharmacophore 125

Table A2: Inactive compounds retrieved by pharmacophore 128

Table A3: Compounds retrieved using pharmacophore model A3H1 133

Table A4: Inactive compounds retrieved using pharmacophore A3H1 133

Table C1: Inhibition of IN activity by PR Peptides of Pol polyprotein 155

Table C2: Inhibition of IN activity by RT Peptides of Pol polyprotein 156

Table C3: Inhibition of IN activity by RNAse H peptides of
Pol polyprotein 159

vii

Table C4: Inhibition of IN activity by IN peptides of Pol polyprotein 160

Table C5: Catalytic and replication activities of IN mutants 167

Table C6: Sequence and IN inhibitory activity of designed peptides 170

Table C7: HIV-1 IN inhibitory activity of Ala-scanned analogs 172

Table C8: HIV-1 IN activity of single-substituted analogs 174

Table C9: HIV-1 IN activity of truncated analogs 175

Table C10: Alpha-helical content of NL6 and NL9 178

Table C11: HIV-1 IN activity of optical isomers of NL6 and NL9 179

Table C12: Alpha-helical content of bridged peptides 191

Table C13: Inhibition of IN by stapled and unstapled peptides 194

Table C14: Inhibition of replication of HIV-1 and native cytotoxicity 195

Table C15: Inhibition of IN catalysis by ‘open’ or ‘closed’ peptides 196

viii
List of Figures

Figure 1: The base excision repair pathway 4

Figure 2: Structure of the abasic site 7

Figure 3: APE1 catalysis 9

Figure 4: Complex of APE1 active site and abasic DNA fragment 9

Figure 5: Functional domain organization of APE1 11

Figure 6: Pharmacophore approach to inhibitor design 18

Figure 7: HIV-1 IN peptide inhibition 21

Figure 8: Structure of abasic site and tetrahydrofuran 22

Figure 9: Representative denaturing polyacrylamide gel of APE1 activity 23

Figure 10: Plasmid map for pQE30 expression vector 24

Figure 11: Transformant selection for maximum protein yield 26

Figure 12: Imidazole elution of APE1 27

Figure 13: Determination of optimal concentration of APE1 for assay 28

Figure 14: APE1 robustness at very low concentrations and high temp. 28

Figure 15: Effect of incubation time on APE1 cleavage activity 29

Figure 16: Enzyme concentration effect on inhibitor behavior 30

Figure 17: Scheme of APE1 catalytic assay 33

Figure 18: APE1 interactions with the abasic site fragment 35

Figure 19: Three-dimensional pharmacophore models 38

Figure 20: Shape-merged pharmacophore model 39

ix
Figure 21: Venn diagram with compounds retrieved from models 41

Figure 22: Polyacrylamide gel image showing most potent molecules 43

Figure 23: Compound 1 mapped onto pharmacophore H1NI2 49

Figure 24: Compound 21 mapped onto pharmacophore A1NI2 54

Figure 25: Compound 17 mapped onto pharmacophore A3NI1 56

Figure 26: Compound 10 at different concentrations of APE1 61

Figure 27: Bioisosteres of carboxylic acids 66

Figure 28: In silico screening protocol for discovery of novel APE1 inhibitors 69

Figure 29: The DD genotype confers a significant survival benefit 78

Figure 30: Gene regulation in the Tet-Off system 80

Figure 31a: Map of the pTet-Off regulator plasmid 82

Figure 31b: Map of pTre-Tight response plasmid and multiple cloning site 82

Figure 32: Validation of real-time PCR primers 83

Figure 33: APE1 mRNA expression in a panel of cell lines 84

Figure 34a: MX slightly potentiates the cytotoxicity of 5FU 86

Figure 34b: MX protects HCt116 p53 wild-type cells to oxaliplatin 86

Figure 34c: MX potentiates cytotoxicity in HCT116 p53 null cells 87

Figure 35: The protective effective of MX is more apparent
at longer drug exposure 87

Figure 36: Cell viability assays show the same result as colony formation assays 88

Figure B1: Colony formation assay with gemictabine 144

Figure B2: Colony formation assay with gemcitabine and lead compounds 145

x
Figure B3: Colony formation assay with methotrexate 147

Figure B4: Colony formation assay with methotrexate and lead compounds 148

Figure C1: The most potent peptides derived from HIV-1 HXB2 Pol 154

Figure C2: Representative image of IN in vitro assay and active compounds 157

Figure C3a: Mechanism of Schiff-base assay 161

Figure C3b: Schiff-base cross-linking experiments 162

Figure C4: Functional motifs of HIV-1 IN 169

Figure C5: The chemical structures of NL6 and NL9 171

Figure C6: Isomeric forms of NL6 and Nl9 show varying activity against IN 173

Figure C7: CD spectra of NL6 and NL9 177

Figure C8: Inhibitory activity of NL6 and NL9 179

Figure C9: Interactions of HIV-1 IN 183

Figure C10: Optional positions on NL6 for side chain bridging 190

Figure C11: CD spectra of NL6 and NLH series peptides 192

Figure C12: Strategy for design of cell permeable nano-needles 199

xi
Abstract

APE1 is an attractive target for the rational design of small-molecule inhibitors in the
field of oncological therapeutic research. It is an essential enzyme in the mammalian
base excision repair pathway, and works in conjunction with other cellular proteins
in the repair of abasic sites within the genome. This enzyme has been implicated in
the resistance of tumors to current chemotherapeutic agents. Extensive effort is
underway in the identification and development of clinically suitable inhibitors to
this essential enzyme. Herein, we present the discovery of a series of inhibitors to
APE1, as well as the identification of a novel class of bioisosteric compounds that
inhibit the catalytic activity of the enzyme. We have also performed preliminary
work in the development and optimization of biochemical models of APE1
expression regulation. This is an ongoing effort to characterize appropriate cellular
systems in colon cancer for the development of cellular APE1 inhibitors that can be
measured using valid biochemical endpoints. Interesting observations have been
made in this regard, for the first time, also implicating the base excision repair
pathway in the repair of platinum-agent induced DNA damage. Finally, as validation
for APE1 as a therapeutic target, we studied the impact of single nucleotide
polymorphisms in two populations of colorectal cancer patients, and found that the
D148E polymorphism confers a measure of survival to males possessing it.

1
Chapter 1: Introduction

1.1 Current treatment strategy for colorectal cancer in the United States

The most common treatment strategies for colorectal cancer to date, and for the past
40 years, include the fluoropyrimidine compound, 5-fluorouracil (5FU). Current
regimens are 5FU-based and include effective combination agents such as irinotecan
(topoisomerase I inhibitor), the platinum agent oxaliplatin, and leucovorin, a folic
acid derivative known to synergistically enhance the effectiveness of 5FU
combination therapy. The response rate of colorectal cancers to single-agent 5FU
therapy, and thereafter to combination therapy has improved from 10% to 50%, also
due to optimization of drug administration scheduling
89
. Other novel agents used in
the treatment of colorectal cancer include monoclonal antibodies such as cetuximab,
that targets the epidermal growth factor receptor; and bevacizumab, targeting the
vascular endothelial growth factor
66
. These biological agents, used in conjunction
with 5FU-based regimens, have shown additional clinical benefit in metastatic
colorectal cancer populations. However, even given the emergence of new effective
chemotherapeutic agents, and given that 5FU continues to remain the mainstay of
therapeutic regimens for the treatment of colorectal and other cancers, approximately
50% of cancer patients treated with 5FU will derive no benefit. Drug resistance
remains a significant limitation to the clinical use of 5FU, underscoring the need for
identification of novel therapeutic targets to exploit in the continued effort to treat
cancer.

2
1.2 DNA repair proteins as molecular targets in cancer therapy
An emerging concept in the identification of novel chemotherapeutic targets is the
design of small-molecule based inhibitors intended to sensitize tumor cells to DNA-
damaging agents. Many chemotherapeutic agents exert their toxicity by way of DNA
damage mechanisms, including 5FU (incorporated as uracil into DNA, as one of its
cytotoxic mechanisms), oxaliplatin (strand breaks and crosslinks within DNA) and
temozolomide (methylation of guanine within the DNA). While it may seem
counter-intuitive to inhibit enzymes that impede the acquisition of mutations within
an already genetically unstable tumor cell, a threshold exists in tumor cells in which
a lack of DNA repair results in cell death
11
. In addition, cancer cells have been
shown to alter the expression of repair proteins in an attempt to escape the
deleterious effect of DNA damaging agents, enhancing DNA repair and thus
contributing to the resistance that is developed to these agents.
This concept can be applied to two front-line approaches to cancer treatment:
chemotherapy and radiation treatment. In both cases, the cytotoxicity associated with
the DNA damage caused is outweighed by recurring resistance to the causative
agent. The careful identification of specific DNA repair proteins is increasingly
appearing to be a viable approach to in the growing array of technologies available to
researchers and clinicians in the treatment of cancer. Many recent studies have
provided proof-of-concept results that show that selective targeting of DNA repair
proteins has the potential to augment currently used chemotherapeutic and radiation
agents. Some of the targets identified, such as PARP-1, have resulted in the

3
development of effective single-agent anti-tumor small molecules. Others have been
identified as targets that, when inhibited, will serve to sensitize cells to the relevant
agent, exerting a synergistic toxicity that overcomes resistance. The human apurinic/
apyrimidinic endonuclease 1 (APE1) is one such enzyme.

1.3 Base excision repair in Mammalian Cells
In mammalian cells, reactive oxygen species, alkylation, deamination and other
processes of cellular metabolism and/or exogenous drug toxicity can cause
endogenous mutagenic and cytotoxic DNA base lesions that are predominantly
repaired by the base excision repair (BER) pathway (Figure 1). BER is initiated by
lesion-specific glycosylases that excise the damaged base from the sugar-phosphate
backbone by cleaving the glycosidic bond. Examples of ubiquitous glycosylases that
often have redundant functionality include uracil DNA glycosylase (UDG), thymine
DNA glycosylase (TDG), oxoguanine glycosylase (OGG1) and single-strand
selective monofunctional uracil-DNA glycosylase (SMUG1). These glycosylases are
themselves attractive anti-cancer targets for development of small-molecule
inhibitors. Glycosylase-mediated base removal results in a potentially cytotoxic
apurinic/ apyrimidinic (AP) site intermediate which becomes the substrate for the
major human AP endonuclease (APE1, also known as Ref-1)
65
. APE1 is a
fundamental protein in this essential repair pathway, and is thought to be responsible
for >95% of total AP endonuclease activity in human cell culture extracts
22, 65
. The
enzyme belongs to the highly conserved ExoIII/ Xth family of endonucleases, of

4
which the prototypical Escherichia coli homolog exonuclease III is a member. A
second family of AP endonucleases, EndoIV, contains the E. coli endonuclease IV,
for which no human counterpart appears to exist. The other mammalian enzyme
responsible for AP site repair is APE2 which belongs to the APE2/APN family of
enzymes, a subclass of the ExoIII/ Xth family, bearing greatest similarity to the
Saccharomyces cerevisiae APN2 gene product
58, 121, 159
.

Figure 1. The base excision repair (BER) pathway is initiated by glycosylases
specific to the type of damage. Removal of the damaged bases results in an
abasic site, that is processed by APE1 (red arrow) to permit DNA polymerase ß
mediated repair and replication. Several other proteins are recruited to the site by
APE1 and Pol ß . The substrate moves along the BER pathway in a ‘passing-the-
baton’ manner.

5
Mammalian base excision repair consists of two sub-pathways: short patch (SP – the
major pathway) and long-patch (LP – the minor pathway). This classification has
been made in regard to the length of the DNA patch that is repaired. In short patch
BER, the patch consists of a single excised nucleotide: the resulting AP site
processed by APE1, and subsequent filling by polymerase ß. The patch is ligated by
DNA ligase III in conjunction with the scaffolding protein XRCC1. The initial stages
of long-patch BER duplicate those of short-patch BER, up to processing of the break
by APE1. However, the length of the patch extends to several nucleotides (between
6-13), and the resulting gap may be filled by polymerase ß, but also polymerases δ/ ε.
Additional required enzymes are proliferating cell nuclear antigen (PCNA) and flap
endonuclease 1 (FEN1)
65, 128
.

1.4 Biological consequences of abasic sites
Abasic sites (apurinic/ apyrimidinic) are non-coding, and can be cytotoxic and
mutagenic. These lesions represent a major proliferative threat to the cell. On
average, about 2 million abasic lesions are generated in a living mammalian cell in a
20-hour period
9
. Chemical forms of the natural abasic site are in an equilibrium
mixture of racemeric hemiacetals, an aldehyde form and a hydrated aldehyde (Figure
2)
159
. The best example of the cytotoxicity of abasic sites is in bacterial cells
containing a temperature-sensitive mutation in the gene for deoxyuridine
triphosphate (dUTPase)
141
. When dUTPase is inactive, at non-permissive
temperatures, dUTP levels are elevated and uracil is incorporated at an increased rate

6
into DNA. Removal of the uracil by uracil DNA glycosylase results in abasic sites
being generated. If a defect in abasic site repair is introduced, the cells are no longer
viable. It is presumed that RNA and DNA polymerases dissociate on encountering an
abasic lesion, which, if left unrepaired, results in strand breaks and aborted
transcriptional events. The cytotoxicity of abasic sites may also be due to their ability
to affect the DNA cleavage activity of topoisomerases by irreversibly trapping
topoisomerase-DNA covalent complexes
159
.

1.5 Substrate-based interruption of BER
As mentioned, the initial steps of BER result in the formation of a non-coding lesion
generated via the spontaneous, chemically-induced, or glycosylase-mediated
hydrolysis of the information containing purine or pyrimidine base from the DNA
backbone. Substrate-based interruption of BER involves the reaction of a
nucleophilic reagent, methoxyamine (MX) that specifically reacts with abasic sites.
MX is an alkoxyamine derivative that blocks APE1 cleavage of abasic sites
88
. The
cytotoxicity caused by this reaction has been shown to potentiate the DNA-damaging
mechanism of temozolomide, a front-line chemotherapeutic agent, in colon, breast
and ovarian cancer cells, as well as in mouse xenograft models
49
,
123
,
86
.
Temozolomide alkylates the N7-guanine in DNA, and attempted repair of this lesion
causes the generation of abasic sites. MX, when allowed to bind to these sites,
stabilizes the cytotoxic abasic site intermediate and effectively blocks recognition of
the lesion by APE1. Glycosylases also do not recognize this lesion, and replication is

7
effectively stalled, causing cell death via the eventual generation of DNA double-
strand breaks
88
.
This study provides proof-of-concept that blocking the catalytic activity of APE1
sensitizes cells to a chemotherapeutic agent, with a synergistic cytotoxic effect being
observed. MX is thus considered an inhibitor of APE1, although not a classical one,
in that MX inhibits APE1 by binding to its substrate and not the enzyme’s active site;
thus preventing recognition, and inhibiting APE1 functionality.

1.6 APE1/ Ref-1 Catalytic Activities
APE1 is a multi-functional enzyme that is an essential player in the BER pathway.
Following excision of the damaged base, APE1 hydrolytically cleaves the
phosphodiester backbone 5’ to the AP site via a divalent metal cation-facilitated
Figure 2. The chemical forms of the natural abasic site are in an equilibrium mixture of
racemeric hemiacetals, an aldehyde form and a hydrated aldehyde. In vitro experiments
indicate that the hemiacetal forms predominate, with the aldehyde form present at about 1%.
Tetrahydrofuran is a stable synthetic analog of an abasic site, and is used in our catalytic assay
as presented.

8
acid-base SN2 catalytic mechanism. This reaction leaves a 3’-hydroxyl and a 5’
abasic deoxyribose phosphate to be processed by the subsequent cascade of BER
enzymes, including DNA polymerase beta and DNA ligase III (Figure 3)
34, 102
.
APE1 activity is requisite for rectification of DNA damage in both the short-patch
and long-patch sub-pathways of BER, although each pathway utilizes different
enzymes to complete repair subsequent to APE1 activity. In addition to its primary
abasic site incision function, APE1 also exhibits 3’ →5’ exonuclease, 3’-
phosphodiesterase, and RNAse H catalysis, and a 3’-phosphatase activity
137, 161
. It
has also been shown that APE1 appears to have endonucleolytic activity as a repair
enzyme within the nucleotide incision repair pathway
94
. The same catalytic active
site—found in the conserved domain of most members of the ExoIII family—
appears to be employed for all of these repair actions
56
. The active site region also
shares some similarity with Mg
2+
-dependent endonucleases, phosphatases and
proteins involved in cell cycle signaling and transduction
37
. In addition to its
repertoire of repair activity, and equally as essential in function, APE1 utilizes a site
located in its N-terminus for redox regulation of important transcription factors such
as NF- ĸB, p53, c-Fos and c-Jun
165, 166
. This redox function of the enzyme was
discovered simultaneously to its endonuclease function, hence an alternate name of
Ref-1
2
. Our investigations presented herein focus on the inhibition of APE1
endonuclease activity.

9

1.7 Functional and structural aspects of APE1
The APE1 protein has a molecular mass of 35.5 kDa excluding post-translational
modifications, and three
consensus nuclear localization
signals in its N-terminus.
Proteolysis studies indicate that
the human protein consists of
globular nuclease domain, and a
flexible, disordered N-terminal
region
56, 121
. The core nuclease
domain shares structural
similarity not only to the ExoIII family of endonucleases, but also to the non-specific
Figure 3. APE1 catalyzes the hydrolytic, Mg-dependent generation of a single-strand break
with a 3’-hydroxyl terminus and a 5’-deoxyribose phosphate. Removal of the 5’-deoxyribose
moiety is catalyzed by polymerase ß (SP) or FEN1 (LP). DNA ligase III or XRCC1 are then
involved in ligating the filled gap.
Figure 4. Complex of APE1 active site and abasic DNA
fragment with important residues for catalytic activity and
substrate recognition.

10
endonuclease, DNAse I. The α/ß sandwich motif of these Mg++-dependent
endonucleases is also present in phosphatases and other cell-cycle signal-
transduction proteins
37
.
APE1 recognizes abasic sites by probing DNA for increased flexibility, a quality
bestowed on the phosphate backbone specific to abasic sites or single nucleotide
gaps. Crystal structures of APE1 complexed with abasic DNA provide evidence that
APE1 is a structure-specific nuclease that locates and productively binds DNA that
can adopt a unique, kinked formation (Figure 4,
174
)
103
.
The two catalytic activities of the APE1/ Ref-1 enzyme are reflected by structural
domain organization as overlapping but distinct entities. The redox active site is
consensus nuclear location signals. Site-directed mutagenesis has indicated that
cysteine residues at positions 65 and 93 are required for redox regulation of
transcription factors
166
. The globular nuclease domain, which is responsible for the
endonuclease functionality of this protein, spans residue 96 to approx. residue 309.
Using site-directed mutagenesis, chemical footprinting techniques and molecular
dynamics simulations several residues have been indicated that give insight into how
APE1 interacts with its metal cation and abasic DNA. APE1 binds predominantly to
the minor groove of abasic DNA using residues R156 and Y128 to contribute to
protein-DNA stability. The APE1 metal-binding site is in close proximity to the
region immediately 5’ to the abasic site. Amino acid residues E96, D70, and D308 of
APE1 are involved in metal co-ordination, with D70 and E96 binding directly to the
metal and D308 co-ordinating the metal. Figure 5 shows the domain organization of

11
APE1/ Ref1 and highlights residues important to both catalytic activities of this
enzyme.

1.8 Basis for targeting Ape1 for cancer therapeutics
DNA anti-sense technology has implicated APE1 in cellular resistance to a variety of
chemotherapeutic agents that cause alkylation and oxidative damage
45
. In turn,
proof-of-concept studies have shown that mammalian cells are rendered sensitive to
methyl methane sulfonate, hydrogen peroxide, bleomycin, temozolomide and
gemcitabine when targeted with anti-sense oligonucleotides specific to APE1
15, 45,
151
. One study used the Saccharomyces cerevisiae abasic endonuclease, Apn1 (which
does not possess redox activity), to restore abasic repair activity in several human
cell lines that had APE1 downregulated by specific siRNA. In the absence of
exogenous DNA damage, Apn1 corrected the decrease in proliferation, increase in
abasic sites and increased levels of apoptosis that were associated with siRNA-
mediated downregulation of APE1
54
. Other studies have demonstrated that knocking
out APE1 or using a dominant-negative form of the protein leads to sensitization to
chemotherapeutic agents
23, 100, 125, 168
. Based on these studies, it is evident that APE1
Figure 5. Functional domain organization of APE1/ Ref-1 and residues important for
each catalytic activity.

12
is fundamental to maintenance of genomic integrity within the cell. Even in the
absence of exogenous DNA damage in the form of chemotherapeutic agent toxicity,
a decrease in APE1 results in poor cell proliferation, an increase in mutagenic and
cytotoxic abasic sites, and increased levels of apoptosis. In a tumor cell population
that is characterized by high cellular proliferative levels, and a high rate of
mutagenesis, it is indicated that APE1 would be a reasonable target for small-
molecule inhibition. The addition of clinically relevant DNA damaging agents which
generate damage that is repaired via BER would be an applicable clinical strategy.
The chemopreventive nature of APE1. An observation that has been made as a
result of certain studies indicates that APE1 may be a productive target of
chemoprevention, wherein the modification of APE1 activity changes the response
of cells and organisms to DNA damaging agents
134
. The extensive physical
interactions of proteins within the BER pathway (see Section 1.10) are potential
targets of chemical control, as is acetylation. Acetylation of APE1 is a requirement
for many protein-protein interactions
14
. Two stimulants of the BER gene expression
reduce the risk of cancer: selenium and oltipraz. Selenomethionine protects cells
from damage by hydrogen peroxide, and a dominant-negative form of APE1 blocks
activation of p53 by selenomethionine, suggesting that APE1 is involved in the
stimulation of BER by selenoproteins
130, 131
. Oltipraz (4-methyl-5-(pyrazinyl)-3H-
1,2-dithiole-3-thione) is an agent that is being investigated for its chemopreventive
aspects. It has been associated with a reduced risk of lung, liver, bladder and breast
cancers and is in Phase I and II clinical trials
27, 126
. Unlike with selenoproteins, it is

13
thought that the redox, rather than the repair function of APE1, is most significant in
this regard
171
.
In addition to demonstrable potentiation of the effects of cytotoxic agents, the value
of concurrent APE1 inhibition with existing anticancer clinical regimens is
underscored by clinical evidence that establishes a relationship between APE1
expression levels and cancer. Alterations in APE1 expression and localization are
thought to have prognostic and/ or predictive significance in several cancers,
including colon, breast and lung
70, 71, 119
. Differential patterns of expression between
normal and cancerous tissue have been observed in cervical, prostate, and epithelial
ovarian cancers, where expression levels are elevated
45, 69, 104, 167
. A predictive
significance in cancer-specific survival following radiotherapy has also been
attributed to APE1 expression
63, 77
.

1.9 Interactions of APE1 with other proteins in BER
The physical protein interactions of APE1 have been shown to increase the
productivity of individual enzymes as well as the BER pathway as a whole. APE1
stimulates the activity of both downstream and upstream enzymes within the repair
cascade. For example, APE1 interacting with FEN1 may play a role in the selection
of which subpathway (short-patch or long-patch) the DNA lesion enters. Both APE1
and DNA polymerase ß are involved in the recruitment of proliferating cell nuclear
antigen, PCNA, which is a DNA-scaffolding protein that contributes to the stability
of the lesion when repaired. In addition to these proteins, APE1 is stimulated by

14
XRCC1 and heat shock protein 70
73, 147
. APE1 plays a role in the stimulation of
several glycosylases, enzymes that are upstream of it in the BER pathway. Among
these are OGG1, whose dissociation from it abasic substrate is prevented by APE1;
UDG, whose binding to abasic DNA is facilitated by APE1, at least temporarily
protecting the cell from toxicity by steric hindrance; TDG, which is displaced by
APE1 from the abasic substrate through a physical interaction; and adenine DNA
glycosylase, whose complex-formation with its substrate is stimulated by APE1
independent of its repair catalytic activity
54, 64, 94, 112, 113, 154, 170
. One of the most
important interactions of APE1 is with p53. It is both stimulated by, and stimulates
the BER pathway. APE1 has been shown to promote the tetramerization of p53 and
promotes the association of p53 dimers into tetramers, which leads to the
enhancement of p53 DNA binding activity
140
.

1.10 Current progress in inhibitor discovery to APE1
As mentioned previously, indirect inhibition of BER has been accomplished by the
use of a alkoxyamine derivative that binds to abasic sites by reacting with the
aldehyde group in the acyclic sugar left in the DNA abasic site, thus preventing
recognition of the abasic site substrate by APE1. The clinical significance of this
activity has been illustrated by its utility with temozolomide, a common clinical
alkylating agent used in the treatment of gliomas – although this effect has been
demonstrated in a variety of tumor cell lines, including colon, ovarian and breast.

15
Small-molecule inhibitors of APE1. The first small-molecule inhibitor to APE1
repair activity was published in 2005
91
. The compound was discovered using a
novel FRET-based high-throughput screen with a library of only 5000 drug-like
compounds. 7-nitroindole-2-carboxylic acid inhibits all the repair activities of APE1,
including its endonuclease, 3’-phosphodiesterase, 3’-phosphatase, and 3’-to 5’-
exonclease activities of APE1 at low micromolar concentrations. It has also been
shown to potentiate the toxicity of methyl methane sulfonate, temozolomide,
hydrogen peroxide and Zeocin in HT1080 fibrosarcoma cells. Our group has tried to
replicate this low micromolar inhibition of APE1 in vitro but have been
unsuccessful.
A second study uncovered antimony-based small molecule inhibitors to APE1 via a
screening of the NCI Diversity Set of small molecules. A further screen of a small,
specific arylstibonic sublibrary identified ligands with IC
50
values between 4 and
300 µM
129
. The negatively charged stibonic acids act by a partial mixed-mode of
inhibition and possibly act as DNA phosphate mimics. In vivo, the most active
arylstibonic acid compound was highly cytotoxic.
Lucanthone. Lucanthone is a small-molecule inhibitor of topoisomerase II used in
the radiation and chemotherapy sensitization of brain and breast tumors
10
. Treatment
with lucanthone in breast cancer cells causes a dose-dependent increase in abasic
sites, presumably due to the inhibition of the repair activity of APE1. The compound
also increases the cell killing effect of methyl methane sulfonate and temozolomide
90
. Due to the high levels of lucanthone necessary for in vitro inhibition of APE1, it

16
is now concluded that its effect is more as a topoisomerase II inhibitor and off-target
events rather than as a specific APE1 inhibitor. Clinically, however, lucanthone is
relatively non-toxic, and has been used in treating schistosomiasis, and in radiation
and chemotherapy sensitization.
Despite the discovery of inhibitors containing acidic moieties that inhibit APE1 with
very low micromolar concentrations, it is necessary to find inhibitors that are of
clinical significance. In this regard, cellular inhibitors of APE1 are likely
characterized by their synergistic ability to potentiate the DNA damaging effect of
clinical chemotherapeutic agents. In the absence of DNA damage, it is unlikely that a
small-molecule targeted to APE1 alone can cause rampant abasic site damage
leading to apoptosis. Rather, cell cycle arrest should occur, and the abasic sites
converted to single-strand breaks that may be cycled into other repair pathways such
as the nucleotide excision repair pathway, NER. NER is responsible for the repair of
such lesions as those caused by the platinum agents, although preliminary work
performed by us indicates for the first time that under certain circumstances, BER
may be involved in the repair of oxaliplatin- and cisplatin-mediated DNA damage.
Moreover, these observations seem also to be affected by p53 status of the cell line.
The clinical success of small-molecule inhibitors directed against other DNA
repair proteins also provides a basis for development of clinically responsive APE1
inhibitors. The selective targeting of repair proteins is a novel therapeutic strategy
given the information that exists regarding the potentiation of DNA damage in cell
lines derived from patients with inherited DNA repair disorders. In addition, various

17
genetic approaches have been taken to study the effects of disabling DNA repair in
mammalian systems, including RNA interference and mouse knockout models – for
APE1, double-knockout mouse models are inviable, underscoring the importance of
this protein even in embryonic development. Table 1 summarizes small-molecule
inhibitors currently in use in clinical settings.
Table 1. Small-molecule inhibitors of DNA repair
Target Inhibitor
O6-methylguanine-DNA methyltransferase
(MGMT)
O6-Benzyl-guanine, O6-4 Bromthenyl-guanine
(Patrin 2), O6-4 Bromthenyl-guanine-C8-B-D-
glucose
APE1/ Ref-1
Methoxyamine (APE1 activity), E3330 (Ref-1
activity)
Poly-ADP-ribose-polymerase (PARP1)
NU 1025, NU 1064, NU 1085, AG 14361, AG
14699, GPI 15427, CEP 6800
DNA dependent protein kinase (DNA-PK)
Wortmannin, OK-1035, LY294002, NU 7026, NU
7441, IC 87102, IC 87361
DNA methyltransferase Decitabine
ATM Kinase KU 0055933

18
1.11 Computational approaches in development of small-molecule inhibitors
Rational design of small molecule APE1 inhibitors is enabled by the availability of
structural information on APE1 in complex with the substrate DNA bearing an
abasic site. A pharmacophore is a three-dimensional arrangement of features of an
active site or a model
inhibitory molecule that is
used to cull small-
molecule databases for
lead compounds that will
exhibit biologically active
properties.

The pharmacophore perception of identifying small molecule inhibitors with diverse
chemical scaffolds is a popular technique in drug design, and has been successfully
used to discover numerous classes of clinically relevant small-molecule inhibitors
targeting a number of pathways, including DNA repair
57, 60, 162
. This technique also
allows the identification of different classes of molecules having varying structural
scaffolds but all potentially exhibiting optimal biological activity. The advantage of
identifying molecules from different classes is that each molecule will have different
pharmacodynamic/ pharmacokinetic properties such as absorption, distribution,
metabolism, excretion and toxicity, etc.
Figure 6. The pharmacophore approach to drug development
utilizes well-defined three dimensional set of structural features
based on a chosen catalytic model that are considered optimal
to identifying active lead inhibitory candidate molecules .

19
In our primary investigation, which is the first report of rationally designed APE1
inhibitors, we developed a set of three-dimensional (3D) pharmacophore models
based on APE1 interactions with the abasic deoxyribose 3’- and 5’- phosphate
backbone in a co-crystal structure of APE1 in complex with substrate abasic DNA
103
. The pharmacophore models represent prominent interactions, and the chemical
nature and shape of the abasic DNA fragment within the APE1 active site. Further
optimization of our small molecules included the use of docking programs to impart
selectivity to those that showed a high degree of inhibition to the human protein.
The pharmacophore technique also allowed us to identify a novel set of
compounds that have not been previously shown to inhibit APE1. This class of
molecules is bioisosteres of carboxylic acids. These bioisosteric molecules are much
more conducive to cellular entry and permeation of the hydrophobic cell membrane
than are carboxylic acids and thus represent a good platform for the optimization and
development of clinically significant inhibitors to this essential repair enzyme.
This dissertation, in addition to the discovery of small-molecule inhibitors as
described above, also presents preliminary work on the biochemical aspects of APE1
small-molecule inhibition. We aim to build biochemical models of APE1 gene
expression manipulation: overexpression via a tetracycline-regulated mammalian
protein expression vector, and downregulation using specific APE1-targeted siRNA
in an attempt to select a cell line suitable for use in the development of cell-based
inhibitors to APE1. In addition, to establish our own proof-of-concept studies within
a cellular environment, we looked at the inhibition of abasic site repair by

20
methoxyamine in conjunction with several current standard-of-care colorectal cancer
chemotherapeutic agents, including 5-FU. Finally, as further validation of APE1 as a
rational target for the development of anticancer therapeutics, we sought to test the
significance of single-nucleotide polymorphisms in response to chemotherapy and
survival rate in two populations of colorectal cancer patients.
In Chapter 2 of this thesis, we demonstrate optimization of the high-
throughput screen and purification of the recombinant enzyme, and optimization of
experimental conditions for inhibitor screening. Protein kinetics observations we
have made with regard to the catalytic activity of the enzyme are also presented, as
these are important considerations to be made when screening for inhibitors to this
enzyme. Chapter 3 represents the pharmacophore-guided discovery of small-
molecule inhibitors to APE1, a study that has since been published in the Journal of
Medicinal Chemistry, 2008. This leads us into Chapter 4, which presents work on the
optimization of bioisosteres of carboxylic acids to behave as cell-penetrable small-
molecule inhibitors of APE1. Chapter 5 indicates preliminary biochemistry work
performed with regard to the biochemical models of APE1 inhibition, and the
clinical research that we have conducted on single-nucleotide polymorphisms of
APE1 in a population of colorectal cancer patients.
Chapter 6 presents research on the peptide inhibition of the human
immunodeficiency virus 1 integrase enzyme. HIV-1 IN is also a rational target for
drug development, and peptides have been presented here as structural tools for the
elucidation of protein-interaction hotspots within the protein that present themselves

21
as attractive structural targets for the development of second-generation inhibitors to
the protein. Figure 7 illustrates the progression of the peptide research that has been
published, as well as the work that is currently in progress for publication.

Figure 7. The optimization of peptides derived from the HIV-1 integrase protein for eventual use as
cell-permeable ‘nano-needles’ that target catalytic hot-spots on the protein, thus inhibiting essential
protein-protein and protein-nucleic acid interactions.

22
Chapter 2: Design and Optimization of a Catalytic Assay for APE1

2.1 Tetrahydrofuran as a synthetic analog of an abasic site

The natural abasic ribose exists as an equilibrium mixture between four chemical
forms – 99% as two hemi-acetal enantiomers ( α- and ß-2-deoxy-D-ribofuranose) and
approximately 1% as ring-opened aldehyde forms. The ring opened forms are subject
to facile beta-elimination reactions that result in DNA strand scission, and general
instability
155
. Consequently, for most biochemical analyses requiring an abasic site
incorporated oligonucleotide, a prototypical synthetic abasic site isosteric with the
predominant cyclic form of the 2-deoxyribose is used: 3-hydroxy-2-hydroxymethyl-
tetrahydrofuran (THF). The stereochemistry of the deoxyribose ring is preserved,
with the electronic properties differing only at the C1 position of the sugar due to the
hydroxyl moiety present at C1. THF is a stable, non-racemic analog, and structural
studies involving the natural abasic ribose usually yield similar results to those using
the analog
159
. Figure 8 shows the side-by-side structures of the natural abasic site
(O) and the tetrahydrofuran (F).
Figure 8. Structure of natural abasic site compared to the synthetic abasic site analog.

23
Our initial attempts at optimizing an oligonucleotide for utilization in a catalytic
activity assay for APE1 were targeted towards using an oligonucleotide containing a
uracil residue. The reaction of this oligonucleotide with UDG removes the base and
leaves a natural abasic site, which would later be cleaved with APE1 in the catalytic
assay. However, some non-specific cleavage was observed, leading to the eventual
design of a 26-mer double-stranded oligonucleotide containing a tetrahydrofuran
residue at position 15 from the P-32 labeled 5’-end. This resulted in a specific
cleavage reaction clearly illustrating the 25-mer radioactively labeled synthetic
oligonucleotide and its 15-mer cleavage product. Figure 9 shows a representative

denaturing polyacrylamide gel of abasic site incision by APE1, and inhibition by a
lead compound.
Figure 9. Visualization of a typical experiment after optimization.

24
Although known to catalyze incision on a variety of substrates including abasic sites
positioned at replication forks, bubble- and loop-configurations, hybrid DNA/RNA
intermediates, and RNA substrates, APE1 has a distinct substrate preference for
double-stranded (ds) DNA
13
. One analysis of APE1 activity against abasic sites in
dsDNA yielded a Km of ~24nM, and a Km of ~428nM for abasic sites centrally
located in single-stranded DNA. The k
cat
for both these substrates was around 4 s
-1
,
indicating that the affinity of APE1 for abasic sites in dsDNA is significantly higher
than for those in ssDNA, The turnover rate for both these substrates, however,
remains the same
94
. To ensure maximally optimum conditions for APE1 binding to
its DNA substrate, we designed the oligonucleotide to be double-stranded, and
located the THF residue centrally on the top strand.
2.2 Optimization of Escherichia coli expression system and protein purification
protocol for APE1.
The expression system used to purify recombinant protein for the purpose of our
assays was the pQE30 expression
system (Qiagen, CA). The plasmid
contained the gene for APE1 cloned
between the BamH1—HindIII sites of
the multiple cloning site. The coding
sequence for APE1 is derived from
deposited GenBank sequence number
Figure 10. Plasmid map for pQE30 expression
vector. Qiagen, Inc. (Valencia, CA).

25
S43127 (CDS 224—180). Flanking the stop codon of the cloned gene sequence is a
6x His-tag for affinity purification of the protein. Figure 10 shows a map of the
pQE30 plasmid, a kind gift from the Curran laboratory. The pQE30 expression
system is based on the T5 promoter transcription-translation system. The T5 phage
promoter is recognized by the E. coli RNA polymerase and two lac operator
sequences that increase lac repressor binding and ensure repression of the T5
promoter. The T5 promoter initiates high transcription rates that are regulated and
repressed by the presence of high levels of the lac repressor protein, present in trans
to the APE1 gene. A ß-lactamase gene confers ampicillin resistance to the cell. The
host strain, E. coli M15, contains a low-copy regulator plasmid [pRep4] which
confers kanamycin resistance and constitutively expresses the lac repressor protein
encoded by the lacI gene which binds to operator sequences to tightly regulate
recombinant APE1 expression. Upon addition of isopropyl-B-D-galactosidase
(IPTG), which binds to the lac repressor protein to inactivate it, the host cell RNA
polymerase is able to transcribe sequences downstream of the promoter. The E. coli
M15 [pRep4] was purchased as a non-competent strain that was made chemically
competent in the lab using a low-temperature, rubidium chloride protocol. Aliquots
of competent host strain were used for fresh transformations of pQE30-APE1 each
time a batch of protein was purified.

26
Transformant
selection. To ensure
maximum yield of
protein from a single
batch (1 liter) culture, we
first undertook to select a
transformant colony that
showed a high level of
protein expression post-
IPTG addition. To this
end, after transformation, several single colonies were each inoculated into separate
tubes containing 10ml of LB broth. Samples of these cultures were taken before and
after addition of IPTG, and electrophoresed at 100V onto an 8% denaturing
polyacrylamide gel. Because the APE1 protein has a molecular size of 35.5 kDa, we
used an aliquot of purified HIV-1 integrase (32 kDa) as a marker to approximate size
of the required observation. Figure 11 shows the selection of transformants as
described. It is clearly observed that upon addition of IPTG to the mini-culture, there
is an increase in expression of a protein of approximately 36 kDa. Glycerol stocks
were made of Clone 1, and used in subsequent purifications.
Protein purification was performed using a protocol that was slightly modified from
the Qiagen protocol. The most significant modifications included lysis in a French
press as opposed to sonication to maximize cell lysis, and inclusion of an imidazole
Figure 11. Transformant selection for maximum protein yield.

27
elution gradient. Elution gradients allow for protein to be more easily concentrated
when subsequently dialyzing protein
fractions. As seen in Figure 12, the
imidazole gradient yielded the most
protein at lower concentrations of
imidazole, and allowed for the dialysis of
the more concentrated fractions together in
order to obtain concentrated protein.

2.3 Optimization of experimental conditions for inhibitor screening.
To ensure optimal catalytic activity of APE1, we also optimized the experimental
conditions of the assay. Our first activity assay was to determine the volume of
protein to be used in a cocktail for a single experiment. We used various volumes of
undiluted and serially diluted protein in the presence of two different divalent
cations, manganese and magnesium, for 30 minutes at 37ºC as shown in Figure 13.
Even at a ten-fold dilution, APE1 actvity has progressed beyond its abasic site
incision activity; at high concentrations of APE1, once all of the substrate has been
cleaved at the abasic site, the enzyme begins to remove nucleotides from the 3’-end
of the cleaved product in accordance with its 3’-5’ exonuclease activity.
In a clear demonstration of the robustness of this enzyme, Figure 14 shows that
dilutions of up to 1/20000 still result in complete cleavage of the DNA substrate. The
Figure 12. Protein elution using imidazole
gradient.

28
concentration of the DNA substrate was maintained at 200nM as described in the
literature. Incubating the protein at 85ºC for 5 minutes, and subsequent incubation
with the DNA substrate at serial dilutions, slightly reduced abasic cleavage, with an
abolishment of activity at a 10000-fold dilution.

Figure 13. Determination of optimal concentration of protein to be used in assay.
Figure 14. APE1 is a robust enzyme. At a serial dilution of 1/20000, APE1 still exhibits complete
cleavage of the DNA substrate. Incubation of this enzyme at high temperatures and subsequent
incubation with the DNA substrate at 37ºC causes a reduction in cleavage activity and complete
abolishment of activity at a dilution of 1:10000.

29
It was observed that at serial dilutions below 1:20000, there was no significant
decrease in APE1 cleavage activity. From these optimization experiments, it was
determined that a final concentration of 0.5 nM enzyme would be appropriate for use
in the screening assay.
To explore an optimum incubation time for 200nM DNA abasic substrate with the
selected concentration of enzyme, we quenched reactions at varying intervals from
30s to 20 minutes. At a serial dilution of 1/20000, a 20-minute incubation resulted in
complete conversion of DNA substrate to cleaved product. The assay now uses a 15-
minute incubation at 37ºC for optimal results. We have also found that changes in
temperature between 25ºC and 37ºC do not show any significant changes in
catalysis.

Figure 15. The effect of incubation time on APE1 catalysis. A 20-minute incubation at 37ºC was
selected for use in the screening assay. To eliminate the possibility that DNA cleavage was not the
result of APE1 catalytic activity, the DNA substrate was incubated with different buffers. It is also
observed that the IC
50
of 7-nitro-indole-2-carboxylic acid, a published inhibitor to APE1 changes
with a change in incubation time (lanes 14-16, 5-minute incubation; 17-19, 15 minute incubation;
20-22, 10 minute incubation. Drug concentrations were 33, 11, and 3.7 µM at each incubation.

30
To eliminate the possibility that DNA cleavage was not the result of APE1 catalytic
activity, the DNA substrate was incubated with B cocktail (100mM magnesium
chloride, 1 mM BSA, 2% B-mercaptoethanol), HEPES buffer (B cocktail containing
HEPES detergent), Tris-NaCl running buffer, HIV-1 integrase (an enzyme that will
not recognize abasic sites) and APE1 buffer (dialysis buffer without protein) + B
cocktail. As shown in Figure 15, the DNA substrate containing the synthetic THF
residue is unaffected by any of the components of the buffers that are used in our
assay. It is also not recognized by the HIV-1 integrase enzyme, which requires a
consensus recognition sequence to bind an oligonucleotide.
2.4 Behavior of 7-nitro-indole-2-carboxylic acid. As a further optimization, we
also wanted to observe the behavior of the enzyme in the presence of
dimethylsulfoxide
and inhibitors. To
this end, we
incubated different
serial dilutions of the
enzyme with the first
published small
molecule inhibitor
APE1. While we could not duplicate the published IC
50
value of this molecule, it
was demonstrated that a change in concentration of the enzyme in the assay had a
significant effect on the IC
50
of the inhibitory compound. Figure 16 presents this
Figure 16. The IC
50
value of 7-nitro-indole-2-carboxylic acid
changes with enzyme concentration. An inversely proportional
relation between concentration and IC
50
value is observed.

31
observation. Interestingly, from Figure 15, it is observed that incubation time is also
inversely proportional to observed IC
50
values for the inhibitor.
With regard to the enzyme, selection of a suitable buffer in which the reaction was to
occur was also necessary. APE1 requires the presence of a divalent metal cation for
catalysis, and as can be observed in Figure 13, there is a very slight difference in
activity between the use of manganese chloride and the use of magnesium chloride in
reaction conditions: the activity of APE1 is slightly higher with the latter, with the
exonuclease activity completely dismantling the cleaved product nucleotide by
nucleotide. This is an acceptable observation, as the use of magnesium chloride in
the assay most closely mimics the native cellular environment in which APE1
normally exists. The ‘B cocktail’ containing the enzyme thus incorporates 100mM
magnesium chloride. It has been reported that differing concentrations of the
magnesium divalent cation alters APE1 nuclease efficiency by affecting the kinetics
and stability of phosphodiester bond cleavage
160
. Higher concentrations of
magnesium cations favor the double-stranded incision activity of APE1, and thus we
use the maximum reported concentration of magnesium at which APE1 activity is
not curtailed.
The enzyme cocktail also contains 1 mg/ml bovine serum albumin (BSA) The
presence of this protein prevents non-specific binding to the DNA substrate. More
importantly, the use of BSA in inhibitor screening assays is useful to prevent false
positive identifications due to promiscuous molecules
99
. Promiscuous molecules
tend to inhibit proteins universally via aggregation structures that bind to the active

32
sites of proteins and prevent catalysis. The presence of BSA prevents aggregation of
such molecules and minimizes the identification of possible false positive leads in a
high-throughput screening environment. Finally, the presence of a reducing agent, 2-
mercaptoethanol, maintains protein structure within the buffer to ensure stable
catalysis.
2.5 Conclusions and implications. Optimization of a catalytic assay for evaluating
inhibitors to APE1 resulted in the observation that the activity of this enzyme is
extremely robust, catalyzing the complete abasic-site cleavage of DNA substrate
even at nanomolar concentrations. It must be noted that the enzyme is also very
sensitive to changes in enzyme concentration and incubation time. Presence of an
inhibitor (with dimethylsulfoxide as solvent) also reduces the basal activity of APE1,
but in turn, changes in enzyme concentration and incubation time inversely affect the
observed IC
50
values of a prototypical inhibitor. To preserve integrity of the assay
during the course of an inhibitor study, it is imperative that the same conditions be
maintained throughout, in order to obtain consistent half-maximal inhibitory
concentrations and avoid large standard errors. The established standard in our
laboratory as a result of these optimization experiments is that the enzyme
concentration is maintained at which there is 100% cleavage of the 26-mer abasic
DNA substrate without subsequent exonuclease activity. Incubation time selection
will follow the same standard. It is difficult to ascertain at which exact concentration
this happens as variations in protein yield and subsequent Km of the purified protein
tend to change with each batch purified. However, if the established conditions are

33
maintained, it will be possible to use the compounds presented in the following
chapter as positive controls for further development of inhibitors to APE1 of
different classes of molecules.
Figure 17 summarizes the scheme of our catalytic assay for the development of
inhibitors to APE1.

Figure 17. Scheme of APE1 catalytic assay. A P
32
-labeled double-stranded oligonucleotide
containing a synthetic abasic site is incubated with recombinant protein. Upon cleavage of the top
strand by APE1, dissociation of the top and bottom strands occurs. The labeled substrate and 15-
mer cleavage product are visualized on an SDS-denaturing polyacrylamide gel.

34
Chapter 3: Discovery of Small-Molecule Inhibitors to APE1
Rational design of small-molecule APE1 inhibitors is enabled by the availability of
structural information on APE1 in complex with the substrate DNA bearing an
abasic site. The pharmacophore perception of identifying small molecule inhibitors
with diverse chemical scaffolds is a popular technique in drug design, and has been
successfully used to discover numerous classes of clinically relevant small-molecule
inhibitors targeting a number of pathways, including DNA repair
57, 60, 162
. We have
developed a set of three-dimensional (3D) pharmacophore models based on APE1
interactions with the abasic deoxyribose 3’- and 5’- phosphate backbone in a co-
crystal structure of APE1 in complex with substrate abasic DNA
103
. The
pharmacophore models represent prominent interactions, and the chemical nature
and shape of the abasic DNA fragment within the APE1 active site. This is the first
report of rationally designed selective APE1 inhibitors.
In an effort to characterize the selectivity and mode of inhibition of these compounds
against APE1, we further evaluated them against the catalytic activity of
Exonuclease III (ExoIII), Endonuclease IV (EndoIV), and HIV-1 integrase. While
both ExoIII and EndoIV are enzymes that possess abasic site cleavage activity, HIV-
1 integrase is a retroviral enzyme that binds to the viral and host DNA and, similar to
APE1, cleaves the phosphodiester backbone 5’ to its recognition site of the viral
DNA, leaving a recessed 3’hydroxyl moiety. As with APE1 and ExoIII, and
characteristic of most DNA-binding enzymes, HIV-1 integrase also requires a
divalent metal cation for catalysis. Importantly, this structurally diverse set of

35
molecules we have identified are selective APE1 inhibitors and are suitable as lead
molecules to establish quantitative structure-activity relationship models for further
development of clinically relevant APE1 inhibitors.
3.1 Design and Generation of Pharmacophore Models. APE1 uses a well-defined
positively charged surface to selectively recognize the flipped-out abasic DNA
fragment, which binds within a unique binding pocket on the bottom of the DNA
binding region of APE1
103
.
Charge interactions, several
strong H-bonding interactions,
and shape complementarity
are observed between
prominent amino acid residues
of APE1 and abasic fragment
of the abasic DNA (Figure
18). An unusual interaction
between the APE1 amino acid
residue Arg177, which inserts
through the major groove of
abasic DNA and the negatively charged 3’-phosphate of the abasic fragment indicate
unique functionality for this enzyme. A hydrophobic pocket surrounded by amino
acid residues Phe253, Trp280 and Ile282, selectively recognizes and binds to the
abasic deoxyribose sugar moiety and prohibits binding of DNA bases and racemized
R17
N17
N21
H30 Mn
2
3
5
Figure 18. APE1 interactions with abasic site fragment in
the co-crystal structures of human APE1 bound AP DNA
(PDB:1DEW, and PDB:1DE9). The abasic site fragment is
rendered as a stick model. The surface model of the abasic
site fragment colored according to electrostatic potentials
is also shown. APE1 active site amino acid residues are
shown as ball & stick models. The green lines indicate
interactions and magenta sphere represents the active site
Mn2+. Some of the interactions are considered in three-
dimensional pharmacophore modeling (177).

36
β-anomer abasic sites. The negatively charged 5’-phosphate is involved in a number
of strong charge and H-bonding interactions with APE1 amino acid residues Asn174,
Asn212, His309 and the Mn
2+
metal ion (Figure 18).
Given the importance and uniqueness of these interactions for APE1 in recognizing
and incising abasic DNA sites, we transformed the prominent ones into a set of 3D
pharmacophore models using the abasic fragment as a template. These
pharmacophore models represent the chemical and electrostatic environment of
specific APE1 interactions with the abasic DNA fragment (Figures 19 – 20). The co-
crystal structures used to generate the pharmacophore models contains a synthetic
analog (tetrahydrofuran) of the abasic deoxyribose. The natural hydrolytic abasic
deoxyribose site harbors a hydroxyl group. Interactions of the protein with the abasic
fragment are represented by a set of pharmacophore components: hydrophobic (H),
H-bond acceptor (A) and negatively ionizable (NI) features. The pharmacophore
models were then used to identify novel molecules bearing desired chemical moieties
that mimic abasic DNA to bind APE1 and disrupt its catalytic functions.
Four plausible pharmacophore models were generated using a combination of H, A,
and NI features (Figures 19 – 20). The pharmacophore H1NI2 is comprised of two
negatively ionizable (NI1 – NI2) and a hydrophobic (H1) feature (Figure 19A). The
NI1 and NI2 features are separated by 8.07 ± 1 Å. The inter-feature distances
between H1 to NI1 and NI2 are 4.13 ± 1 and 5.81 ± 1 Å, respectively. Mapping the
H1NI2 pharmacophore onto the abasic fragment demonstrates that the NI1 and NI2
features represent the negatively charged 3’- and 5’-phosphates and the H1 feature

37
corresponds to the tetrahydrofuran that mimics the abasic deoxyribose (Figure 19B).
Pharmacophore A1NI2 consists of one A feature and two NI (NI1 and NI2) features.
Similar to pharmacophore H1NI2, the NI1 and NI2 features are separated by a
distance of 8.07 ± 1 Å. The mapping of A1NI2 onto the abasic fragment shows a
feature mapping pattern similar to that of pharmacophore H1NI2 with respect to the
NI1 and NI2 features. The H-bond acceptor feature is mapped onto the O atom of the
tetrahydrofuran (Figures 19C – 19D). Inter-feature distances between A1 to NI1 and
NI2 are 5.08 ± 1 and 4.70 ± 1 Å, respectively. The pharmacophore A3NI1 is
comprised of three H-bond acceptor (A1 – A3) features and a negatively ionizable
feature (NI1). The NI1 feature represents the negatively charged 3’-phosphate of the
abasic fragment. One of the three H-bonding acceptor features (A1) represents the
interaction of the tetrahydrofuran oxygen atom with the side chain amine of Asn174.
H-bonding acceptor features A2 – A3 represent interactions of APE1 amino acid
residues Asn174, His309 with the O5’, and O1P oxygen atoms of the 5’- phosphate.
The 3D arrangement of features and mapping of pharmacophore A3NI1 onto the
abasic fragment is shown in Figures 19E – 19F. In parallel to the above-mentioned
three pharmacophore models, the model A3H1 was also generated, which is devoid
of a negatively ionizable feature.
The pharmacophore A3H1 consists of three H-bond acceptor features (A1 – A3) and
a hydrophobic feature (H1). The 3D arrangement and inter-feature distances of
A3H1 are shown in Figure 20A. Mapping the A3H1 pharmacophore onto the abasic
fragment correlates the APE1 – abasic DNA fragment interactions (Figure 20B).

38

The pharmacophore A3H1 consists of three H-bond acceptor features (A1 – A3) and
a hydrophobic feature (H1). The 3D arrangement and inter-feature distances of
A3H1 are shown in Figure 20A. Mapping the A3H1 pharmacophore onto the abasic
fragment correlates the APE1 – abasic DNA fragment interactions (Figure 20B). The
H-bond acceptor feature A1 represents the 3’- phosphate interaction with the side
chain guanidine of Arg177. The hydrophobic feature H1 demonstrates the chemical
Figure 19. Three-dimensional pharmacophore models H1NI2 (A), A1NI2 (C), and A3NI1 (E) are
generated to represent human APE1 interactions with abasic DNA in the co-crystal structure of the
APE1 bound to abasic DNA (1DEW.PDB). The abasic fragment (3’-phosphate, abasic
deoxyribose, and 5’-phosphate) was used as a template for pharmacophore modeling.
Pharmacophore features are: Hydrophobic (H, cyan), negatively ionizable (NI, blue), H-bond
acceptor (A, green). Inter-features distances are given in Å. (B, D, and F) The pharmacophore
models are mapped on to the abasic fragment (177).
4.13 ±
5.81 ±
8.07 ±
5.08 ±
4.70 ±
8.07 ±
5.07 ±
4.67 ±
6.39 ±
2.42 ±
A C
NI
NI
H
NI
NI
A
NI
A
A
A
B D

39
nature and environment of the tetrahydrofuran that mimics the natural abasic
deoxyribose. The H-bond acceptor features A2 – A3 correlate to the interactions of
residues Asn174, His309 with the O5’, and O1P oxygen atoms of the 5’- phosphate.
In addition to considering interaction features, the shape of the abasic fragment
backbone was also generated and merged with pharmacophore A3H1 to enhance its
selectivity (Figure 20C). The shape feature functions as an additional filter in the
database search process and restricts the shape and size of hits that can be retrieved
by pharmacophore A3H1.

3.2 Database Search and Compound Selection. A library of 365,000
commercially available small molecule compounds was collected and converted to a
searchable multi-conformer 3D database. The small molecule database was searched
for potential compounds using the 3D pharmacophore models H1NI2, A1NI2, A3NI,
Figure 20. (A) Three-dimensional pharmacophore model (A3H1) was based on the human APE1
interactions with abasic DNA in the co-crystal structure of human APE1 bound to abasic DNA
(1DEW.PDB). The abasic fragment was used as a template for pharmacophore modeling. (B) The
pharmacophore model was mapped on to the abasic fragment. (C) The shape of the abasic fragment
was generated and merged with the H1A3 pharmacophore. Pharmacophore features are:
Hydrophobic (H, cyan), negatively ionizable (NI, blue), H-bond acceptor (A, green). The light blue
mesh shows shape of the abasic fragment (177).

40
and A3H1 as search queries. A search using pharmacophore H1NI2 retrieved 424
compounds (0.0011 %). The database search using pharmacophore models A1NI2
and A3NI1 retrieved 267 (0.0007%) and 1496 (0.0041 %) compounds, respectively.
The pharmacophore A3H1 retrieved a large number of hits, approximately 10% of
the database. The high hit rate corresponds to the non-selective nature of the
pharmacophore. In addition, the short inter-feature distances of various features in
A3H1 contributed to the high hit rate (Figure 20A). On the contrary, pharmacophore
S1A3H1, which incorporated the shape of the nucleotide backbone, retrieved 1830
compounds (0.005 %). Pharmacophore models H1NI2, A1NI2 and A3NI1 possessed
considerable similarity in nature and arrangement of pharmacophore features.
Pharmacophore models H1NI2 and A1NI2 have two NI features in a similar 3D
arrangement. Pharmacophore models A1NI2 and A3NI1 have a NI and an A feature
in common, while H1NI2 and A3NI1 have a NI feature in common. The observed
similarity among H1NI2, A1NI2 and A3NI1 prompted us to look for compounds
common for these three pharmacophore models. Interestingly, a significant number
of compounds were found to be included in all models that share at least a negatively
ionizable feature (Figure 21A). 158 compounds were common for pharmacophore
models H1NI2, A1NI2 and A3NI1. 74 compounds were found common for H1NI2
and A1NI2. Pharmacophore A3NI retrieved 27 and 16 were compounds common to
A1NI2 and H1NI2, respectively. Of the 1754 total hits retrieved by pharmacophore
models H1NI2, A1NI2 and A3NI1, 80 compounds were selected from common and
individual pharmacophore hits for an in vitro screening assay against APE1 (Figure

41
21B). Additionally, 80 compounds were selected from 1830 hits retrieved by the
shape merged pharmacophore S1A3H1. Compounds were selected based on
pharmacophore fit value, pharmacophore mapping pattern, predicted binding
orientation and interactions of the compounds within the active site of APE1.
Docking conformations for all of the molecules discussed were performed using
GOLD
1
.

3.3 Selective Inhibition of APE1 Catalytic Activity by Structurally Diverse
Small Molecules. Two sets of structurally diverse compounds were selected to
screen in an in vitro assay specific to APE1. These molecules were tested in an
electrophoretic APE1 activity assay to determine their percent inhibition of APE1
catalytic activity, relative to a protein-only control. The assay used to derive activity
data for compounds is schematically shown in Figure 17. Of the 80 compounds
selected for screening, 46 compounds inhibited APE1 catalytic activity with an IC
50

value less than 100 μM (Figure 21B). Approximately 57 % of screened compounds
Figure 21. (A) Venn diagram shows the number of compounds retrieved by pharmacophore
models A3N1, H1NI2 and A1NI2 from a database of 362,360 compounds. The compounds in
intersection area are common hits for respective pharmacophore models. (B) Venn diagram shows
number of compounds tested in an in vitro assay specific to APE1. Compounds in parenthesis
inhibited APE1 with an IC
50
value <100 μM (177).

42
that were retrieved by pharmacophore models (H1NI2, A1NI2, and A3NI) bearing at
least one negatively ionizable feature showed potential inhibitory activity against
APE1. Structure and APE1 inhibitory activity of the most active compounds 1 – 21
are given in Table 2. The moderately active compounds S1 – S25 (IC
50
values of >
26 μM and < 100 μM) and inactive compounds S26 – S59 (IC
50
> 100 μM) are given
in Tables S1 and S2 (Appendix I). Surprisingly, a very low number of compounds
retrieved by the shape-merged pharmacophore model S1A3H1 inhibited APE1
activity. Of the 80 compounds screened, only two compounds (S60 and S61) showed
moderate inhibitory potency against APE1 catalytic activity (Table S3, Appendix I).
The structures of inactive compounds S62 – S139 that were retrieved by
pharmacophore S1A3H1 are given in Table S4 (Appendix I). The poor performance
of the pharmacophore S1A3H1 retrieved compounds in an in vitro assay is attributed
to absence of a negatively ionizable feature in the pharmacophore. The presence of a
negatively ionizable feature in the moderately active compounds S60 and S61
(Appendix I) supports this observation. This study demonstrates the importance of
selection of pharmacophore features when pharmacophore hypotheses are generated
exclusively based on the chemical nature of key interactions observed between
substrate and receptor. For APE1, the negatively ionizable feature represents key
interactions between the abasic DNA and the active site. As envisaged by the
pharmacophore models H1NI2 and A1NI2, the most potent compounds 1, 2, 5, 10,
17 – 20, possessing two negatively ionizable carboxylate groups, inhibited APE1
activity with an IC
50
value of < 10 μM (Table 2). Conversely, some of the inactive

43
compounds possessing one or two negatively ionizable groups indicate that mere
presence of the negatively ionizable features alone does not confer activity to
compounds (Table S2, Appendix I). The presence and optimal arrangement of
functional groups complementary to the chemical environment of the APE1 active
site, including negatively ionizable features, are required for a compound to inhibit
APE1 catalytic activity. In general, the structure and inhibitory profile of screened
compounds demonstrate that the presence of at least one negatively ionizable feature
is required for APE1 inhibition. Activity profiles of some of the potent compounds
that inhibited APE1-mediated cleavage of a synthetic oligonucleotide containing an
abasic site analog (tetrahydrofuran) with IC
50
values below 10 μM are shown in
Figure 22 (Table 2).

Figure 22. A representative SDS-PAGE of the most active APE1 inhibitory molecules in the assay.
Lane 1: DNA only, lanes 2 and 27: APE1 + DMSO only, lanes 3-6: compound 22; lanes 7-10:
compound 23; lanes 11-14: compound 1; lanes 15-18: compound 25; lanes 19-22: compound 15;
lanes 23-26: compound 7. All compounds were tested at 100, 33, 11, and 3.7 µM.

44
Compounds 1 and 2 were retrieved by pharmacophore model H1NI2 and inhibited
APE1 activity with IC
50
values of 4 and 9, respectively. The mapping of compound
1 onto model H1NI2 shows an excellent agreement with pharmacophore fit value of
2.84 out of 3.0 between chemical features of 1 and pharmacophoric features of
model H1NI2 (Figure 23A). Pharmacophore model H1NI2 demonstrated high
predictive ability by retrieving (80%) active compounds (Figure 21B). Of the six
compounds selected for screening from H1NI2, 5 compounds 1 – 2, and S1 - S3
inhibited APE1 activity with an IC
50
value of < 100 μM (Table 2 and Table S1,
Appendix I).

S.No. Structure
Activity
IC
50
( μM)
1
H1NI2
N
CO
2
H
Cl
CO
2
H

4 ± 1
2
H1NI2
N
CO
2
H
CO
2
H
S

9 ± 1
3
A1NI2
N N
S
S S
CO
2
H
HO
2
C

17 ± 3
4
H1-A1-NI2
N
N
N
N
CO
2
H
S
CO
2
H

15 ± 6

Table 2. APE1 Inhibitory Activity of Compounds Retrieved Using Pharmacophore Models H1NI2,
A1NI2, A3NI1

45
Table 2. contd.

5
H1-A1-NI2
S
OH
O
S
O
O CO
2
H
CO
2
H

4 ± 1

6
H1-A1-NI2
N N
S
CO
2
H
HO
2
C

15
7
H1-A1-NI2
N N
O
CO
2
H
HO
2
C

16 ± 4
8
H1-A1-NI2
O
HO
2
C
S
CO
2
H

22
9
H1-A1-NI2
O
HO
2
C
S
O
HO
2
C

20
10
H1-A1-NI2 N
O
HO
2
C
S
N
O
S
CO
2
H

6 ± 3
11
H1-A1-NI2
N
O
HO
2
C
S
N
O
S
CO
2
H

12 ± 2
12
H1-A1-NI2
O O O
CO
2
H
O
CO
2
H

8 ± 2

46
Table 2. contd.
13
H1-A1-NI2
O
O
O
O
CO
2
H
HO
2
C

11 ± 2
14
H1-A1-NI2
N
OH
CO
2
H HO
2
C

20
15
H1-A1-NI2
N
N S S CO
2
H HO
2
C

19 ± 9
16
H1-A1-NI2
O
HN
N
H
O O
HO
2
C
CO
2
H

26 ± 6
17
A1NI2-A3NI1
O
CO
2
H
O
CO
2
H

6 ± 1
18
A1NI2-A3NI1
O
N
S
S
O CO
2
H
HO
2
C

6 ± 1
19
A1NI2-A3NI1
O
N N
O
O
O
O CO
2
H CO
2
H

8 ± 1
20
A1NI2-A3NI1
N
N
N
N
HO
2
C
CO
2
H

4 ± 1

47

Table 2. contd.
21
A3NI1
N
N
NH
N
H
2
N
S
O
HO
20 ± 2

The promising APE1 inhibitory activity and high fit value of compounds 1 and 2
onto pharmacophore H1NI2 encouraged us to explore the small molecule databases
to identify readily available analogs. Two retrieved analog compounds of 1 and 2
were screened in the APE1 assay. Compound 22 bearing a para tolyl and 23 with a
4-fluoro phenyl at 5-position of the pyrrole core inhibited APE1 activity with IC
50

values of 12 ± 3 and 9 ± 1 μM (Table 3).

S.No. Structure
Activity
IC
50
( μM)
1
H1NI2
N
CO
2
H
Cl
CO
2
H

4 ± 1
2
H1NI2
N
CO
2
H
CO
2
H
S

9 ± 1
22
N
CO
2
H
CO
2
H

12 ± 3
23
N
CO
2
H
F
CO
2
H

9 ± 1
Table 3. APE1 Inhibitory Activity of Analogs of Compounds 1 and 2.

48
All the compounds were docked on to the active site of APE1 using GOLD. The
docking was carried out to predict the possible binding conformations and
interactions of the compounds inside the active site of APE1. The predicted binding
conformation of compound 1 inside the active site of APE1 demonstrates excellent
agreement between binding interactions and pharmacophore features of H1NI2
(Figure 23B). It also shows that the 1-methyl carboxylate on the pyrrole core of
compound 1 interacts with the side chain guanidine of Arg177. This charge
interaction is represented by one of the NI features in pharmacophore H1NI2. The 2-
propanoate on the pyrrole core of compound 1 represents the second NI feature of
pharmacophore H1NI2 and forms strong H-bonding interactions with Asn174,
Asn212, and His309. The pyrrole core of compound 1 representing the hydrophobic
H1 feature occupied a deep hydrophobic cavity surrounded by amino acid residues
Phe253, Trp280 and Ile282. The superimposition shows a high similarity in
structural and chemical features of bound conformation of compound 1 and the
abasic fragment from the co-crystal structure of APE1 in complex with abasic DNA
(Figure 23C). The limited structure-activity relationship information and predicted
binding conformation of compound 1 indicate that further modifications on the
pyrrole core to fill the deep hydrophobic cavity of the active site of APE1 could
enhance activity.

49
Figure 23. (A) Compound 1 mapped onto pharmacophore H1NI2. There is a good agreement
between chemical features of compound 1 and pharmacophoric features of H1NI2. (B) The predicted
bound conformation of compound 1 inside the abasic DNA binding site of APE1. The blue surface
represents abasic DNA binding region of APE1. Compound 1 is shown as a stick (green) model.
Compound 1 interactions with prominent amino acid residues of APE1 are shown as dashed-lines.
There is a strong agreement between mapped pharmacophoric features of compound 1 and its
interactions with prominent amino acid residues of APE1 active site. (C) The predicted bound
conformation of compound 1 is superimposed onto the bound orientation of abasic unit in the co-
crystal structure of APE1-abasic DNA complex (PDB:1DEW) (177).

Compound 3 was retrieved using pharmacophore A1NI2. Similar to H1NI2, A1NI2
also demonstrated high predictive ability and excellent hit rate (85%) in retrieving
potential APE1 inhibitors (Figure 21B). Of the seven compounds retrieved by
pharmacophore A1NI2, 5 compounds (3, S4 – S7) inhibited APE1 activity with an
IC
50
value of < 100 μM (Table 2 and Table S1, Appendix I). Compound 3, a
symmetric dicarboxylate with a thio-diazole central core inhibited APE1 activity
with an IC
50
value of 17 ± 1 μM (Table 2). Compounds 4 – 16 are common hits for
pharmacophore H1NI2 and A1NI2 (Table 2). Pharmacophore models H1NI2 and
A1NI2 together demonstrated a significant predictive ability (76.6 %) in retrieving
active compounds compared to their individual performance (Figure 21B). Of the 30
common compounds selected for screening, 23 compounds (4 – 16 and S8 – S17)
inhibited APE1 activity with an IC
50
value of < 100 μM (Table 2 and Table S1,

50
Appendix I). The APE1 inhibitory profile of common hits with pharmacophore
models H1NI2 and A1NI2 indicates that along with the two terminal fingerprint NI
features, an optimum sized central hydrophobic core with or without a favorably
substituted H-bond acceptor functional group is essential for a compound to be
recognized by APE1 and to inhibit its activity. Compound 5 with three carboxylate
groups on central aryl system inhibited APE1 activity with IC
50
values of 4 ± 1 μM.
The tricarboxylate 5 and its two analogs (24 - 25) showed strong APE1 inhibitory
activity (Tables 2 and 4). Compounds 24 and 25 inhibited APE1 activity with IC
50

values of 6 ± 2 and 10 μM, respectively.

S.No. Structure
Activity
IC
50
( μM)
5
H1-A1-NI2
S
HO
2
C
S
O
O
CO
2
H
CO
2
H

4 ± 1
24
S
HO
2
C
S
O
O CO
2
H
HO
2
C

6 ± 2
25
O
HO
2
C
S
O
O
CO
2
H
CO
2
H

10
Compounds 6 - 9 are dicarboxylates bearing a substituted central hydrophobic core
(Table 2). These compounds possessed reasonably sized central hydrophobic cores
substituted with H-bond acceptor groups. The inter-feature distance between the two
Table 4. APE1 Inhibitory Activity of Analogs of Compound 5

51
NI features varies across these compounds, for example, compounds 6 and 7. The
APE1 inhibitory activity of compounds 6 and 7 indicates that these compounds have
an optimum arrangement of two NI, hydrophobic and H-bond acceptor features. The
furan-2-carboxylate analogs 8 and 9 inhibited APE1 activity with IC
50
values of 22
and 20 μM, respectively. Compounds 8 and 9 possessed two NI features and an H-
bond acceptor in the central part of the compounds. The slight reduction in potency
of these compounds may be due to lack of an optimum sized hydrophobic central
core. Insertion of an optimum sized hydrophobic group between two NI features of
compounds 8 and 9 could further enhance APE1 inhibitory activity of the
compounds.
Compounds 10 and 11, possessing a 2-oxoindole and a thioxothiazolidinone core
group flanked by two carboxylate groups, exhibited strong APE1 inhibitory activity.
Several analogs (26 - 30) of compounds 10 and 11 also showed strong APE1
inhibitory activity (Table 5). Compound 27 is the most potent among this set of
compounds and inhibited APE1 activity with an IC
50
value of 3 ± 1 μM.
Interestingly, compounds 28 – 30 possessing only one carboxylate group exhibited
APE1 inhibitory activity in a range similar to that of compounds 10 and 11. The
activity profile of compounds 28 – 30 indicates that at least one carboxylate group is
required for a compound to be recognized by APE1 (Table 5). Compound 12 is a
dicarboxylate, possessing a bulky hydrophobic benzocoumarin central core, inhibited
APE1 activity with an IC
50
value of 8 ± 2 μM. Dicarboxylate 13 possessing a bulky
benzodifuranone central core also showed strong APE1 inhibitory activity.

52
Compound 13 inhibited APE1 activity with an IC
50
value of 11 ± 2 μM. Similarly,
dicarboxylates 14 – 16 possessing moderate to bulky hydrophobic central cores
showed moderate APE1 inhibitory activity (Table 2).

S.No. Structure
Activity
IC
50
( μM)
10
H1-A1-NI2 N
O
HO
2
C
S
N
O
S
CO
2
H

6 ± 3
11
N
O
HO
2
C
S
N
O
S
CO
2
H

12 ± 2
26
N
O
HO
2
C
S
N
O
S
CO
2
H

13 ± 6
27
N
O
HO
2
C
S
N
O
S
CO
2
H

3 ± 1
28
N
O
HO
2
C
S
N
O
S
H
N
O
HO

11 ± 2

Table 5. APE1 Inhibitory Activity of Analogs of Compounds 10 and 11

53
Table 5. contd.

29
N
O
HO
2
C
S
N
O
S
Br

6 ± 3
30
N
O
HO
2
C
S
NH
O
S

9 ± 3

Compounds 17 – 20 are hits common to pharmacophore models A1NI2 and A3NI1.
All four compounds selected for screening inhibited APE1 activity with an IC
50

value of ≤ 8 μM. Together, pharmacophore models A1NI2 and A3NI1 showed
excellent predictive ability (100%) in retrieving potential APE1 inhibitors (Figure
21B). Compound 17, a 2, 5-dimethyl-3-carboxy-4-furyl-methyl salicylic acid
inhibited APE1 activity with an IC
50
value of 6 ± 1 μM. The activity profile of the
compound supports the observed agreement between mapped A1NI2 pharmacophore
features onto compound 17 and its predicted binding interactions inside the APE1
active site (Figure 24).

54
Figure 24. (A) Compound 21 mapped onto pharmacophore A1NI2. (B) The predicted bound
conformation of compound 21 inside the abasic DNA binding site of APE1. The blue surface represents
abasic DNA binding region of APE1. Compound 21 is shown as a green stick model. Compound 21
interactions with prominent amino acid residues (stick models) of APE1 are shown as yellow dashed-
lines. (C) The predicted bound conformation of compound 21 is superimposed onto the bound
orientation of the abasic unit in the co-crystal structure of APE1-abasic DNA complex (PDB:1DEW)
(177).

Taking into consideration its binding interactions with APE1 active site and mapping
pattern onto pharmacophore models A1NI2 and A3NI1, a set of additional analogs
were selected and screened. Most of the tested analogs (31 – 35) inhibited APE1
with an IC
50
value similar to the parent compound (Table 6).

55

S. No. Structure
Activity
IC
50
( μM)
17
A1NI2-3NI1
O
CO
2
H
O
HO
2
C

6 ± 1.0
31
O
HO
2
C
S
S
O
CO
2
H

5 ± 2
32
O
CO
2
H
S
O
HO
2
C

6 ± 1
33
O
CO
2
H
O

11 ± 2
34
O
CO
2
H
S
N

27 ± 1
35
O
S
O
HO
2
C

16 ± 5

The inhibitory profile of compound 17 and its analogs shows an inherent SAR in this
class of compounds. Compound 18, a dicarboxylate containing a central core with
hydrophobic and H-bond acceptor features also showed strong APE1 inhibitory
activity. The symmetric dicarboxylates 19 and 20 also inhibited APE1 activity with
IC
50
values of 8 ± 1 and 4 ± 1 μM, respectively. It is interesting to note that several
APE1 inhibitors reported here, for example compounds 3, 6, 7, 13, 16, 19, and 20 are
Table 6. APE1 Inhibitory Activity of Analogs of Compound 17

56
symmetric molecules. The inhibitory profile of the symmetric compounds and the
symmetric nature of pharmacophore models H1NI2 and A1NI2 reveal the
architecture and chemical nature of APE1 active site.
Compound 21 was retrieved by pharmacophore A3NI1. Compared to H1NI2 and
A1NI2, pharmacophore A3NI1 showed low predictability (22.5 %) in retrieval of
potential APE1 inhibitors. Of 31 compounds selected for screening, 7 compounds
(21, S18 – S23) inhibited APE1 activity with an IC
50
value of < 100 μM (Table 2
and Table S1, Appendix I). Compound 21, a 6-amino-9H-purin-8-ylthio-3-
propanoate inhibited APE1 activity with an IC
50
value of 20 ± 2 μM. An excellent
agreement is observed between pharmacophore mapping and predicted binding
interactions of compound 21 inside the active site of APE1 (Figure 25).
Figure 25. (A) Compound 17 mapped onto pharmacophore A3NI1. There is a good agreement
between chemical features of compound 17 and pharmacophoric features of A3NI1. (B) The predicted
bound conformation of compound 17 inside the abasic DNA binding site of APE1. The blue surface
represents abasic DNA binding region of APE1. Compound 17 is shown as a green stick model.
Compound 17 interactions with prominent amino acid residues (stick models) of APE1 are shown as
dashed-lines. There is a strong agreement between mapped pharmacophoric features of compound 17
and its interactions with prominent amino acid residues of APE1 active site. (C) The predicted bound
conformation of compound 17 is superimposed onto the bound orientation of abasic unit in the co-
crystal structure of APE1-abasic DNA complex (PDB:1DEW) (177).

57
The predicted binding conformation of compound 21 demonstrates that the
carboxylate (propanoate) group of the compound that is mapped by a NI feature of
pharmacophore A3NI1 forms a strong charge interaction with the side chain
guanidine of Arg177.
Similarly, the 6-aminopurinyl core of the compound 21, which is mapped by two H-
bond acceptor features of the pharmacophore A3NI1, is involved in two H-bond
acceptor interactions with amino acid residues His309 and Asn174 (Figure 25A –
25B). The superimposition of the predicted bound conformation of compound 21
onto the abasic DNA fragment shows a high similarity in nature and position of key
chemical features of the molecules (Figure 25C). A good match between the
mapping patterns of key pharmacophore features with predicted binding interactions
inside the active site supports the APE1 inhibitory profile of compound 21.
3.4 Selective inhibition of APE1 catalysis versus inhibition of its bacterial
homolog and other DNA-binding enzymes. To assess the specificity towards APE1
of a sample of the most potent APE1 inhibitors, we examined their activity in three
other proteins that are either functionally or structurally homologous to APE1
(ExoIII), originating from a different family of abasic endonucleases (EndoIV), or
exhibiting similar DNA phosphate backbone cleavage activity (HIV-1 integrase).
ExoIII and EndoIV represent two distinct protein families based on the two major
abasic endonucleases found in E. coli, but do not exhibit any similarity in either
structure or catalytic mechanism to each other
159
. ExoIII shares significant
functional homology with APE1 and is thus a valid candidate to experimentally

58
determine the overall specificity of these molecules towards this family of abasic
endonucleases as well as preferential inhibition of human APE1 over its bacterial
counterpart using our assay. As predicted, the overall trend of inhibition by these
molecules is prevalent with ExoIII. However, the IC
50
value and selectivity index of
the inhibitors for the same experimental concentrations of ExoIII are markedly
higher (up to 20-fold) than the relative values for APE1, suggesting a molecular
selectivity for the human enzyme based on differences in key ligand-interacting
amino acid residues within the active site of the respective proteins (Table 7).

compd
APE1
catalysis
(IC
50
, µM)
ExoIII
catalysis
(IC
50
, µM)
Selectivity
Index for
APE1
Inhibitors
with ExoIII
EndoIV
catalysis
(IC
50
, µM)
HIV1 IN
catalysis
(IC
50
, µM)
17 6 ± 1 98 20 >100 >100
18 6 ± 1 70 15 >100 >100
1 4 ± 1 >100 ~26 >100 >100
20 4 ± 1 80 16 >100 >100
10 6 ± 3 60 10 >100 >100
5 4 ± 1 90 21 >100 >100

In contrast to the prevalent, though not comparable, activity of our lead
molecules against APE1 and ExoIII, the co-screen performed using EndoIV and
HIV-1 integrase indicates complete loss of activity. EndoIV is a major abasic
endonuclease in bacteria for which a human counterpart has not been identified, and
operates to cleave at abasic sites in DNA via a similar hydrolytic mechanism to the
ExoIII-like endonucleases. Despite showing comparable incision rates for double-
Table 7. Relative Inhibition Activities of APE1 Inhibitors with Exonuclease III, Endonuclease
IV, and HIV-1 Integrase

59
stranded synthetic oligonucleotides containing a tetrahydrofuranyl residue, EndoIV
displays substantial difference in substrate preference over ExoIII
35
. The
hydrophobic pockets of ExoIII (comprising residues Trp212, Leu226 and Ile228) and
APE1 (comprising Phe266, Trp280 and Leu282) contribute to substrate selectivity of
the ExoIII family members and the results of these experiments indicate selective
inhibition by these compounds of the family of ExoIII-like proteins. Docking scores
were generated using eHITS software for each of the selected small molecules with
APE1 and ExoIII respectively also indicate selectivity for the human enzyme with
the exception of compound 20 (Table 8)
178
.

HIV-1 integrase is a retroviral enzyme that mediates integration of viral DNA into
the host genome. Cleavage of the sugar-phosphate backbone at a dinucleotide
adjacent to a conserved CA on reverse-transcribed viral DNA results in a 3’-
hydroxyl moiety for subsequent enzymatic catalysis. Similar to APE1 catalysis, and
indeed that of most DNA-binding enzymes, the presence of a divalent metal cation is
required for efficient integration
8
. The most potent APE1-inhibitory compounds did
compd
Inhibition of
APE1 catalysis
(IC
50
, µM)
eHITS docking
score for APE1
eHITS docking score
for ExoIII
17 6 ± 1 -2.904 -1.923
18 6 ± 1 -2.858 -0.465
1 4 ± 1 -4.734 -1.890
20 4 ± 1 -1.789 -3.585
10 6 ± 3 -3.289 -2.602
5 4 ± 1 -4.664 -2.601
Table 8. Docking scores of the most potent APE1 inhibitors for APE1 and ExoIII

60
not show any activity against HIV-1 integrase. Given that the predominant functional
moieties of these compounds are dicarboxylate groups, the mechanism of inhibition
of these molecules might be suggestive of metal chelation. Indeed, beta-diketo acids
form a well-studied class of HIV-1 integrase inhibitors whose primary mode of
inhibition is metal chelation
30
. However, since the HIV-1 integrase reaction in our
assays takes place in the presence of a divalent Mn
2+
cation and proceeds unchecked
in the presence of the APE1 inhibitory molecules, it is unlikely that metal chelation
reaction by free acid groups is the mode of inhibition in this case. Table 6
summarizes the relative activities of the six most potent APE1 inhibitors with ExoIII,
EndoIV and HIV-1 integrase. The selectivity index is the calculated fold-change in
IC
50
value of the APE1 inhibitors for ExoIII.
The results of these experiments to ascertain specificity of these molecules for
inhibition of APE1 and other ExoIII-like endonucleases also negates the possibility
that the mechanism of inhibition involved is direct binding to the abasic site itself
within DNA, such as that of methoxyamine
87
. Because of the distinct selectivity of
these molecules to prevent incision by APE1 and ExoIII, it may be argued that they
target the catalytic or DNA-binding residues within the respective active sites of
these enzymes. In addition to the small molecules presented in this study as direct
inhibitors to APE1 catalysis, molecules found to target APE1 allosteric sites leading
to altered DNA binding; or the interruption of protein-protein interactions essential
for DNA repair, may also find clinical success as potentiators of DNA-damaging
agents by specific disruption of the BER pathway.

61
Figure 26. Compound 10 was tested at different concentrations of APE1. Lane 1: DNA only,
lanes 2-6, 7-11, 12-16, 17-21 and 22-26 are as follows: DMSO control, compound 10 at 100, 33,
11, 3.7 µM, respectively. IC
50
values for compound 10 at 0.0005 µM were obtained using further
dilutions of the compound not shown in this figure. Percent inhibition at each dose was plotted
on a log scale to obtain IC
50
values. The box to the right shows the concentration of APE1 and
corresponding basal activity in DMSO.
As mentioned in Chapter 2, to determine optimal assay conditions for the evaluation
of compounds against APE1 catalysis, we had examined the activity of our purified
APE1 protein in a variety of experimental conditions, including protein
concentration, incubation time and temperature. Observing a variation in the
concentration of APE1 used in inhibitor screening assays in prior studies reporting
APE1 inhibitors, we sought to define a non-arbitrary concentration for our assays by
performing dose response experiments at different protein concentrations.

62

As observed in Figure 26, the IC
50
value of compound 10 varies proportionally with
enzyme concentration. Thus, we have reported all of our inhibitor IC
50
values as
assayed at an APE1 concentration of 0.5 nM, at which there is complete enzymatic
cleavage of the 25-mer abasic substrate to cleaved product without subsequent
APE1-mediated exonuclease activity on the cleaved product. Substrate concentration
is the approximate K
m
for the AP lesion
161
. Concentrations lower than 0.5 nM show
low basal abasic site incision activity and result in submicromolar concentrations for
all of our top APE1 inhibitors. Incision reactions were allowed to proceed for 10
minutes at 37ºC, considering the single-time point end measurement nature of our
assay. Since reaction velocity will eventually decrease with substrate depletion, the
effect of an inhibitor should be most prominent within the initial phase of the
reaction. Longer incubation times of 20, 30, and 40 minutes at 37ºC, 30ºC, and at
room temperature, respectively, results in a corresponding increase in IC
50
value,
whereas shorter incubation times produce lower values.
Our report on APE1 assay optimization experiments delineates a need for
standardized assay conditions when screening for inhibitors of APE1. In addition to
our own inhibitors we have demonstrated a similar strong protein concentration-
dependent effect on dose response to APE1 inhibition with 7-nitro-indole carboxylic
acid, a compound formerly shown to inhibit APE1 activity in vitro
91
. The wild-type
enzyme is a robust endonuclease that is highly efficient at incising the DNA-sugar
backbone 5’ to abasic sites, with an approximate maximal single-turnover rate of 850

63
s
-1 93
. Given the high enzyme velocity of APE1, it is perceived that the IC
50
values of
potential small molecule inhibitors to APE1 will be affected by the parameters of the
screening assay.
3.5 Conclusion and implications. In this report we have demonstrated
successful rational design of small-molecule APE1 inhibitors, using a set of 3D
pharmacophore models that were generated based on key interactions of the abasic
DNA within the APE1 catalytic active site. The most potent and selective inhibitors
with IC
50
values below 10 µM are those containing two carboxylate functional
groups in a 3D arrangement with hydrophobic groups, similar to that of the
arrangement of the 3’- and 5’- deoxyribose phosphate groups on abasic DNA. The
APE1 inhibitory profile of some of the most potent compounds indicates that along
with two terminal fingerprint negatively ionizable features or bioisostere groups of
negatively ionizable features, an optimum sized central hydrophobic core with or
without a favorably substituted H-bond acceptor/donor functional group is essential
for a compound being recognized by APE1 and inhibit its catalytic activity. The
observed resemblance between mapping of representative pharmacophore models
such as H1NI2 and A1NI2 on to the most potent compounds and their predicted
binding conformations (key interactions) inside APE1 active site reveals the
chemical environment and architecture of the active site of APE1. The presence and
optimal arrangement of functional groups complementary to the chemical
environment of the APE1 active site are required for a compound to inhibit APE1
catalytic activity. Our assessment of the chemical, steric, and electrostatic nature of

64
the APE1 active site has refined the identification of interactions between abasic
DNA and residues within the active site that translate to efficient inhibition of
catalysis. Furthermore, limited structure-activity analysis of all of the small
molecules tested as part of this study, as well as the potent compounds themselves,
will form an informative structural platform for the rational optimization of lead
compounds to selective small molecule inhibitors of APE1.
The pharmacophore modeling and GOLD docking simulations presented in this
chapter were performed by Dr. Raveendra Dayam. Docking and superimposition of
ligands within the bacterial and human endonucleases for purposes of the co-screen
were conducted by Dr. Jinxia Deng.

65
Chapter 4: Bioisosteres of carboxylic acids show inhibitory activity
against APE1 catalysis

4.1 Bioisosteres of carboxylic acids: tetrazoles. In medicinal chemistry, the use of
bioisosteric molecules in the development of clinically relevant molecules has been
well documented
115
. Lead compounds with desired pharmacological activity may
have associated structural characteristics that adversely affect any number of other
factors such as metabolism, side effects, excretion and bioavailability. Bioisosterism
– the replacement of a functional group with a physicochemically similar group to
elicit parallel biological activity –represents a useful approach for improving the
pharmacokinetic properties of the original target molecule. Bioisosteres can be
classified as either classical or non-classical. Classical bioisosteres have similar
steric and electronic features and the same number of atoms of the groups they
replace
115
.
Non-classical bioisosteres, such as the tetrazoles, do not have the same number of
atoms as the substituent moiety for which they are used as a replacement. They are
capable of maintaining similar biological activity by mimicking the spatial
arrangement or electronic properties of the moiety that is critical for functionality.
The most commonly used bioisostere of the carboxylate moiety, tetrazoles, are non-
classical bioisosteres. They are ionized at physiological pH and exhibit planar
structure. The most attractive feature of tetrazoles is that they are almost 10-fold

66
more lipophilic than the corresponding carboxylate moiety, while having similar
acidity (pKa 4.5-4.9 vs. 4.2-4.4, respectively). This increase in lipophilicity could be
attributed to the increased membrane permeability of tetrazoles over their
corresponding carboxylate analog
59
. Another major advantage of tetrazoles over
carboxylic acids is their resistance to most metabolic degradation pathways.
Tetrazoles are able to escape the Phase II transformation of carboxylic acids, and
thus have a longer absorption half-life. Some examples of tetrazoles as carboxylic
acid bioisosteres include protein tyrosine phosphatase inhibitors, metabotropic
glutamate (mGlu1) receptor agonists, growth hormone secretagogues, and cysteinyl
leukotriene D4 receptor agonists
28, 55, 85, 116
.

Given that the most potent compounds in our study as presented in Chapter 3 are
carboxylic acids, we sought to evaluate their potential as synergistic agents to a
variety of chemotherapeutic agents. As mentioned earlier, and inferred from the
literature, true APE1 inhibitors are more likely to cause cell cycle arrest rather than
cause complete apoptosis in a cell. In the presence of exogenous DNA damage from
a chemotherapeutic agent, the effect of a molecule that inhibits APE1 activity is
Figure 27. Tetrazoles as carboxylic acid bioisosteres

67
exacerbated. The assay of choice for investigating the potential of a molecule as a
synergistic agent is the colony formation assay. The colony formation assay allows
the evaluation of potential agents to cause cell cycle arrest. The measurable endpoint
of this assay is the cell colonies that proliferate from singly seeded cells after
exposure to the agent of interest for a given period of time. Unlike cytotoxicity
assays that indicate the number of cells that are killed by an agent, colony formation
assays also known as cell survival assays, enumerate the ability of cells to recover
from cell cycle arrest and proliferate thereafter.
Optimization of colony formation assay for 96-well format. Because of the large
number of compounds evaluated in this study, we optimized the cell cycle assay for
use in a 96-well format. Different cell numbers of HCT116 p53 wild-type colon
cancer cells, from 75 cells/ well to 1000 cells/ well were plated on a 96-well plate
with and without DMSO. The DMSO was used to simulate the toxicity of a potential
cytotoxic agent. After exposure for 48 hours, cell medium was changed and the cells
allowed to grow in fresh medium for a further 72 hours. Care was taken not to allow
the cells to become confluent due to the small volume of the plating vessel. The
amount of time, and the number of cells seeded per well for which cells are allowed
to grow until distinct colonies (~50 cells/ colony) can be visualized, varies with cell
type and turnover. For HCT116 p53 wild-type cells that have a doubling rate of
approximately ~16 hours, we have a standard of 200 cells/ well in a 96-well format.
Normal growth time post-removal of medium containing the agent of interest has
been determined to be about five days.

68
The colony formation assays performed in HCT116 p53 wildtype cells with active
compounds from our pharmacophore study did not yield any small molecules that
were clearly synergistic at 20µM with either gemcitabine or methotrexate at their
relative IC
25
values. Gemcitabine is a nucleoside analog to deoxycytidine used in the
treatment of non-small cell lung cancer, pancreatic, bladder, and breast cancer, with
less debilitating side effects than other chemotherapeutic agents. Phosphorylation of
gemcitabine causes it to act as an antimetabolite that inhibits DNA synthesis. The
siRNA-mediated knockdown of APE1 has been shown to potentiate the cytoxicity of
gemcitabine
81
. Methotrexate is also an antimetabolite that acts by inhibiting the
metabolism of folic acid. It competitively and reversibly inhibits dihydrofolate
reductase, an enzyme that participates in tetrahydrofolate synthesis. Folic acid is
necessary for the de novo synthesis of thymidine. Its cytotoxicity occurs specifically
during nucleic acid synthesis, and is thus cytotoxic during the S-phase of the cell
cycle. The results of our colony formation assays of our APE1 inhibitory compounds
are shown in Appendix II. We believe that the highly charged nature of these
dicarboxylate molecules prevent cell membrane permeability, and are unable to exert
a synergistic effect in a cellular environment. Thus, we turned our attention to
bioisosteres of carboxylic acids to evaluate as potential APE1 inhibitors, not in vitro
and as cell-based synergistic agents.
4.2 Optimization of tetrazoles-containing compounds as small-molecule
inhibitors of APE1. In an attempt to find tetrazole containing compounds with god
biological activity, we docked a database of 20,000 molecules using the GOLD

69
docking program into the active site of APE1 (PDB ID: 1DE9). From this run, the
top 2000 compounds were selected for a second round of docking, of which 500
compounds were docked using various filters. The top 100 compounds were visually
inspected and 50 compounds were selected for the in vitro assay based on the
following criteria: structural diversity of the compounds, high docking values, or
adoption of a reasonable binding orientation at the active site. On the basis of the
above observations, a total of 50 compounds were selected for testing. Chemical
structures and purity of each compounds was confirmed before initiation of the
biological assay.

As shown in Figure 1, of the 50 compounds selected for biological testing, nine
compounds inhibited APE1 catalytic activity with an IC
50
value less than 100 μm.
The biological data and GOLD docking fitness values are summarized in Table 9.
The run also picked out a compound that was previously reported by us to be one of
the most potent inhibitors to APE1.

Figure 28. In silico screening protocol implemented in the discovery of novel APE1inhibitors.
The 7-8 times and 2 times are 7-8 speedup and 2 times speed-up procedures implemented in
GOLD. Courtesy: Dr. Srinivas Odde, Neamati laboratory.

70

ID Structure
GOLD
Fitness
Activity
IC
50
( μM)
1
N
O
OH
HO
O
Cl

53.5 4±1
2
H
N
O
OH
Br
Br
O HO
O

58.72 22 ± 8
3 S
OH
O
NH
O
F
F
F

53.5 48 ± 13
4
N
H
O
N
H
OH O
O

53.67 27 ± 11
5
N
H
O
OH
O
S
NH
N
N

44.01 51 ± 7
6
N
S
S
N
H
O
HO
O

44.3 33 ± 6

Table 9. Compounds yielded by in silico docking of structurally diverse
compounds

71

Table 9, continued
7
NH
N
H
O
O
OH
O

43.11 15
8
N
N
F
F
F
N
HO
O

54.17 47 ± 6
9
F
F
F
N
S
N
HO
H
N
O
Cl

40.67 90 ± 14

We next conducted a similarity and sub-structure database search of an in-house
database using the tetrazole chemical motif. This search resulted in additional set of
five active tetrazole compounds reported in Table 10 along with their APE1
inhibitory activity.

S.No Structure
GOLD
Fitness
Activity
IC
50
( μM)
10
N
N
N
HN
O
OH

39.89 42 ± 14
11
HN
N
N
H
N
N
N
N
N

38.84 21 ± 4

Table 10. Tetrazole compounds yielded by substructure search

72
Table 10, continued
12
O
F
F
F
HN
N
N
N

47.81 35 ± 13
13 N
N
S
N
N
N
H
N
O

45.88 50 ± 12
14
HN O
NH
N
N
N
O
O

37.45 57 ± 6

As can be seen in Table 10, the five compounds containing the tetrazole moiety do
not have comparable activity to the reported carboxylic acids, whose half-maximal
inhibitory concentrations are below 10 μM. In addition, colony formation assays of
these compounds with gemcitabine and methotrexate do not show synergy, nor do
they exhibit cellular toxicity on their own. While it is cannot be ruled out that these
agents may show synergy with other DNA damaging agents, their inherent low
APE1 inhibitory activity implies that synergy may not be observed.
4.3 Conclusions and implications. The idea of tetrazole bioisosteres of carboxylic
acids for the development of cell-based inhibitors to APE1 is viable given their
success in other areas of therapeutics. As shown in Chapter 3, when designing
inhibitors to APE1, the shape of the molecule is important. Linear molecules with
acidic moieties flanking the ‘ends’ presumably mimic the phosphate backbone of an
abasic DNA fragment, and are thus able to bind within the active site and inhibit
catalysis. It should be noted that none of the tetrazole molecules presented herein

73
possess that linear shape, which appears to be an important structural component for
an optimum APE1 inhibitor. A second important observation is that the most potent
inhibitors contain more than one carboxylate functional group, whereas all but one of
the tetrazoles tested possess a single tetrazole moiety. The activity of these tetrazoles
parallels the activity of certain monocarboxylate molecules that show limited APE1
inhibitory activity. Given that Compound 11, containing a ditetrazole moiety, also
has the corresponding best half-maximal inhibitory concentration of the compounds
tested, it is possible that such molecules – if also possessing a linear shape – may
prove to exhibit excellent inhibitory activity. Indeed, further development of
tetrazoles in this regard could entail simple replacement of the carboxylate moiety on
small molecules that have been shown to have low IC
50
values. Such a venture may
not only result in the elegant development of potential cellular inhibitors to APE1,
but also provide a wealth of insight into the mechanism of action of the
dicarboxylate inhibitors of APE1. Nevertheless, this observation is the first in current
literature to implicate tetrazoles as a novel class of APE1 inhibitory compounds.
The computational work presented in this chapter for the tetrazole compounds was
performed by Dr. Srinivas Odde, of the Neamati laboratory.

74
Chapter 5: Biochemical Validation of APE1 as an Anticancer
Therapeutic Target

This chapter presents work performed towards biochemical validation of APE1 as an
anticancer target for drug development. Our approach was two-fold: the clinical
impact of APE1 polymorphisms in colorectal cancer, and the establishment of
biochemical models of APE1 gene manipulation to better understand the
mechanisms of APE1 inhibition. In addition, we experimented with methoxyamine, a
previously described indirect inhibitor of BER, to establish proof-of-concept studies
with interesting observations.
5.1 APE1 polymorphisms in cancer disease studies
APE1 polymorphisms are of significance because of its important role in the
maintenance of genome integrity. However, to correlate a particular disease state to a
single polymorphism that causes a decrease in repair activity is often difficult due to
differences in cancer types, patient populations and exogenous determinants of
cancer progression. In general, it is believed that polymorphisms in clusters of genes
that regulate such processes as DNA repair concurrently contribute to a cancer
phenotype
75
. While trying to correlate a single polymorphism to cancer
susceptibility does not reflect the multi-faceted nature of cancer causation, several
studies suggest that BER gene polymorphisms are important in cancer risk. Common
polymorphisms of APE1 include Q51H, I64V and D148E. The D148E
polymorphism has been well-characterized and the purified mutant protein does not

75
show any difference in endonuclease activity from wild-type
58
. Polymorphisms in
APE1 in conjunction with other repair proteins have been shown to impact certain
cancers including melanoma, prostate, breast, lung, endometrial, and pancreatic
21, 58,
67, 83, 117
. APE1 gene expression has also been associated with some prognostic
significance. In cervical, prostate, ovarian, myosarcoma, and germ cell cancers,
APE1 protein levels are increased compared to normal tissue
45
. In other types,
protein localization have been shown to differ between normal and cancerous tissue.
Altered patterns of APE1 localization need to be further characterized on a
mechanistic basis to fully establish a relationship between expression patterns and
cancer. Table 12 summarizes the current information available for APE1
localization.
Table 11. Summary of APE1 polymorphisms and their impact in cancer

76

Table 12. Summary of APE1 localization patterns in normal and cancerous tissue

5.2 The APE1 D148E polymorphism in colorectal cancer survival
We examined the APE1 D148E polymorphism alone and in combination with the
XRCC1 R399Q in two populations of colorectal cancer patients. The first was a
group of 197 patients with locally advanced colon cancer that were being treated
with 5-FU based adjuvant chemotherapy. [bolus 5-FU: n = 30 (15%); infusional 5-
FU: n = 167 (85%)]. The second group of 318 patients had metastatic colorectal
cancer and were enrolled in at least one chemotherapy clinical trial while being
treated at USC/ Norris Cancer Center or Los Angeles County hospital. All patients
were treated with 5FU-based chemotherapy regimens. Although the treatment
regimens varied between patients, the majority were exposed to similar
chemotherapies. All 318 patients received treatment with FU, 298 patients (94%)

77
received treatment with FU/ irinotecan, and 279 patients (88%) received treatment
with FU/ oxaliplatin. The primary endpoint of this study was survival (OS). OS was
determined by calculating the difference between the date of first treatment at the
relevant institution and the date of last follow-up appointment or date of death from
disease. The association of survival with a patient’s clinicopathic characteristics
(age, sex, race, stage, tumor grade, and type of chemotherapy) were assessed using
univariate survival analyses (log-rank test).
Results. In both metastatic and adjuvant populations, the APE1 D148E
polymorphism was not correlated to survival both alone and in conjunction with the
XRCC1 R399Q SNP.
However, an interesting observation was made with regard to the APE1 D148E
polymorphism in the metastatic population. In the univariate analysis, we
observed that patients carrying DE or EE (median overall survival
(OS) = 10.3 months) had significantly shorter overall survival compared to
patients carrying DD (median OS = 13.1) in males (p = 0.031), but not in
females (p = 0.87).

Table 13. Distribution of APE1 genotypes by gender
APE1 Males Females
DD 107 (72%) 86 (72%)
DE 38 (26%) 33 (28%)
EE 3 (2%) 0

78
Genotyping. Whole blood was collected and genomic DNA was extracted using the
QIAamp kit (Qiagen, California).
Figure 29. The DD genotype confers a significant survival benefit on males with metastatic colorectal
cancer (P = 0.031).
This observation is the first to be made that correlates APE1 polymorphisms to
gender. Similar analyses of larger population should provide more validation of this
observation and could extend to other types of cancer. The statistical work in this
chapter was performed by Dr. Dong Yun Yang of the Heinz-Josef Lenz laboratory,
Norris Cancer Center. All RFLP analyses were performed under the supervision of
Dr. Wu Zhang on samples available in the lab.

79
5.3 Biochemical models to characterize potential inhibitor behavior in
clinically contingent pharmacological backgrounds
To be able to qualitatively understand the effect of a BER inhibitor within a cell, we
proposed a research plan linking biochemical models of manipulation of APE1
expression in a pharmacological cellular environment, with structural studies both in
silico and in vitro to identify potential small-molecule inhibitors to APE1 activity. A
mammalian expression vector used to overexpress APE1 in a mammalian cancer cell
line would provide the opportunity to demonstrate the cellular response of commonly
used colorectal chemotherapeutic agents—such as fluoropyrimidines and platinum
agents—to overexpressing an enzyme known to play a crucial role in the essential
mammalian base excision repair (BER) pathway. Using RNA interference and
chemical methods, we aimed to characterize the consequences of downregulating
this enzyme and promoting aberrant BER within different clinically contingent
pharmacological backgrounds to simulate the potential effects of an inhibitor to
APE1 catalytic activity. The biochemical endpoints resulting from the alteration of
APE1 expression in these models were assayed using conventional growth
inhibition, cell cycle, and recovery assays.
Construction of an APE1 overexpression vector. The expression system selected
is a tetracycline-inducible expression vector, the Tet-Off Gene Expression System
(Clontech). Figure 29 shows a schematic of gene regulation in the Tet-Off system.

80
Figure 30. Gene regulation in the Tet-Off system. (Courtesy: Clontech, Mountain View, CA).

In bacterial systems, the Tet repressor protein (TetR) negatively regulates the genes
of the tetracycline-resistance operon. TetR binds to the tet operator sequences (tetO)
in the absence of tetracycline. TetR and tetO provide the basis of regulation and
induction for use in the mammalian experimental system.
The first component of the Tet system is the regulatory protein, based on TetR.
Addition of a viral activation domain converts the TetR protein from repressor to
transcriptional activator, tTA. tTA is encoded by the pTet-Off regulator plasmid,
shown in Figure 30a. The plasmid includes a neomycin resistance gene to permit
selection of stably transfected cells. The second critical component is the response
plasmid which expresses, in this case, the APE1 gene cloned into the EcoR1 –
BamH1 restriction sites within the multiple cloning site. This vector, pTRE-Tight,
contains the TRE, tetracycline response elements, which consists of seven direct

81
repeats of a 42-bp sequence containing tetO, located upstream of a CMV promoter.
this minimal promoter lacks the strong enhancer elements normally associated with
CMV promoters, and thus ensures that there is no background expression of APE1
from the TRE in the absence of binding by the TetR domain of tTA. The expression
system is achieved by setting up a double-stable Tet cell line which contains both the
regulatory and response plasmids. When cells contain both plasmids, APE1 is only
expressed upon binding of tTA to the TRE (Figure 30). In the Tet-Off system, tTA
binds the TRE and activates transcription in the absence of tetracycline or
doxycycline. Stable transfection of the pTet-Off plasmid was performed to generate
a pTet-Off HCT116 p53 wild-type as described in Chapter 6, Materials and methods.
APE1 was successfully cloned into the pTRE-Tight vector using cDNA generated
from HCT116 p53 wild-type cells.

82

Figure 31a. Map of the pTet-Off regulator plasmid. (Courtesy: Clontech, Mountain View, CA)

Figure 31b. Map of the pTRE-Tight response plasmid, showing multiple cloning site (Clontech,
Mountain View, CA). As shown on the right, APE1 was cloned into the EcoR1 and BamH1 sites of
the pTRE-Tight.

83

Optimization of real-time PCR-based method for validation of cellular APE1
gene expression. Before conducting experiments of gene overexpression or siRNA-
mediated downregulation of APE1, we attempted to ascertain basal levels of
expression in a panel of cancer cell lines using Western blotting. This method of
protein quantitation repeatedly showed that
APE1 seemed to be constitutively expressed
across several cell lines. We then decided to
develop a real-time PCR based method of
protein expression quantitation to validate
and confirm those results obtained by
Western blotting. Real-time PCR primers
were manually designed to flank exon-exon
border sequences so that they would not be
able to hybridize within genomic DNA. In
this case, the primers hybridize between base
pair 229 and base pair 520 (GenBank accession NM_080648. Figure 32 shows a
regular PCR performed with these primers amplifies a very specific product of ~300
base pairs, which corresponds to the expected cDNA-derived product of 291 base
pairs for this primer set. Sequences are as follows:
Forward primer: 3’-AAGAAGAAAGGATTAGATTGGGTAAA
Reverse primer: 3’-CCTTCCTGATCATGCTCCTC
Figure 32. Validation of real-time PCR
primers. The expected product for these
primers is a 291-bp product as observed.
Lane 1, marker. Lanes 2-5: PCR product
obtained using serial dilutions of cDNA:
1:2, 1:4, 1:8 and 1:16, respectively.

84
These primers have subsequently been used to determine the basal expression in a
panel of cancer cell lines by a post-doctoral fellow in our group (Figure 33). Further
optimization of the Western blotting protocol is required, as this method shows
consistently shows the same levels of APE1 expression across all cell lines, while the
real-time PCR exhibits a different result.
Figure 33. APE1 mRNA expression in a panel of cell lines relative to HT29 expression. (Courtesy:
Dr. Huabiao Zhao.)

Transfection of colon cancer cell lines with pTre-Tight:APE1 and siRNA
targeted to APE1. siRNA sequences targeted to the APE1 gene were ordered from
Dharma on, according to published sequences. These, as well as the cloned
overexpression vector were transfected into cells several times, as described in
Chapter 6. However, protein quantitation by Western blot did not yield conclusive
results.
Currently, with a working real-time based expression quantitation method in place, it
is hoped that these molecular biological ‘tools’ of APE1 gene expression will prove

85
useful in establishing the proposed biochemical models by which inhibition of
APE1, and consequent disruption of BER may be better studied and understood.
5.4 Identification of a role for BER in oxaliplatin toxicity.
To elucidate the role of BER in the mechanism of toxicity for chemotherapeutic
agents currently used in colon cancer treatment, we utilized the BER inhibitor MX in
HCT116 colon cancer cell lines. Cells were treated with 5-fluorouracil (5-FU) and
oxaliplatin in the presence and absence of MX, and clonogenic survival assays were
carried out to observe the effect of MX on 5-FU- and oxaliplatin-mediated
cytotoxicity.
Methoxyamine potentiates the cytotoxicity of 5-U but not oxaliplatin. As shown
in Figure 34a, MX at a concentration of 10mM shows a slight potentiation of 5-FU
mediated toxicity HCT116 p53 wild-type cells. This effect is seen at both 48 and 72
hour drug exposures. This suggests that direct inhibition of APE1 activity may have
a more significant effect in enhancing 5-FU mediated cytotoxicity.
A surprising result is observed with oxaliplatin. In HCT116 p53 wild-type cells a
statistically significant increase in cellular recovery in the MX-treated cells
compared to MX-untreated cells at 48 and 72 hours is observed (Figure 34b). Most
interestingly, the recovery afforded to cells in the presence of MX treatment with
oxaliplatin is not observed in p53 null HCT116 colon cancer cells, suggesting that
BER plays a p53-mediated role in the correction of oxaliplatin-mediated DNA
damage. Similar results are observed with cisplatin. The protective effect of MX is
also observed with MTS assays. While these results may be opposite to what is

86
expected, because MX is an indirect inhibition of APE1 activity, these initial
observations are not indicative of the effect direct inhibition of APE1 may have on
oxaliplatin treatment. Moreover, the results provide a basis for elucidating a
previously unknown role for BER in platinum agent cytotoxicity. In addition, they
may provide the answer to the question that arises of It is of even greater interest to
observe cellular response to direct inhibition of APE1 in the same experimental
conditions.
Figure 34a. MX slightly potentiates 5-FU mediated toxicity in HCT116 p53 wild-type cells.

Figure 34b. MX has a protective effect to oxaliplatin treatment in HCT116 p53 wild-type cells.

Figure 34c. MX potentiates oxaliplatin toxicity in HCT116 p53 null cells.

87
Figure 34c. MX has a protective effect in cells that do not express WT p53.

We repeated the oxaliplatin experiments in HCT116 p53 WT cells at longer
treatment times to make time point comparisons. The protective effect of
methoxyamine on these cells was made more apparent at longer drug exposures
(Figure 35).
Figure 35. The protective effect of MX is more apparent at longer drug exposures.

88
Finally, we duplicated the same conditions in cell viability (MTT) assays to see if the
same effect was produced. MX protected p53 null cells from oxaliplatin toxicity, and
the same result was observed for experiments conducted with cisplatin.
Figure 36. Cell viability assays show the same result as colony formation experiments.

5.5 Conclusions and implications. These studies should provide insight into the
contribution of mammalian DNA repair pathways to cytotoxicity and resistance of
colorectal cancer cells to core cancer treatments. This study would be the first to
show that downregulation of a key mammalian DNA repair enzyme confers
sensitivity to fluoropyrimidines in human colorectal cancer cells. While the effects of
downregulating APE1 on other cancer treatments such as the platinum agents is not
known, prior studies have shown that oxidative agents, alkylating agents and
ionizing radiation are indeed potentiated by modulation of BER enzymes, including
APE1 (reviewed in Fishel et al, 2007)
50
. The biochemical validation of APE1/ BER
as a modulator of chemotherapeutic response in colorectal cancer provides a
platform for rational drug design of clinically promising APE1 inhibitors.
Potentiation of cytotoxicity by an APE1 inhibitor should show synergistic effects
with current core anticancer treatments and potentially lead to reduced drug dosages

89
and corresponding reduction in chemotherapeutic side effects. This hypothesis is
enhanced by the testing in clinical trials of a small molecule, methoxyamine (MX),
in conjunction with temozolomide (Temodar) treatment in human gliomas and
ovarian cancer
49, 168
. Methoxyamine binds irreversibly to abasic sites and interrupts
the recognition of the abasic site by APE1, and has been shown to potentiate
temozolomide toxicity, among that of other agents
49, 132, 168, 169
. While this inhibition
of APE1 is indirect, it underscores the ability of BER to modulate tumor response to
alkylating agents. Additionally, methoxyamine shows no toxicity at clinically
relevant doses on its own
132, 168, 169
. We believe that direct inhibition of APE1 will
have a more potent effect on sensitizing cancer cells to treatment and provide an
opportunity to scrutinize the mechanisms of cytotoxicity of core chemotherapeutic
agents that are heretofore not known, such as that of oxaliplatin. As with existing
DNA repair inhibitors to PARP-1, DNA-PK, MGMT and ATM, it is believed that
inhibition of APE1/BER may also provide selective tumor killing to cancers with
sub-optimal repair activity, and refractory cancers for which no standard treatment
exists.
The statistical work in this chapter was performed by Dr. Dong Yun Yang from the
laboratory of Dr. Heinz-Josef Lenz. RFLP analyses were carried out by the author
under guidance from Dr. Wu Zhang. All other biochemical assays and experiments
were conducted by the author under guidance of Dr. Peter Wilson, Dr. William
Fazzone and Melissa LaBonte of the Ladner laboratory.

90
Chapter 6: Materials and methods

6.1 Methods used for APE1 catalytic assay development (Chapters 2-4)
6.1.1 Expression and purification of recombinant APE1
His-tagged APE1 was purified from Escherichia coli M15 cells (Qiagen, Valencia,
CA). Qiagen protocol was used to generate competent M15 cells. A pellet of
bacterial cells from an overnight culture was resuspended in TFB1 buffer (100 mM
RbCl, 50 mM MnCl
2
, 30 mM potassium acetate, 10 mM CaCl
2
, 15% glycerol) and
incubated on ice for 90 min. The cells were then recollected by centrifugation at
4000g for 5 min at 4°C and then resuspended in TFB2 buffer (10 mM MOPS, 10
mM RbCl, 75 mM CaCl
2
, 15% glycerol). An aliquot of these cells were transformed
with a Qiagen pQE30 plasmid containing the APE1 gene sequence and ampicillin
resistance. The APE1 plasmid was expressed in the M15 expression strain after
induction by IPTG (1 mM) at an absorbance of 0.6 - 0.8 optical density at 595 nm.
The culture was allowed to grow for an additional 3-4h at 37ºC. This was followed
by centrifugation of the cells at 3000 rpm in a bucket rotor centrifuge (Beckman) for
20 min. Pelleted cells were resuspended in lysis buffer (20 mM HEPES, pH 7.5, 5
mM imidazole, 100 mM NaCl) and passed twice through a French Press (Thermo
Spectronic, Madison, WI). Lysate was centrifuged at 31,000 g and the pellet was
solubilized in a buffer containing 20 mM HEPES, pH 7.5, 5 mM imidazole, and 10
mM CHAPS. Recombinant APE1 protein was purified using Ni-affinity
chromatography with Swell-gel Nickel-chelated discs (Pierce, Rockford, IL). The

91
protein was eluted from the column with increasing concentrations of imidazole from
40mM to 1M. An aliquot of each concentration post-elution was run on an SDS-
PAGE gel and fractions containing protein were dialyzed in Spectra/Por molecular
porous membrane tubing, MWCO 12-14,000 (Spectrum Laboratories, Inc., Houston,
TX). The protein was dialyzed in buffer containing 20 mM HEPES, pH 7.5, 500 mM
NaCl, 40% glycerol, 0.2 mM EDTA, and 1 mM dithiothreitol (DTT). After dialysis,
the purified enzyme solution contained 50 mM NaCl, 1mM HEPES, pH 7.5, 50 µM
EDTA, 50 µM DTT, and 10% glycerol (w/v). Aliquots of the protein were then
stored at -80ºC and removed when necessary for use.
6.1.2 Enzymatic Assays. To determine the extent of abasic residue cleavage by
APE1, ExoIII and EndoIV, or integration by HIV-1 integrase, recombinant APE1
was preincubated at a final concentration of 0.05 nM with the potential inhibitors in
reaction buffer (50 mM NaCl, 1 mM HEPES, pH 7.5, 50 µM EDTA, 50 µM
dithiothreitol, 10% glycerol (w/v), 7.5 mM MnCl
2
, 0.1 mg/ml bovine serum
albumin, 10 mM 2-mercaptoethanol, 10% DMSO, and 25 mM MOPS, pH 7.2) at 30
ºC for 10 min. Then, 200 nM of the 5'-end
32
P-labeled linear oligonucleotide
substrate was added, and incubation was continued for an additional ten minutes.
Reactions were quenched by the addition of an equal volume (16 µl) of loading dye
(98% deionized formamide, 10 mM EDTA, 0.025% xylene cyanol and 0.025%
bromophenol blue). An aliquot (5 µl) was electrophoresed on a denaturing 20%
polyacrylamide gel (0.09 M tris-borate pH 8.3, 2 mM EDTA, 20% acrylamide, 8M
urea). The gels were dried, exposed in a PhosphorImager cassette, analyzed using a

92
Typhoon 8610 Variable Mode Imager (Amersham Biosciences), and quantitated
using ImageQuant 5.2. The percent inhibition (% I) was calculated using the
following equation: %I = 100 x [1-(D – C)/(N – C)] where C, N, and D are the
fractions of 26-mer substrate converted to 13-mer incision products (or integration
products) for DNA alone, DNA plus enzyme, and enzyme plus drug, respectively.
The IC
50
values were determined by plotting the logarithm of drug concentration
against percent inhibition of enzymatic activity. The optimum concentration of APE1
protein in the assay is the concentration at which there is complete conversion of the
synthetic abasic-site containing oligonucleotide substrate to its cleaved product
without subsequent exonuclease activity. Additional assay optimization information
can be found in Supplementary Information and Figure S1.
6.1.3 Restriction Endonuclease Assays. Primers (BamH1 sequence:
GATCGGATCCTCACAGTGCTAGGT ATAGGGT; EcoR1 sequence:
CCGGAATTCATGCCGAAGCGTGGAAAAAGG) containing the appropriate
restriction sites were incubated with the enzyme and the inhibitory molecules at 10
µM for 3 hours at 37ºC. The products of these reactions were electrophoresed on a
0.5% agarose gel containing ethidium bromide and visualized under UV light using a
Biorad ChemiDoc XRS system (Biorad).

93
6.1.4 Preparation of Oligo Substrate.
Top strand (5’ATTTCACCGGTACG(F)TCTAGAATCCG 3’) containing the
tetrahydrofuran synthetic abasic residue (F) is radiolabeled with γ-P
32
using T4
polynucleotide kinase (Epicenter, Madison WI).
The complementary strand (3’-TAAAGTGGCCATGC(C)AGATCTTAGGC-5’) is
added in 1.5 molar excess after deactivation of the kinase by heating to 95ºC, and
annealed by allowing to cool slowly to room temperature. The resulting cooled
mixture is centrifuged through a Spin-25 mini-column (USA Scientific, Ocala, FL)
to separate annealed double-stranded oligonucleotide from unincorporated material.
The oligonucleotides used in the HIV-1 activity assays, 21-top (5'-
GTGTGGAAAATCTCTAGCAGT-3') and 21-bot (5'-
ACTGCTAGAGATTTTCCACAC-3') were labeled using a similar protocol.
6.1.5 Oligonucleotides. For the APE1, ExoIII, and EndoIV enzymes the following
oligonucleotides were used: The top strand (5’-
ATTTCACCGGTACG(F)TCTAGAATCCG-3’) containing the tetrahydrofuran
synthetic abasic residue (F) and the bottom strand (3’-
TAAAGTGGCCATGC(C)AGATCTTAGGC-5’). The HIV-1 integrase catalytic
assay uses 21-top (5'-GTGTGGAAAATCTCTAGCAGT-3') and 21-bot (5'-
ACTGCTAGAGATTTTCCACAC-3') oligonucleotides specifically recognized by
the enzyme for catalysis.
6.1.6 Biological Materials, Chemicals and Enzymes. All compounds were
dissolved in DMSO to a final stock concentration of 10 mM and stored at -20ºC.

94
Dilutions were performed in DMSO. The expression system for APE1 was a kind
gift from the laboratory of Dr. Tom Curran, Department of Developmental
Neurobiology, St. Jude Children's Research Hospital, Memphis, TN. Wild-type
integrase was purified from an expression system generously provided by Dr. Robert
Craigie, Laboratory of Molecular Biology, NIDDK, NIH, Bethesda, MD. ExoIII,
EndoIV and the restriction endonucleases were purchased from New England
Biosciences (Ipswich, MA). The synthetic oligonucleotide containing a
tetrahydrofuran abasic site analog was purchased from The Midland Certified
Reagent Company (Midland, TX). The oligonucleotides used in the HIV-1 integrase
activity assay were purchased from Integrated DNA Technologies (Coralville, Iowa).

6.2 Computational methods (Chapter 3-4)
6.2.1 Generation of Pharmacophore Models. The 3D pharmacophore models were
generated using the abasic DNA fragment as a template. The abasic fragment, 3’, 5’-
deoxyribose phosphate was extracted from the co-crystal structures of human APE1
bound abasic DNA (1DEW.PDB, and 1DE9.PDB).
103
The coordinates of abasic
fragment were exported to Catalyst (Accelrys, Inc) for pharmacophore mapping.
58

Using the Pharmacophore View/Map work bench, the abasic fragment was explored
to identify presence of various pharmacophore features. A set of pharmacophore
features were selected to represent key interactions observed between the abasic site
and the active site of APE1 in the co-crystal structures of human APE1 bound abasic
DNA. The pharmacophore features selected are negatively ionizable (NI),

95
hydrophobic (H) and H-bond acceptor (A) features. Features were first mapped onto
the abasic fragment. Reasonable features were then selected and merged into
searchable 3D pharmacophore models. Four pharmacophore models H1NI2, A1NI2,
A3NI1 and H1A3, were generated by combining NI, H and A features.
6.2.2 GOLD Docking Simulations. Predicted binding interactions of compounds
within the APE1 active site were generated using GOLD docking software.
Compounds with free carboxyl groups were modeled in their carboxylate form as
this form is considered biologically relevant. All the water molecules present in
protein were removed, and hydrogen atoms were added to the protein considering
appropriate ionization states for both the acidic and basic amino acid residues.
Docking was performed using version 3.2 of the GOLD: Genetic Optimization for
Ligand Docking (Cambridge Crystallographic Data Centre) software package.
1, 145,
146
A 20 Å radius active site was defined using the backbone N atom of amino acid
residue A174 as the center of the active site. All the compounds were docked into the
active site of the APE1. On the basis of the GOLD fitness score, for each molecule, a
bound conformation with high fitness score was considered as the best bound
conformation. All docking runs were carried out using standard default settings with
a population size of 100, a maximum number of 100 000 operations, and a mutation
and crossover rate of 95. The fitness function that was implemented in GOLD
consisted basically of H-bonding, complex energy, and ligand internal energy terms.
6.2.3 Docking Studies using eHITS. Docking studies were carried out using GOLD
and eHiTS programs. eHiTS evaluates all the possible protonation states for the

96
receptor and ligands automatically for every receptor-ligand pair and systematically
covers the part of the conformational and positional search space to avoid severe
steric clashes.

Our docking tasks were performed on APE1 (PDB ID: 1DE9) and
ExoIII (PDB ID: 1AKO). The two structures were first aligned based on their pair-
wised sequence alignment. The high degree of sequence identity and homology
between the two enzymes allowed us to use a highly conserved aspartic acid residue
(Asp151 in ExoIII and Asp210 in APE1) as a clip file to align enzymes.

6.3 Molecular biology protocols (Chapter 5)
6.3.1 Western Blotting
6.3.1.1 Protein extraction. Monolayer cultures were rinsed with PBS,

and harvested
in PBS containing a 1:1000 dilution of the Protease

Inhibitor Cocktail (100 mM
AEBSF, HCl, 80 µM aprotinin,

5 mM bestatin, 1.5 mM E-64, 2 mM leupeptin
hemisulfate, and

1 mM pepstatin A) purchased from Calbiochem (La Jolla, CA).

Cells were lysed by sonication at 1 minute intervals three times. The lysate was
centrifuged at 14000 rpm at 4ºC and the supernatant removed. The supernatant
containing proteins of interest was either used immediately (see Preparation of
protein samples), or stored at 80 ºC for later use.
6.3.1.2 Protein quantification. Protein concentration was

determined using the BCA
protein assay (Pierce)

as directed. Absorbance was measured at 570nm using a
microplate reader (Molecular Devices).

97
6.3.1.3 Preparation of protein samples. 2x Western loading dye containing 10% β-
mercaptoethanol was added to 20-30 μg of protein and the samples were denatured at
95
o
C for 5 minutes.
6.3.1.4 SDS-polyacrylamide gel electrophoresis. Proteins were separated on a 8-
12% SDS-polyacrylamide gel. The resolving gel (30% acrylamide, 1.5 mM Tris pH
8.8, 10% SDS, ammonium persulfate, Temed) was overlaid with water-saturated
butanol and allowed to polymerize at room temperature for 20 minutes. The water-
saturated butanol was then poured off, the sample loading comb set in place and the
stacking gel (30% acrylamide, 1.0 mM Tris pH 6.8, 10% SDS, ammonium
persulfate, Temed) poured and allowed to polymerize at room temperature for 15
minutes. Cell extracts (30 µg) were separated at 20mA in 1x Western running buffer
(10x: 1.92M glycine, 0.25M Tris, 1% SDS) until sufficiently resolved.
6.3.1.5 Western Blots. Following separation by SDS-polyacrylamide gel
electrophoresis, proteins were transferred to nitrocellulose membranes pre-soaked in
1x transfer buffer (1.92M glycine, 0.25M Tris, 20% methanol). Proteins were
transferred on ice at 100V for an hour.
6.3.1.6 Immunoblotting. Primary mouse monoclonal APE1antibody (Novus
Biologicals) was added in a 1:1000 dilution to 5%m fat-free milk containing 0.1%
TBST. Blots were probed overnight with primary antibody at 4ºC, washed 3 times in
5% fat-free milk/ 0.1% TBST, and followed by probing with secondary antibody
rabbit anti-mouse IgG (Santa Cruz) in the same fat-free milk composition for two

98
hours at room temperature. The membrane was washed 3 times for 5 minutes before
detecting using the Supersignal detection reagent.
6.3.1.7 Detection. The Supersignal immunodetection kit (Pierce) was used in
accordance with the manufacturer’s instructions. The membrane was placed between
two transparencies and exposed to X-ray film which were subsequently developed.

6.3.2 Colony formation assay. HCT116 p53
+/+
parental cells were seeded at the
appropriate cell density for 24 hours in a 6-well plate or 96-well plate. Triplicate
samples were treated with the agent(s) of interest in the appropriate manner for 48
hours. After 48 hours, the media containing treatment was replaced with fresh media
and the cells allowed to proliferate for 72 hours. After 72 hours, the culture media
was removed and the cells were fixed with 3ml of methanol for 5 minutes. Methanol
was then removed and replaced with 3ml of crystal violet stain for 5 minutes.
Background stain was removed by several distilled water washes. The crystal violet
stain was re-absorbed by addition of 1ml 0.2M sodium citrate and allowed to shake
for 20 minutes. 200µl of solution was then removed from each sample in triplicate
and the absorbance measured at 570nm using an ELISA microplate reader
(Molecular devices).
6.3.3 Routine culture of cell lines. Cells were grown in an atmosphere of 5% CO
2

at 37
o
C in a humidified incubator. All cell lines were maintained in their respective
growth medium in T75 tissue culture flasks and were passaged every 4 days.
Following the removal of medium from a pre-confluent flask, cells were rinsed once

99
in 1x phosphate buffered saline (1x PBS) and then 2ml of 1x trypsin was added.
Cells were incubated at 37
o
C until the monolayer had detached. Aliquots of the
trypsinized cells were then removed and transferred to new flasks containing the
appropriate medium and returned to the incubator.
6.3.4 Cell viability assay. Cell viability was assessed by 3-(4, 5-dimethylthiazol-2-
yl)-2,5-diphenyltetrazolium bromide (MTT) analysis (Sigma). Cells were treated
with a range of concentrations of relevant drug(s) for 72-96 hours. 20 μl of MTT dye
(5mg/ml) was then added to each well and the cells incubated at 37
o
C for a further 3
hours. The culture medium was then removed and formazan crystals reabsorbed in
200μl of dimethyl sulfoxide (DMSO). Cell viability was determined by reading the
absorbance of each well at 570nm using an ELISA microplate reader (Molecular
devices).
6.3.5 Flow cytometry
6.3.5.1 Harvesting cells for flow cytometry. Cells that had been incubated with the
appropriate concentration of cytotoxic drug were washed once in 1xPBS before
harvesting in 2ml of 1xPBS/0.5mM EDTA and pelleted by centrifugation at
2,400rpm/4
o
C for 5 minutes. Cell pellets were washed with 1xPBS/1% foetal calf
serum (FCS) and pelleted by centrifugation at 2,500rpm/4
o
C for 5 minutes before
being resuspended in 0.5ml of ice-cold 1xPBS and fixed in 7ml of -80
o
C ethanol.
Fixed cells were then pelleted and washed with 5ml of 1xPBS/1% FCS. Cells were
then re-pelleted and the supernatant decanted. Cells were resuspended in residual
1xPBS/1% FCS and 360 μl of propidium iodide/RNase A solution) was added. Cells

100
were subsequently incubated at 37
o
C for 30 minutes before being analyzed using an
EPICS ELITE flow cytometer (Coulter) equipped with a 15mW Argon laser
(excitation beam 488nm).
6.3.5.2 Flow cytometric analysis. Viable cells were gated on a dot plot display of
forward scatter versus side scatter and cell cycle populations were quantified using
histogram analysis software.
6.3.6 RNAi
6.3.6.1 siRNA oligonucleotides. siRNA oligonucleotides purchased from

Dharmacon (Valencia, CA) were designed according to the literature to target a
portion of the APE1 open reading

frame [(sense, 5'-
GUCUGGUACGACUGGAGUACCUUd(TT); and antisense, 5'-
GGUACUCCAGUCGUACCAGACUUd(TT)].

The oligonucleotide arrived
converted to the 2’-hydroxyl, annealed and desalted. Addition of buffer provided as
part of the package brought the concentration of the solution to 200uM, which was
aliquoted and stored at –20°C.
A second set of siRNA oligonucleotides was ordered from the USC core facility of
the same sequence. The double-stranded siRNAs were generated by mixing the

corresponding pair of sense and antisense oligonucleotides in

sterile DEPC water to
obtain a 20 µM

solution. The reaction mixture was heated to 95°C for 1

min and then
incubated at 37°C for 1 h, aliquoted, and stored

at –20°C.
6.3.6.2 siRNA transfections. HCT116 p53
+/+
cells were seeded at the appropriate
density and allowed to grow for 24 hours in Optimem plus 10% FCS or

101
Lipofectamine RNAimax. For each transfection, 16 μl of transfection reagent
(Invitrogen) per tissue culture plate or 8µl per individual well on a 6-well plate was
mixed with 240 μl or 80 μl of serum-free Optimem (SFO) and incubated at room
temperature for 5 minutes. 6.3 μl of 1µM APE1 siRNA or scrambled control siRNA
per well or 18 μl per p90 was then added to the SFO:oligofectamine/ Lipofectamine
mix, inverted 3 times and incubated at room temperature for a further 30 minutes.
Meanwhile, the cells to be transfected were washed once with SFO and 860 μl of
SFO added per well or 3.2ml per p90. The oligofectamine/ Lipofectamine:siRNA
complex was added dropwise to each well (300 μl) or p90 (900 μl). Four hours post-
transfection, 3x growth medium was added to the cells (2.2ml per p90 and 860µl per
well). After 1 hour, the appropriate concentration of cytotoxic drug was added. The
cells were harvested 72 hours post-treatment and analyzed by flow cytometry or
Western blot analysis.
6.3.7 Transfection of mammalian cells. HCT116 pTet-Off cell lines were
seeded

in a 24-well plate. Transient transfections were performed using

plasmid
DNA normalized to 0.25 µg/µl using Expressfect
TM
(Denville) for HCT116 cells,
according to manufacturer's instructions.

The pTre-Tight:APE1 overexpression
constructs were transfected alone

or with the pTre-Tight:dUTPase expression
construct. All wells were transfected with

0.2 µg of the appropriate construct with
co-transfections

totaling 0.4 µg DNA. Six hours post-transfection,

cells were washed
with PBS and incubated in either fresh media

or media containing a cytotoxic agent
at the appropriate concentration.

After an additional 30 h, cells were washed once

102
with PBS, lysed

and quantified as per western blotting methodology to account

for
variation in cell number. Histograms are presented as fold-change compared to
control as appropriate. The differences

between comparative transfections were
analyzed for statistical

significance using a two-tailed unpaired Student's t-test
(Graphpad).

6.3.8 Cloning
6.3.8.1 PCR amplification of APE1 coding sequence. The APE1 coding sequence
was amplified using cDNA obtained from HCT116 p53
+/+
cells. Primer sequences
for PCR reaction were: APE1 Forward primer: 5’-
CCGGAATTCATGCCGAAGCGTGGAAAAAGG (30) (Tm = ~63ºC). APE1
Reverse primer: 5’- GATCGATATCTCACAGTGCTAGGTATAGGGT (31) (Tm =
~60ºC). Forward primer has EcoR1 restriction site, reverse primer has EcoRV
restriction site incorporated (highlighted). Primers (40 nmole scale) were
reconstituted in 1 mL sterile water. The same primer sequences were used, but
incorporated other restriction sites (eg., BamHI, HindIII) as necessary.
6.3.8.2 Subcloning of APE1 using the TOPO TA expression kit. The subcloning
of APE1 into the TOPO vector was carried out according to the manufacturer’s
protocol (Invitrogen). Briefly, the PCR product underwent the addition of poly-A
tails and was then phenol/chloroform extracted and ethanol precipitated. The PCR
product was then mixed with the TOPO vector and transformed into DH5-alpha

103
E.coli cells. Positive clones (blue-white selection) were then selected for further
analysis.
6.3.8.3 Phenol chloroform extraction and ethanol precipitation of DNA. The
aqueous DNA solution was made up to 200 μl with ddH
2
O and an equal volume of
phenol/chloroform/isoamyl alcohol (25:24:1) was added. The mixture was vortexed
for 1 minute, incubated at room temperature for 2 minutes and centrifuged at 13,000
rpm/4
o
C for 15 minutes. The upper aqueous layer was removed to a fresh eppendorf
and one-tenth the volume of 3M sodium acetate (pH 5.2) (Appendix 1) and 2.5 times
the volume of 100% ethanol were added. The DNA solution was vortexed and stored
at -80
o
C for 30 minutes. The mixture was centrifuged at 13,000rpm for 20 minutes
and the DNA pellet washed once with 70% ethanol before being air-dried and
resuspended in ddH
2
O.
6.3.8.4 Transformation of plasmid DNA. Transformation was carried out
according to manufacturers’ instructions. Briefly, one vial of TOP10/pLysS BL21/
M15 [pRep4]/ DH5alpha competent cells was thawed on ice and mixed with 2 μl of
the appropriate DNA and incubated on ice for 10 minutes. The cells were then heat
shocked at 42
o
C for 30 seconds and placed on ice, before adding 250 μl of LB broth
and incubating at 37
o
C in an orbital incubator for one hour. 200μl of the cell
suspension was then plated onto LB-agar plates containing appropriate antibiotic(s)
and grown overnight at 37
o
C.
6.3.8.5 Cloning of APE1-TOPO product into pTre-Tight empty vector. The
pTre-Tight was digested sequentially with EcoR1 and BamH1, as was the APE1-

104
TOPO vector. Both digestion reactions were electrophoresed on a 1% agarose gel,
followed by phenol extraction and ethanol precipitation. Plasmid DNA was analysed
by restriction digestion using endonuclease enzymes (New England Biolabs) and
standard protocols. In general, 1 unit of enzyme was used to digest 1 μg of DNA in
10μl of 1X enzyme reaction buffer at 37
o
C for 1 hour. Ligation was carried out at
14
o
C overnight, with varying ratios of product:vector to ensure efficient ligation.

6.3.9 Overexpression of APE1. HCT116 pTet-Off cells were seeded on 6 cm
plates and 3 h after

plating the cells were washed with PBS and growth media
containing

10% tet-approved FBS (BD Clontech) added. Cells were transfected

after
24 h with 2 µg pTre-Tight:APE1 for 6 h, washed

in PBS and fresh media added.
Twenty-four hours post-transfection,

media containing the appropriate cytotoxic
agent was added.

Cells were harvested for protein after 48 h incubation and inducible

expression of dUTPase confirmed using Western blotting and enzyme

activity assay.

105
References

1. (2005) GOLD 3.2. The Cambridge Crystallographic Data Centre, Cambridge,
UK

2. Abate C, Patel L, Rauscher FJ, 3rd, Curran T (1990) Redox regulation of fos
and jun DNA-binding activity in vitro. Science 249: 1157-1161

3. Al-Mawsawi LQ, Fikkert V, Dayam R, Witvrouw M, Burke TR, Jr.,
Borchers CH, Neamati N (2006) Discovery of a small-molecule HIV-1
integrase inhibitor-binding site. Proc Natl Acad Sci U S A 103: 10080-10085

4. Al-Mawsawi LQ, Neamati N (2007) Blocking interactions between HIV-1
integrase and cellular cofactors: an emerging anti-retroviral strategy. Trends
Pharmacol Sci 28: 526-535

5. Al-Mawsawi LQ, Christ F, Dayam R, Debyser Z, Neamati N (2008)
Inhibitory profile of a LEDGF/p75 peptide against HIV-1 integrase: Insight
into integrase-DNA complex formation and catalysis. FEBS Lett

6. Al-Mawsawi LQ, Hombrouck A, Dayam R, Debyser Z, Neamati N (2008)
Four-tiered pi interaction at the dimeric interface of HIV-1 integrase critical
for DNA integration and viral infectivity. Virology 377: 355-363

7. Asante-Appiah E, Skalka AM (1997) Molecular mechanisms in retrovirus
DNA integration. Antiviral Research 36: 139-156

8. Asante-Appiah E, Skalka AM (1999) HIV-1 integrase: structural
organization, conformational changes, and catalysis. Adv Virus Res 52: 351-
369

9. Atamna H, Cheung I, Ames BN (2000) A method for detecting abasic sites in
living cells: age-dependent changes in base excision repair. Proc Natl Acad
Sci U S A 97: 686-691

10. Bases RE, Mendez F (1997) Topoisomerase inhibition by lucanthone, an
adjuvant in radiation therapy. Int J Radiat Oncol Biol Phys 37: 1133-1137

11. Belzile JP, Choudhury SA, Cournoyer D, Chow TY (2006) Targeting DNA
repair proteins: a promising avenue for cancer gene therapy. Curr Gene Ther
6: 111-123

106
12. Bernal F, Tyler AF, Korsmeyer SJ, Walensky LD, Verdine GL (2007)
Reactivation of the p53 tumor suppressor pathway by a stapled p53 peptide. J
Am Chem Soc 129: 2456-2457

13. Berquist BR, McNeill DR, Wilson DM, 3rd (2008) Characterization of abasic
endonuclease activity of human Ape1 on alternative substrates, as well as
effects of ATP and sequence context on AP site incision. J Mol Biol 379: 17-
27

14. Bhakat KK, Yang SH, Mitra S (2003) Acetylation of human AP-
endonuclease 1, a critical enzyme in DNA repair and transcription regulation.
Methods Enzymol 371: 292-300

15. Bobola MS, Blank A, Berger MS, Stevens BA, Silber JR (2001)
Apurinic/apyrimidinic endonuclease activity is elevated in human adult
gliomas. Clin Cancer Res 7: 3510-3518

16. Brass AL, Dykxhoorn DM, Benita Y, Yan N, Engelman A, Xavier RJ,
Lieberman J, Elledge SJ (2008) Identification of host proteins required for
HIV infection through a functional genomic screen. Science 319: 921-926

17. Brown PO (1999) Integration. Cold Spring Harbor Press, Cold Spring
Harbor

18. Brown PO (1999) Integration. In Retroviruses (Editors: Coffin, J. C.; Hughes,
S. H.; Varmus, H. E.): Cold Spring Harbor Press

19. Bujacz G, Alexandratos J, Qing ZL, Clement-Mella C, Wlodawer A (1996)
The catalytic domain of human immunodeficiency virus integrase: ordered
active site in the F185H mutant. FEBS Lett 398: 175-178

20. Bushman FD, Engelman A, Palmer I, Wingfield P, Craigie R (1993)
Domains of the integrase protein of human immunodeficiency virus type 1
responsible for polynucleotidyl transfer and zinc binding. Proc Natl Acad Sci
U S A 90: 3428-3432

21. Chang-Claude J, Popanda O, Tan XL, Kropp S, Helmbold I, von Fournier D,
Haase W, Sautter-Bihl ML, Wenz F, Schmezer P, Ambrosone CB (2005)
Association between polymorphisms in the DNA repair genes, XRCC1,
APE1, and XPD and acute side effects of radiotherapy in breast cancer
patients. Clin Cancer Res 11: 4802-4809

107
22. Chen DS, Herman T, Demple B (1991) Two distinct human DNA diesterases
that hydrolyze 3'-blocking deoxyribose fragments from oxidized DNA.
Nucleic Acids Res 19: 5907-5914

23. Chen DS, Olkowski ZL (1994) Biological responses of human apurinic
endonuclease to radiation-induced DNA damage. Ann N Y Acad Sci 726:
306-308

24. Chen H, Engelman A (1998) The barrier-to-autointegration protein is a host
factor for HIV type 1 integration. Proc Natl Acad Sci U S A 95: 15270-15274

25. Chen JC, Krucinski J, Miercke LJ, Finer-Moore JS, Tang AH, Leavitt AD,
Stroud RM (2000) Crystal structure of the HIV-1 integrase catalytic core and
C-terminal domains: a model for viral DNA binding. Proc Natl Acad Sci U S
A 97: 8233-8238

26. Chorev M, Goodman M (1995) Recent developments in retro peptides and
proteins--an ongoing topochemical exploration. Trends Biotechnol 13: 438-
445

27. Clapper ML (1998) Chemopreventive activity of oltipraz. Pharmacol Ther
78: 17-27

28. Costantino G, Maltoni K, Marinozzi M, Camaioni E, Prezeau L, Pin JP,
Pellicciari R (2001) Synthesis and biological evaluation of 2-(3'-(1H-tetrazol-
5-yl) bicyclo[1.1.1]pent-1-yl)glycine (S-TBPG), a novel mGlu1 receptor
antagonist. Bioorg Med Chem 9: 221-227

29. Dayam R, Neamati N (2003) Small-molecule HIV-1 integrase inhibitors: the
2001-2002 update. Curr Pharm Des 9: 1789-1802

30. Dayam R, Deng J, Neamati N (2006) HIV-1 integrase inhibitors: 2003-2004
update. Med Res Rev 26: 271-309

31. Dayam R, Gundla R, Al-Mawsawi LQ, Neamati N (2008) HIV-1 integrase
inhibitors: 2005-2006 update. Med Res Rev 28: 118-154

32. de Soultrait VR, Desjobert C, Tarrago-Litvak L (2003) Peptides as new
inhibitors of HIV-1 reverse transcriptase and integrase. Curr Med Chem 10:
1765-1778

108
33. de Soultrait VR, Desjobert C, Tarrago-Litvak L (2003) Curr Med Chem 10:
1765-1778

34. Demple B, Harrison L (1994) Repair of oxidative damage to DNA:
enzymology and biology. Annu Rev Biochem 63: 915-948

35. Demple B, Harrison L, Wilson DM, 3rd, Bennett RA, Takagi T, Ascione AG
(1997) Regulation of eukaryotic abasic endonucleases and their role in
genetic stability. Environ Health Perspect 105 Suppl 4: 931-934

36. Desjobert C, de Soultrait VR, Faure A, Parissi V, Litvak S, Tarrago-Litvak L,
Fournier M (2004) Identification by phage display selection of a short peptide
able to inhibit only the strand transfer reaction catalyzed by human
immunodeficiency virus type 1 integrase. Biochemistry 43: 13097-13105

37. Dlakic M (2000) Functionally unrelated signalling proteins contain a fold
similar to Mg2+-dependent endonucleases. Trends Biochem Sci 25: 272-273

38. Dyda F, Hickman AB, Jenkins TM, Engelman A, Craigie R, Davies DR
(1994) Crystal structure of the catalytic domain of HIV-1 integrase: similarity
to other polynucleotidyl transferases. Science 266: 1981-1986

39. Engelman A, Mizuuchi K, Craigie R (1991) HIV-1 DNA integration:
mechanism of viral DNA cleavage and DNA strand transfer. Cell 67: 1211-
1221

40. Engelman A, Craigie R (1992) Identification of conserved amino acid
residues critical for human immunodeficiency virus type 1 integrase function
in vitro. J Virol 66: 6361-6369

41. Engelman A (1999) In vivo analysis of retroviral integrase structure and
function. Adv Virus Res 52: 411-426

42. Engelman A, Cherepanov P (2008) The lentiviral integrase binding protein
LEDGF/p75 and HIV-1 replication. PLoS Pathog 4: e1000046

43. Espeseth AS, Felock P, Wolfe A, Witmer M, Grobler J, Anthony N,
Egbertson M, Melamed JY, Young S, Hamill T, Cole JL, Hazuda DJ (2000)
HIV-1 integrase inhibitors that compete with the target DNA substrate define
a unique strand transfer conformation for integrase. Proc Natl Acad Sci U S
A 97: 11244-11249

109
44. Esposito D, Craigie R (1999) HIV integrase structure and function. Adv
Virus Res 52: 319-333

45. Evans AR, Limp-Foster M, Kelley MR (2000) Going APE over ref-1. Mutat
Res 461: 83-108

46. Farnet CM, Bushman FD (1997) HIV-1 cDNA integration: requirement of
HMG I(Y) protein for function of preintegration complexes in vitro. Cell 88:
483-492

47. Fawell S, Seery J, Daikh Y, Moore C, Chen LL, Pepinsky B, Barsoum J
(1994) Tat-mediated delivery of heterologous proteins into cells. Proc Natl
Acad Sci U S A 91: 664-668

48. Ferns RB, Partridge JC, Tisdale M, Hunt N, Tedder RS (1991) Monoclonal
antibodies define linear and conformational epitopes of HIV-1 pol gene
products. AIDS Res Hum Retroviruses 7: 307-313

49. Fishel ML, He Y, Smith ML, Kelley MR (2007) Manipulation of base
excision repair to sensitize ovarian cancer cells to alkylating agent
temozolomide. Clin Cancer Res 13: 260-267

50. Fishel ML, Kelley MR (2007) The DNA base excision repair protein
Ape1/Ref-1 as a therapeutic and chemopreventive target. Mol Aspects Med
28: 375-395

51. Fletcher MD, Campbell MM (1998) Partially Modified Retro-Inverso
Peptides: Development, Synthesis, and Conformational Behavior. Chem Rev
98: 763-796

52. Foged C, Nielsen HM (2008) Cell-penetrating peptides for drug delivery
across membrane barriers. Expert Opin Drug Deliv 5: 105-117

53. Forood B, Feliciano EJ, Nambiar KP (1993) Stabilization of alpha-helical
structures in short peptides via end capping. Proc Natl Acad Sci U S A 90:
838-842

54. Fung H, Demple B (2005) A vital role for Ape1/Ref1 protein in repairing
spontaneous DNA damage in human cells. Mol Cell 17: 463-470

55. Garcia-Marcos L, Schuster A (2001) Antileukotrienes in asthma: present
situation. Expert Opin Pharmacother 2: 441-466

110
56. Gorman MA, Morera S, Rothwell DG, de La Fortelle E, Mol CD, Tainer JA,
Hickson ID, Freemont PS (1997) The crystal structure of the human DNA
repair endonuclease HAP1 suggests the recognition of extra-helical
deoxyribose at DNA abasic sites. Embo J 16: 6548-6558

57. Griffin RJ, Fontana G, Golding BT, Guiard S, Hardcastle IR, Leahy JJ,
Martin N, Richardson C, Rigoreau L, Stockley M, Smith GC (2005) Selective
benzopyranone and pyrimido[2,1-a]isoquinolin-4-one inhibitors of DNA-
dependent protein kinase: synthesis, structure-activity studies, and
radiosensitization of a human tumor cell line in vitro. J Med Chem 48: 569-
585

58. Hadi MZ, Wilson DM, 3rd (2000) Second human protein with homology to
the Escherichia coli abasic endonuclease exonuclease III. Environ Mol
Mutagen 36: 312-324

59. Hansch C, Hoekman D, Leo A, Zhang L, Li P (1995) The expanding role of
quantitative structure-activity relationships (QSAR) in toxicology. Toxicol
Lett 79: 45-53

60. Hardcastle IR, Cockcroft X, Curtin NJ, El-Murr MD, Leahy JJ, Stockley M,
Golding BT, Rigoreau L, Richardson C, Smith GC, Griffin RJ (2005)
Discovery of potent chromen-4-one inhibitors of the DNA-dependent protein
kinase (DNA-PK) using a small-molecule library approach. J Med Chem 48:
7829-7846

61. Hazuda DJ, Felock P, Witmer M, Wolfe A, Stillmock K, Grobler JA,
Espeseth A, Gabryelski L, Schleif W, Blau C, Miller MD (2000) Inhibitors of
strand transfer that prevent integration and inhibit HIV-1 replication in cells.
Science 287: 646-650

62. Hehl EA, Joshi P, Kalpana GV, Prasad VR (2004) Interaction between
human immunodeficiency virus type 1 reverse transcriptase and integrase
proteins. J Virol 78: 5056-5067

63. Herring CJ, West CM, Wilks DP, Davidson SE, Hunter RD, Berry P, Forster
G, MacKinnon J, Rafferty JA, Elder RH, Hendry JH, Margison GP (1998)
Levels of the DNA repair enzyme human apurinic/apyrimidinic endonuclease
(APE1, APEX, Ref-1) are associated with the intrinsic radiosensitivity of
cervical cancers. Br J Cancer 78: 1128-1133

111
64. Hill JW, Hazra TK, Izumi T, Mitra S (2001) Stimulation of human 8-
oxoguanine-DNA glycosylase by AP-endonuclease: potential coordination of
the initial steps in base excision repair. Nucleic Acids Res 29: 430-438

65. Hoeijmakers JH (2001) Genome maintenance mechanisms for preventing
cancer. Nature 411: 366-374

66. Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J, Heim
W, Berlin J, Baron A, Griffing S, Holmgren E, Ferrara N, Fyfe G, Rogers B,
Ross R, Kabbinavar F (2004) Bevacizumab plus irinotecan, fluorouracil, and
leucovorin for metastatic colorectal cancer. N Engl J Med 350: 2335-2342

67. Ito H, Matsuo K, Hamajima N, Mitsudomi T, Sugiura T, Saito T, Yasue T,
Lee KM, Kang D, Yoo KY, Sato S, Ueda R, Tajima K (2004) Gene-
environment interactions between the smoking habit and polymorphisms in
the DNA repair genes, APE1 Asp148Glu and XRCC1 Arg399Gln, in
Japanese lung cancer risk. Carcinogenesis 25: 1395-1401

68. Jayaraman G, Kumar TK, Arunkumar AI, Yu C (1996) 2,2,2-
Trifluoroethanol induces helical conformation in an all beta-sheet protein.
Biochem Biophys Res Commun 222: 33-37

69. Kakolyris S, Kaklamanis L, Engels K, Turley H, Hickson ID, Gatter KC,
Harris AL (1997) Human apurinic endonuclease 1 expression in a colorectal
adenoma-carcinoma sequence. Cancer Res 57: 1794-1797

70. Kakolyris S, Kaklamanis L, Engels K, Fox SB, Taylor M, Hickson ID, Gatter
KC, Harris AL (1998) Human AP endonuclease 1 (HAP1) protein expression
in breast cancer correlates with lymph node status and angiogenesis. Br J
Cancer 77: 1169-1173

71. Kakolyris S, Giatromanolaki A, Koukourakis M, Kaklamanis L, Kanavaros
P, Hickson ID, Barzilay G, Georgoulias V, Gatter KC, Harris AL (1999)
Nuclear localization of human AP endonuclease 1 (HAP1/Ref-1) associates
with prognosis in early operable non-small cell lung cancer (NSCLC). J
Pathol 189: 351-357

72. Katz RA, Skalka AM (1994) The retroviral enzymes. Annu Rev Biochem 63:
133-173

112
73. Kenny MK, Mendez F, Sandigursky M, Kureekattil RP, Goldman JD,
Franklin WA, Bases R (2001) Heat shock protein 70 binds to human
apurinic/apyrimidinic endonuclease and stimulates endonuclease activity at
abasic sites. J Biol Chem 276: 9532-9536

74. Kilby JM, Hopkins S, Venetta TM, DiMassimo B, Cloud GA, Lee JY,
Alldredge L, Hunter E, Lambert D, Bolognesi D, Matthews T, Johnson MR,
Nowak MA, Shaw GM, Saag MS (1998) Potent suppression of HIV-1
replication in humans by T-20, a peptide inhibitor of gp41-mediated virus
entry. Nat Med 4: 1302-1307

75. Kiyohara C, Takayama K, Nakanishi Y (2006) Association of genetic
polymorphisms in the base excision repair pathway with lung cancer risk: a
meta-analysis. Lung Cancer 54: 267-283

76. Klutch M, Woerner AM, Marcus-Sekura CJ, Levin JG (1998) Generation of
HIV-1/HIV-2 cross-reactive peptide antisera by small sequence changes in
HIV-1 reverse transcriptase and integrase immunizing peptides. J Biomed Sci
5: 192-202

77. Koukourakis MI, Giatromanolaki A, Kakolyris S, Sivridis E, Georgoulias V,
Funtzilas G, Hickson ID, Gatter KC, Harris AL (2001) Nuclear expression of
human apurinic/apyrimidinic endonuclease (HAP1/Ref-1) in head-and-neck
cancer is associated with resistance to chemoradiotherapy and poor outcome.
Int J Radiat Oncol Biol Phys 50: 27-36

78. Krajewski K, Long YQ, Marchand C, Pommier Y, Roller PP (2003) Design
and synthesis of dimeric HIV-1 integrase inhibitory peptides. Bioorg Med
Chem Lett 13: 3203-3205

79. Krajewski K, Marchand C, Long YQ, Pommier Y, Roller PP (2004)
Synthesis and HIV-1 integrase inhibitory activity of dimeric and tetrameric
analogs of indolicidin. Bioorg Med Chem Lett 14: 5595-5598

80. Latham PW (1999) Therapeutic peptides revisited. Nat Biotechnol 17: 755-
757

81. Lau JP, Weatherdon KL, Skalski V, Hedley DW (2004) Effects of
gemcitabine on APE/ref-1 endonuclease activity in pancreatic cancer cells,
and the therapeutic potential of antisense oligonucleotides. Br J Cancer 91:
1166-1173

113
82. Leavitt AD, Shiue L, Varmus HE (1993) Site-directed mutagenesis of HIV-1
integrase demonstrates differential effects on integrase functions in vitro. J
Biol Chem 268: 2113-2119

83. Li C, Liu Z, Wang LE, Strom SS, Lee JE, Gershenwald JE, Ross MI,
Mansfield PF, Cormier JN, Prieto VG, Duvic M, Grimm EA, Wei Q (2006)
Genetic variants of the ADPRT, XRCC1 and APE1 genes and risk of
cutaneous melanoma. Carcinogenesis 27: 1894-1901

84. Li HY, Zawahir Z, Song LD, Long YQ, Neamati N (2006) Sequence-based
design and discovery of peptide inhibitors of HIV-1 integrase: insight into the
binding mode of the enzyme. J Med Chem 49: 4477-4486

85. Liljebris C, Larsen SD, Ogg D, Palazuk BJ, Bleasdale JE (2002)
Investigation of potential bioisosteric replacements for the carboxyl groups of
peptidomimetic inhibitors of protein tyrosine phosphatase 1B: identification
of a tetrazole-containing inhibitor with cellular activity. J Med Chem 45:
1785-1798

86. Liu L, Taverna P, Whitacre CM, Chatterjee S, Gerson SL (1999)
Pharmacologic disruption of base excision repair sensitizes mismatch repair-
deficient and -proficient colon cancer cells to methylating agents. Clin
Cancer Res 5: 2908-2917

87. Liu L, Gerson SL (2004) Therapeutic impact of methoxyamine: blocking
repair of abasic sites in the base excision repair pathway. Curr Opin Investig
Drugs 5: 623-627

88. Liuzzi M, Talpaert-Borle M (1985) A new approach to the study of the base-
excision repair pathway using methoxyamine. J Biol Chem 260: 5252-5258
89. Longley DB, Harkin DP, Johnston PG (2003) 5-fluorouracil: mechanisms of
action and clinical strategies. Nat Rev Cancer 3: 330-338

90. Luo M, Kelley MR (2004) Inhibition of the human apurinic/apyrimidinic
endonuclease (APE1) repair activity and sensitization of breast cancer cells to
DNA alkylating agents with lucanthone. Anticancer Res 24: 2127-2134

91. Madhusudan S, Smart F, Shrimpton P, Parsons JL, Gardiner L, Houlbrook S,
Talbot DC, Hammonds T, Freemont PA, Sternberg MJ, Dianov GL, Hickson
ID (2005) Isolation of a small molecule inhibitor of DNA base excision
repair. Nucleic Acids Res 33: 4711-4724

114
92. Maertens G, Cherepanov P, Pluymers W, Busschots K, De Clercq E, Debyser
Z, Engelborghs Y (2003) LEDGF/p75 is essential for nuclear and
chromosomal targeting of HIV-1 integrase in human cells. J Biol Chem 278:
33528-33539

93. Maher RL, Bloom LB (2007) Pre-steady-state kinetic characterization of the
AP endonuclease activity of human AP endonuclease 1. J Biol Chem 282:
30577-30585

94. Marenstein DR, Wilson DM, 3rd, Teebor GW (2004) Human AP
endonuclease (APE1) demonstrates endonucleolytic activity against AP sites
in single-stranded DNA. DNA Repair (Amst) 3: 527-533

95. Maroun RG, Krebs D, El Antri S, Deroussent A, Lescot E, Troalen F,
Porumb H, Goldberg ME, Fermandjian S (1999) Self-association and
domains of interactions of an amphipathic helix peptide inhibitor of HIV-1
integrase assessed by analytical ultracentrifugation and NMR experiments in
trifluoroethanol/H(2)O mixtures. J Biol Chem 274: 34174-34185

96. Maroun RG, Krebs D, Roshani M, Porumb H, Auclair C, Troalen F,
Fermandjian S (1999) Conformational aspects of HIV-1 integrase inhibition
by a peptide derived from the enzyme central domain and by antibodies
raised against this peptide. Eur J Biochem 260: 145-155

97. Maroun RG, Gayet S, Benleulmi MS, Porumb H, Zargarian L, Merad H, Leh
H, Mouscadet JF, Troalen F, Fermandjian S (2001) Peptide inhibitors of
HIV-1 integrase dissociate the enzyme oligomers. Biochemistry 40: 13840-
13848

98. Mazumder A, Neamati N, Pilon AA, Sunder S, Pommier Y (1996) Chemical
trapping of ternary complexes of human immunodeficiency virus type 1
integrase, divalent metal, and DNA substrates containing an abasic site.
Implications for the role of lysine 136 in DNA binding. J Biol Chem 271:
27330-27338

99. McGovern SL, Caselli E, Grigorieff N, Shoichet BK (2002) A common
mechanism underlying promiscuous inhibitors from virtual and high-
throughput screening. J Med Chem 45: 1712-1722

100. McNeill DR, Wilson DM, 3rd (2007) A dominant-negative form of the major
human abasic endonuclease enhances cellular sensitivity to laboratory and
clinical DNA-damaging agents. Mol Cancer Res 5: 61-70

115
101. Miller MD, Farnet CM, Bushman FD (1997) Human immunodeficiency virus
type 1 preintegration complexes: studies of organization and composition. J
Virol 71: 5382-5390

102. Mitra S, Hazra TK, Roy R, Ikeda S, Biswas T, Lock J, Boldogh I, Izumi T
(1997) Complexities of DNA base excision repair in mammalian cells. Mol
Cells 7: 305-312

103. Mol CD, Izumi T, Mitra S, Tainer JA (2000) DNA-bound structures and
mutants reveal abasic DNA binding by APE1 and DNA repair coordination
[corrected]. Nature 403: 451-456

104. Moore DH, Michael H, Tritt R, Parsons SH, Kelley MR (2000) Alterations in
the expression of the DNA repair/redox enzyme APE/ref-1 in epithelial
ovarian cancers. Clin Cancer Res 6: 602-609

105. Mulder LC, Chakrabarti LA, Muesing MA (2002) Interaction of HIV-1
integrase with DNA repair protein hRad18. J Biol Chem 277: 27489-27493

106. Neamati N, Mazumder A, Sunder S, Owen JM, Tandon M, Lown JW,
Pommier Y (1998) Highly potent synthetic polyamides, bisdistamycins, and
lexitropsins as inhibitors of human immunodeficiency virus type 1 integrase.
Mol Pharmacol 54: 280-290

107. Neamati N, Marchand C, Pommier Y (2000) HIV-1 integrase inhibitors: past,
present, and future. Adv Pharmacol 49: 147-165

108. Neamati N (2001) Structure-based HIV-1 integrase inhibitor design: a future
perspective. Expert Opin Invest Drugs 10: 281-296.

109. Nilsen BM, Haugan IR, Berg K, Olsen L, Brown PO, Helland DE (1996)
Monoclonal antibodies against human immunodeficiency virus type 1
integrase: epitope mapping and differential effects on integrase activities in
vitro. J Virol 70: 1580-1587

110. Oz-Gleenberg I, Avidan O, Goldgur Y, Herschhorn A, Hizi A (2005)
Peptides derived from the reverse transcriptase of human immunodeficiency
virus type 1 as novel inhibitors of the viral integrase. J Biol Chem

111. Oz I, Avidan O, Hizi A (2002) Inhibition of the integrases of human
immunodeficiency viruses type 1 and type 2 by reverse transcriptases.
Biochem J 361: 557-566

116
112. Parikh SS, Mol CD, Slupphaug G, Bharati S, Krokan HE, Tainer JA (1998)
Base excision repair initiation revealed by crystal structures and binding
kinetics of human uracil-DNA glycosylase with DNA. EMBO J 17: 5214-
5226

113. Parker A, Gu Y, Mahoney W, Lee SH, Singh KK, Lu AL (2001) Human
homolog of the MutY repair protein (hMYH) physically interacts with
proteins involved in long patch DNA base excision repair. J Biol Chem 276:
5547-5555

114. Parrill AL (2003) HIV-1 integrase inhibition: binding sites, structure activity
relationships and future perspectives. Curr Med Chem 10: 1811-1824

115. Patani GA, LaVoie EJ (1996) Bioisosterism: A Rational Approach in Drug
Design. Chem Rev 96: 3147-3176

116. Perrier L, Bourrie M, Marti E, Tronquet C, Masse D, Berger Y, Magdalou J,
Fabre G (1994) In vitro N-glucuronidation of SB 47436 (BMS 186295), a
new AT1 nonpeptide angiotensin II receptor antagonist, by rat, monkey and
human hepatic microsomal fractions. J Pharmacol Exp Ther 271: 91-99

117. Pieretti M, Khattar NH, Smith SA (2001) Common polymorphisms and
somatic mutations in human base excision repair genes in ovarian and
endometrial cancers. Mutat Res 432: 53-59

118. Pommier Y, Johnson AA, Marchand C (2005) Integrase inhibitors to treat
HIV/AIDS. Nat Rev Drug Discov 4: 236-248

119. Puglisi F, Aprile G, Minisini AM, Barbone F, Cataldi P, Tell G, Kelley MR,
Damante G, Beltrami CA, Di Loreto C (2001) Prognostic significance of
Ape1/ref-1 subcellular localization in non-small cell lung carcinomas.
Anticancer Res 21: 4041-4049

120. Puras Lutzke RA, Eppens NA, Weber PA, Houghten RA, Plasterk RH (1995)
Identification of a hexapeptide inhibitor of the human immunodeficiency
virus integrase protein by using a combinatorial chemical library. Proc Natl
Acad Sci U S A 92: 11456-11460

121. Ramotar D (1997) The apurinic-apyrimidinic endonuclease IV family of
DNA repair enzymes. Biochem Cell Biol 75: 327-336

117
122. Restle T, Pawlita M, Sczakiel G, Muller B, Goody RS (1992) Structure-
function relationships of HIV-1 reverse transcriptase determined using
monoclonal antibodies. J Biol Chem 267: 14654-14661

123. Rinne M, Caldwell D, Kelley MR (2004) Transient adenoviral N-
methylpurine DNA glycosylase overexpression imparts chemotherapeutic
sensitivity to human breast cancer cells. Mol Cancer Ther 3: 955-967

124. Roberts NA, Martin JA, Kinchington D, Broadhurst AV, Craig JC, Duncan
IB, Galpin SA, Handa BK, Kay J, Krohn A, et al. (1990) Rational design of
peptide-based HIV proteinase inhibitors. Science 248: 358-361

125. Robertson KA, Hill DP, Xu Y, Liu L, Van Epps S, Hockenbery DM, Park JR,
Wilson TM, Kelley MR (1997) Down-regulation of apurinic/apyrimidinic
endonuclease expression is associated with the induction of apoptosis in
differentiating myeloid leukemia cells. Cell Growth Differ 8: 443-449

126. Ruggeri BA, Robinson C, Angeles T, Wilkinson Jt, Clapper ML (2002) The
chemopreventive agent oltipraz possesses potent antiangiogenic activity in
vitro, ex vivo, and in vivo and inhibits tumor xenograft growth. Clin Cancer
Res 8: 267-274

127. Sakurai K, Chung HS, Kahne D (2004) Use of a retroinverso p53 peptide as
an inhibitor of MDM2. J Am Chem Soc 126: 16288-16289

128. Sattler U, Frit P, Salles B, Calsou P (2003) Long-patch DNA repair synthesis
during base excision repair in mammalian cells. EMBO Rep 4: 363-367

129. Seiple LA, Cardellina JH, 2nd, Akee R, Stivers JT (2008) Potent inhibition of
human apurinic/apyrimidinic endonuclease 1 by arylstibonic acids. Mol
Pharmacol 73: 669-677

130. Seo YR, Kelley MR, Smith ML (2002) Selenomethionine regulation of p53
by a ref1-dependent redox mechanism. Proc Natl Acad Sci U S A 99: 14548-
14553

131. Seo YR, Sweeney C, Smith ML (2002) Selenomethionine induction of DNA
repair response in human fibroblasts. Oncogene 21: 3663-3669

132. She M, Pan I, Sun L, Yeung SC (2005) Enhancement of manumycin A-
induced apoptosis by methoxyamine in myeloid leukemia cells. Leukemia
19: 595-602

118

133. Singh SB, Herath K, Guan Z, Zink DL, Dombrowski AW, Polishook JD,
Silverman KC, Lingham RB, Felock PJ, Hazuda DJ (2002) Integramides A
and B, two novel non-ribosomal linear peptides containing nine C(alpha)-
methyl amino acids produced by fungal fermentations that are inhibitors of
HIV-1 integrase. Org Lett 4: 1431-1434

134. Sobol RW, Foley JF, Nyska A, Davidson MG, Wilson SH (2003) Regulated
over-expression of DNA polymerase beta mediates early onset cataract in
mice. DNA Repair (Amst) 2: 609-622

135. Sourgen F, Maroun RG, Frere V, Bouziane M, Auclair C, Troalen F,
Fermandjian S (1996) A synthetic peptide from the human immunodeficiency
virus type-1 integrase exhibits coiled-coil properties and interferes with the in
vitro integration activity of the enzyme. Correlated biochemical and
spectroscopic results. Eur J Biochem 240: 765-773

136. Sticht J, Humbert M, Findlow S, Bodem J, Muller B, Dietrich U, Werner J,
Krausslich HG (2005) A peptide inhibitor of HIV-1 assembly in vitro. Nat
Struct Mol Biol 12: 671-677

137. Suh D, Wilson DM, 3rd, Povirk LF (1997) 3'-phosphodiesterase activity of
human apurinic/apyrimidinic endonuclease at DNA double-strand break
ends. Nucleic Acids Res 25: 2495-2500

138. Tasara T, Maga G, Hottiger MO, Hubscher U (2001) HIV-1 reverse
transcriptase and integrase enzymes physically interact and inhibit each other.
FEBS Lett 507: 39-44.

139. Tasara T, Maga G, Hottiger MO, Hubscher U (2001) HIV-1 reverse
transcriptase and integrase enzymes physically interact and inhibit each other.
FEBS Lett 507: 39-44

140. Taverna P, Liu L, Hwang HS, Hanson AJ, Kinsella TJ, Gerson SL (2001)
Methoxyamine potentiates DNA single strand breaks and double strand
breaks induced by temozolomide in colon cancer cells. Mutat Res 485: 269-
281

141. Taylor AF, Weiss B (1982) Role of exonuclease III in the base excision
repair of uracil-containing DNA. J Bacteriol 151: 351-357

119
142. Van Maele B, Debyser Z (2005) HIV-1 integration: an interplay between
HIV-1 integrase, cellular and viral proteins. AIDS Rev 7: 26-43

143. Varmus H (1988) Retroviruses. Science 240: 1427-1435

144. Verdine GL, Walensky LD (2007) The challenge of drugging undruggable
targets in cancer: lessons learned from targeting BCL-2 family members. Clin
Cancer Res 13: 7264-7270

145. Verdonk ML, Cole JC, Hartshorn MJ, Murray CW, Taylor RD (2003)
Improved protein-ligand docking using GOLD. Proteins 52: 609-623

146. Verdonk ML, Chessari G, Cole JC, Hartshorn MJ, Murray CW, Nissink JW,
Taylor RD, Taylor R (2005) Modeling water molecules in protein-ligand
docking using GOLD. J Med Chem 48: 6504-6515

147. Vidal AE, Boiteux S, Hickson ID, Radicella JP (2001) XRCC1 coordinates
the initial and late stages of DNA abasic site repair through protein-protein
interactions. EMBO J 20: 6530-6539

148. Vinga-Martins C, Schneider T, Werno A, Roenspeck W, Pauli G, Mueller-
Lantzsch N (1992) Mapping of immunodominant epitopes of the HIV-1 and
HIV-2 integrase proteins by recombinant proteins and synthetic peptides.
AIDS Res Hum Retroviruses 8: 1301-1310

149. Walensky LD, Kung AL, Escher I, Malia TJ, Barbuto S, Wright RD, Wagner
G, Verdine GL, Korsmeyer SJ (2004) Activation of apoptosis in vivo by a
hydrocarbon-stapled BH3 helix. Science 305: 1466-1470

150. Walensky LD, Pitter K, Morash J, Oh KJ, Barbuto S, Fisher J, Smith E,
Verdine GL, Korsmeyer SJ (2006) A stapled BID BH3 helix directly binds
and activates BAX. Mol Cell 24: 199-210

151. Walker LJ, Craig RB, Harris AL, Hickson ID (1994) A role for the human
DNA repair enzyme HAP1 in cellular protection against DNA damaging
agents and hypoxic stress. Nucleic Acids Res 22: 4884-4889

152. Wang JY, Ling H, Yang W, Craigie R (2001) Structure of a two-domain
fragment of HIV-1 integrase: implications for domain organization in the
intact protein. EMBO J 20: 7333-7343

120
153. Wang JY, Ling H, Yang W, Craigie R (2001) Structure of a two-domain
fragment of HIV-1 integrase: implications for domain organization in the
intact protein. EMBO J 20: 7333-7343.

154. Waters TR, Gallinari P, Jiricny J, Swann PF (1999) Human thymine DNA
glycosylase binds to apurinic sites in DNA but is displaced by human
apurinic endonuclease 1. J Biol Chem 274: 67-74

155. Weiss B, Grossman L (1987) Phosphodiesterases involved in DNA repair.
Adv Enzymol Relat Areas Mol Biol 60: 1-34

156. Wild CT, Shugars DC, Greenwell TK, McDanal CB, Matthews TJ (1994)
Peptides corresponding to a predictive alpha-helical domain of human
immunodeficiency virus type 1 gp41 are potent inhibitors of virus infection.
Proc Natl Acad Sci U S A 91: 9770-9774

157. Willetts KE, Rey F, Agostini I, Navarro JM, Baudat Y, Vigne R, Sire J
(1999) DNA repair enzyme uracil DNA glycosylase is specifically
incorporated into human immunodeficiency virus type 1 viral particles
through a Vpr-independent mechanism. J Virol 73: 1682-1688

158. Williams IG (2003) Enfuvirtide (Fuzeon): the first fusion inhibitor. Int J Clin
Pract 57: 890-897

159. Wilson DM, 3rd, Barsky D (2001) The major human abasic endonuclease:
formation, consequences and repair of abasic lesions in DNA. Mutat Res 485:
283-307

160. Wilson DM, 3rd (2005) Ape1 abasic endonuclease activity is regulated by
magnesium and potassium concentrations and is robust on alternative DNA
structures. J Mol Biol 345: 1003-1014

161. Winters TA, Henner WD, Russell PS, McCullough A, Jorgensen TJ (1994)
Removal of 3'-phosphoglycolate from DNA strand-break damage in an
oligonucleotide substrate by recombinant human apurinic/apyrimidinic
endonuclease 1. Nucleic Acids Res 22: 1866-1873

162. Woon EC, Threadgill MD (2005) Poly(ADP-ribose)polymerase inhibition -
where now? Curr Med Chem 12: 2373-2392

121
163. Wu T, Guo J, Bess J, Henderson LE, Levin JG (1999) Molecular
requirements for human immunodeficiency virus type 1 plus-strand transfer:
analysis in reconstituted and endogenous reverse transcription systems. J
Virol 73: 4794-4805

164. Wu X, Liu H, Xiao H, Conway JA, Hehl E, Kalpana GV, Prasad V, Kappes
JC (1999) Human immunodeficiency virus type 1 integrase protein promotes
reverse transcription through specific interactions with the nucleoprotein
reverse transcription complex. J Virol 73: 2126-2135

165. Xanthoudakis S, Miao G, Wang F, Pan YC, Curran T (1992) Redox
activation of Fos-Jun DNA binding activity is mediated by a DNA repair
enzyme. Embo J 11: 3323-3335

166. Xanthoudakis S, Miao GG, Curran T (1994) The redox and DNA-repair
activities of Ref-1 are encoded by nonoverlapping domains. Proc Natl Acad
Sci U S A 91: 23-27

167. Xu Y, Moore DH, Broshears J, Liu L, Wilson TM, Kelley MR (1997) The
apurinic/apyrimidinic endonuclease (APE/ref-1) DNA repair enzyme is
elevated in premalignant and malignant cervical cancer. Anticancer Res 17:
3713-3719

168. Yan L, Bulgar A, Miao Y, Mahajan V, Donze JR, Gerson SL, Liu L (2007)
Combined treatment with temozolomide and methoxyamine: blocking
apurininc/pyrimidinic site repair coupled with targeting topoisomerase
IIalpha. Clin Cancer Res 13: 1532-1539

169. Yan T, Seo Y, Schupp JE, Zeng X, Desai AB, Kinsella TJ (2006)
Methoxyamine potentiates iododeoxyuridine-induced radiosensitization by
altering cell cycle kinetics and enhancing senescence. Mol Cancer Ther 5:
893-902

170. Yang H, Clendenin WM, Wong D, Demple B, Slupska MM, Chiang JH,
Miller JH (2001) Enhanced activity of adenine-DNA glycosylase (Myh) by
apurinic/apyrimidinic endonuclease (Ape1) in mammalian base excision
repair of an A/GO mismatch. Nucleic Acids Res 29: 743-752

171. Yao KS, Hageboutros A, Ford P, O'Dwyer PJ (1997) Involvement of
activator protein-1 and nuclear factor-kappaB transcription factors in the
control of the DT-diaphorase expression induced by mitomycin C treatment.
Mol Pharmacol 51: 422-430

122
172. Yoder KE, Bushman FD (2000) Repair of gaps in retroviral DNA integration
intermediates. J Virol 74: 11191-11200

173. Zawahir Z, Neamati N (2006) Inhibition of HIV-1 integrase activity by
synthetic peptides derived from the HIV-1 HXB2 Pol region of the viral
genome. Bioorg Med Chem Lett 16: 5199-5202

174. Zawahir Z, Dayam R, Deng J, Pereira C, Neamati N (2009) Pharmacophore
guided discovery of small-molecule human apurinic/apyrimidinic
endonuclease 1 inhibitors. J Med Chem 52: 20-32

175. Zhang H, Zhao Q, Bhattacharya S, Waheed AA, Tong X, Hong A, Heck S,
Curreli F, Goger M, Cowburn D, Freed EO, Debnath AK (2008) A cell-
penetrating helical peptide as a potential HIV-1 inhibitor. J Mol Biol 378:
565-580

176. Zhao L, O'Reilly MK, Shultz MD, Chmielewski J (2003) Interfacial peptide
inhibitors of HIV-1 integrase activity and dimerization. Bioorg Med Chem
Lett 13: 1175-1177

177. Zhu K, Dobard C, Chow SA (2004) Requirement for integrase during reverse
transcription of human immunodeficiency virus type 1 and the effect of
cysteine mutations of integrase on its interactions with reverse transcriptase. J
Virol 78: 5045-5055

123
Appendices

The research presented in the main body of this presentation reflects work done
pursuant to my thesis project targeting base excision repair in human colon cancer
models as a novel therapeutic strategy. As a medicinal chemistry laboratory, our goal
is the design and development of inhibitors to validated targets in various arenas of
therapeutic utility. This goal is largely achieved in this thesis for APE1, a ubiquitous
enzyme implicated in the resistance of many tumors to clinically relevant
chemotherapeutic agents.
Appendix I presents additional data to that reported in Chapter 3. These tables show
the structures and activity of molecules with moderate inhibitory activity to APE1, as
well as those that were found to be inactive. These molecules are important in that
they also provide structure-activity information for further endeavors in APE1
inhibitor development.
Appendix II presents colony formation assay data for compounds reported in Chapter
3. This initial screen sought to establish cytotoxicity and synergistic potential of
these compounds with gemcitabine and methotrexate, two antimetabolite agents
currently used for the treatment of various cancers.
During the course of this project, I have also been involved in research pertaining to
peptide inhibition of HIV-1 integrase (HIV-1 IN). HIV-1 IN is also a clinically
relevant target, being an essential enzyme for HIV-1 replication. Our focus on
peptide inhibition of HIV-1 IN is based on the premise that peptides can serve as
biological tools by which the protein-protein and protein-DNA interactions of HIV-1

124
integrase might be elucidated. To this end, we have published two manuscripts, with
one currently in progress, detailing the inhibitory activity and potential use of these
peptides as structural probes to gain insight into binding modes and interactions of
this enzyme. A key characterization of these peptides is that they are derived from
the HXB2 Pol genome of HIV1 (Appendix IIIa) and HIV-1 IN itself (Appendix IIIb
and IIIc).

125
Appendix A
Table A1. Moderately Active Compounds Retrieved Using
Pharmacophore H1NI2, A1NI2, A3NI1
S.No. Structure
Activity
IC
50
( μM)
S1
H1NI2
HO O
H
N OH
O
Br
32
S2
H1NI2
OH
O
O
HO

50
S3
H1NI2
HN OH O
S S
O
O
OH
O
O
HO

27
S4
A1INI2
NN OH
O
HO
O
NO
2
50
S5
A1NI2
N
N
H
N OH
O
HO
O O
O O

33
S6
A1NI2
S
O
O
O
OH
O
OH

40
S7
A1NI2
S
O
O
OH
O
OH

40

126

Table A1, continued
S8
H1-
A1NI2
S
O
OH
O
HO

33

S9
H1-
A1NI2
O
N
H
O HO
OH
O

55
S10
H1-
A1NI2
N
S
N
O
OH
O
OH

60
S11
H1-
A1NI2 O
S
S
NH
HN
O
O
O
O
OH
O
HO
O
50
S12
H1-
A1NI2
O
O
OH
S
O
O
HO

40
S13
H1-
A1NI2
HO
O
H
N
O
O
OH

40
S14
H1-
A1NI2
S
O
O HO
N
O
O
O
HO

30
S15
H1-
A1NI2
O
N
O
OH
O
S
O
O
HO

80

127
Table A1, continued
S16
H1-
A1NI2
HO
O
H
N
O N
N
N
H
N

30 ± 2
S17
H1-
A1NI2

N
H
N
HO
O
O
OH

80

S18
A3NI1
S
N
O O
N
N
N
N
N
OH
F
O HO
33

S19
A3NI1
S
O HO
N
O O
N
N
N
N
N
OH
Cl

87
S20
A3NI1
S
O HO
N
O O
N
N
O
N

40
S21
A3NI1
NN
S
H
N
S
O
O
H
N
O
OH
O
80
S22
A3NI1
S
NH
O
N
N HO
O
O
O

50

128
Table A1, continued
S23
A3NI1
N
S
O
O
S
O
O
HO
O

80
S24
Common S
S
OH
O O
HO
O
O
O
O
35
S25
Common
N
H
O
O OH
H
N
O
HO O

30

Table A2. Inactive Compounds Retrieved Using
Pharmacophore Models H1NI2, A1NI2, A3NI1
S.No. Structure
Activity
IC
50
( μM)
S26
H1NI2
S
S
NH
2
H
2
N
O
O
OH
O
O
HO
>100
S27
A1NI2
N N
S
S S
S
O
O
S
O
O
OH
HO

>100
S28
A1NI2
OH
S S
S O O
OH
O
O
OH
O
O
HO

>100

129

Table A2, continued
S29
H1-
A1NI2
O
O
O
O
N
N
N
N
N
N
H
N
NH

>100
S30
H1-
A1NI2
N
S
S
S
O
O
OH
S
O
O
OH

>100
S31
H1-
A1NI2
P
O
OH HO
O
O
POH HO
O

>100

S32
H1-
A1NI2
N
H
S HO
O
O
N N
H
N
S HO
O
O

>100

S33
H1-
A1NI2
O
HO
O
S
O
OH
O

>100
S34
H1-
A1NI2
O
OH
O
HO
O

>100
S35
H1-
A1NI2
O
O
OH
O
N
N
HN
N

>100

130
Table A2, continued

S36
A3NI1
N
N
N
N
S
S
S
O
O
O
O
N
O
O
N
S
O
O
N
O
O
O
N
O
>100

S37
A3NI1
N
N
S
OH
O
NH
2
O
O

>100
S38
A3NI1
S
N
N HO
O
O O
S
O
O

>100
S39
A3NI1
O
N
H
S
N
O
OH
O
>100
S40
A3NI1
O
OH
N
S
S
O
O

>100
S41
A3NI1
N N
N
H
N
H
N
O
N
N
O
Cl

>100
S42
A3NI1
S
N
O
O
O
N
H
O
OH
O
Cl

>100
S43
A3NI1
S
S
O
N
H
H
N
O
O
OH
O
>100

131
Table A2, continued
S44
A3NI1
S
S
O
N
H
H
N
O
O
O
HO

>100
S45
A3NI1
N
S
HN
S
O
OH
O
O

>100
S46
A3NI1
N
S
N
S
H
N
O
O
HO
O

>100
S47
A3NI1
H
N
O
O HO S
O
O
N
H
N
N
O
>100
S48
A3NI1
N
N
H
S
O
O
O
O HO
>100
S49
A3NI1
HN
N
H
N
O
O
H
N
OH
O

>100
S50
A3NI1
HN
N
H
O
O
NH
O
O
OH

>100
S51
A3NI1
S
S
H
N
O
O
OH
O
>100

132

Table A2, continued
S52
A3NI1
N
N NH
OH
O
N
H
S
O
O
NO
2
>100
S53
A3NI1
S
N
S
S
OH
O
O
OH

100
S54
A3NI1
S
S
O
OH O
O

>100
S55
A3NI1
NO
2
S
OH
O
HO
O
100
S56
A3NI1
S
N
H
O
OH O
O
OH

>100
S57
A3NI1
S
N
S
N
S
S
S
O
O
HO

>100
S58
A3NI1
NH
S
N
N O
O
HO
O

>100
S59
A3NI1
N
S
HO
O
O
O
H
N
O
HO
O

>100

133
Table A3. APE1 Inhibitory Activity of Compounds Retrieved Using Pharmacophore Model A3H1
S.No. Structure
Activity
IC
50
( μM)
S60
N
S
N
H
HO
O
O
S
NH
2
O
27

S61
NH
O
H
N
O
OH
O

35

Table A4. Inactive Compounds Retrieved Using Pharmacophore Model A3H1
S.No. Structure
Activity
IC
50

( μM)
S62
O
N
N
S
O
N
N
OH
O
2
N

>100
S63
O
N
N
S
N
N
H
N
OH
O
2
N

>100
S64
S
O
O
O
N
O
2
N

>100
S65
N
S O
O O
S
O O

>100

134
Table A4, continued
S66
S
N
O
O
NH
O
Cl
OH

>100
S67
S
N
O
O
NH
O
OH

>100
S68
N
S
O
O
O
S
Cl
NH
O
O

>100
S69
S
N
NH
O
O
NH
O
O

>100
S70
S
S
NH
O
O
NH
S
O
O
S

>100
S71
NH
O
S
N
N S
O
O

>100
S72
S
N
O
O
NH
O
O
O

>100
S73
O
S
S
NH
O
O
N H
O
O

>100

135
Table A4, continued
S74
N
N
N
N
S
N
O
O
O

>100
S75
S N
O
O
O
O
O

>100
S76
N
H
O
O
O
O
O

>100
S77
N
O
O
NH
O
N

>100
S78
N
H
O
O
O
O
O

>100
S79
N
H
O
O
O
O
O
O

>100
S80
S N
O
O
O O
N
O

>100

136
Table A4, continued
S81
N
N
N
S
NH
O
N
S
N

>100
S82
S
N
O
O
NH
O
OH
O

100
S83
O
O
O
S
O
S
O
O

>100
S84
N
S
N
OOH
O

100
S85
S
O
O
S
NN
NH
O

>100
S86
S
O
O
S
NN
NH
O
Cl

>100
S87
S
N
O
O
NH
O
Cl
OH

>100
S88
OH
O
O
F
F
F
N
S
O
O

>100

137
Table A4, continued
S89
S
O
NH
O
O
S
O H
O
O
O

>100
S90 N
N
O
O
O
S
N H
2
O
O

>100
S91
N
H
S
N
O
O
O
O
O
O

>100
S92
N N
N
S
N
S
N
N
N
N

>100
S93
S
N
O
O
NH
O
Cl
NN
S

>100
S94
O O
O O
O
I
O N

>100
S95
N
N
S
NH
O
O
O
O

>100

138

Table A4, continued
S96
N
S S
N
O
NH
S
S
O
O
O F
F
F

>100
S97
NH
O
S
NH
O
OH
O

100
S98
S
N
O
O
NH
O
F

>100
S99
NH
NH
O
N
Cl
S
O
O

>100
S100
N N
N
S
N
O
S
N
N

>100
S101
N N
N
N
S
NH
O
O
O

>100
S102
NH N
N
N
S
S
O
O

>100
S103
NH N
N
N
S
NH
2
S
O
O
O

>100
S104
NH N
N
N
S
O
S
O
O
O

>100
S105
N
N
N
S
N H
2
O
S
N
O
O

>100

139

Table A4, continued
S106
N
N N
S
O
O H
O
S
N
O
O
O

>100
S107
N
N
N
S
NH
O
S
S
O
O

>100
S108
N
S
O
O
F
S
N
O
O

>100
S109
N
O
N
N
N
S
N
O
O
Cl
N

>100
S110
N
O
N
N
N
S
N
O
O
O
Br

>100
S111
N
N
N
N
S
NH
O
O

>100

140
Table A4, continued
S112
N
O
N
N
N
O
N H
O
O O
O

>100
S113
N
O
N
N
N
O
S
NH
O
O
O
O

>100
S114
N
O
N
N
N
S
NH
O
O

>100
S115
N
S
O
O
S
NH
O
S NH
O
O

>100
S116
N
N
N
S
N
S
NH
O
N
S

>100
S117 S
O
O
S
N N
NH
O

>100
S118
NH
O
F
N
N
S
N
S
N
O
O

>100

141
Table A4, continued
S120
N
N
N
N
N
OH
N
S
O
O

>100
S121
N
N
N
N
N
OH
N
S
F
O
O
F

>100
S122
N
N
N
N
N
OH
N
S
O
O

>100
S123
N
O
N
N
S
O
O
F

>100
S124
N
O
N
N
S
O
O

>100
S125 S
N
O
O
S
NH
O
O
>100
S126
N
O
NH
O
S
O
O

>100

142
Table A4, continued
S127
N
N
N
N
N
OH
Cl
N S
O
O
O
F
F
F

>100
S128
N
S
N
N
N
O
O
N
S

>100
S129
N
N
S
N
NH
N
S
O
O

>100
S130
S
N
O
O
NH
O
N
O

>100
S131
S N
O
O
O
O
OH
O

100
S132 S
N
O
O
NH
O
N
O
N

>100

143

Table A4, continued
S133
S
NH
O
O
O
NH
O
O

>100
S134
S
N
O
O
NH
O
N
O
O

>100
S135
N
S
O
O
O
NH
S
O
O

>100
S136
N
S
O
O
NH
S
O
O

>100

144
Appendix B
The IC
50
value for gemcitabine and methotrexate in HCT116 p53 wild-type cells for
colony formation assays was found to be 100nM and 8.8 nM, respectively. Cells
were seeded at 300 cells/ well, and treated with IC25 values of gemcitabine and
methotrexate for 24 hours. At 24 hours, compounds to be tested were added at a final
concentration of 20 µM, and thereafter incubated for a further 24 hours. Media was
the removed and the cells allowed to proliferate for 72h before crystal violet was
added.

Figure B1. Each 24-well plate containing a screening compound was tested in the presence and
absence of gemcitabine at 0.18 µM. The most potent compounds from our APE1 inhibitors to
date were tested. Red circles indicate compounds that may show potential synergy.

145

Figure B2. Each 24-well plate containing a screening compound was tested in the presence and
absence of gemcitabine at 0.18 µM. We also tested additional compounds that showed activity
and had tetrazole moieties (BT). Red circles indicate compounds that may show potential
synergy.

146

Figure B2. continued

147

Figure B3. Each 24-well plate containing a screening compound was tested in the presence and
absence of methotrexate at 0.18 µM. The most potent compounds from the APE1 inhibitory
screen were tested. Red circles indicate compounds that may show potential synergy.

148

Figure B4. Each 24-well plate containing a screening compound was tested in the presence and
absence of methotrexate at 0.18 µM. We also tested additional compounds that showed activity
and had tetrazole moieties (BT). Red circles indicate compounds that may show potential
synergy.

149
Figure B4, continued

150
Appendix C
C.1 Inhibition of HIV-1 Integrase Activity by Synthetic Peptides Derived
from the HIV-1 HXB2 Pol Region of the Viral Genome
C.1.1 Introduction
The Pol genomes of retroviruses encode three essential enzymes—protease
(PR), reverse transcriptase (RT) and integrase (IN)—which are involved in their
replication (reviewed in
72, 143
). The integration of the reverse transcribed human
immunodeficiency virus (HIV) genome into host cell DNA is an essential step in
viral replication, catalyzed by IN in two main catalytic events
17
. The first of these
events is 3’-processing, in which IN removes two nucleotides adjacent to a
conserved CA dinucleotide from each end of linear viral cDNA. In the subsequent
strand transfer event, IN inserts the newly processed viral DNA into the host
genome. Host repair enzymes such as human Rad 18 are thought to complete the 5’-
end joining reaction to produce the provirus
105, 172
. Integration and events prior to it
are believed to occur as part of a pre-integration complex (PIC), in which IN, RT,
and other viral and cellular proteins interact, in the cytosol (early PIC) and in the
nucleus (late PIC)
101
.
IN is an attractive target for the development of anti-HIV therapeutics
because of its essential nature to the virus, and lack of host cellular counterparts. In
contrast to RT and PR, few selective inhibitors have been described for IN, and none
yet have FDA approval for clinical use
29
. Currently approved anti-HIV therapy

151
includes small-molecule and peptidomimetic PR and RT inhibitors and T-20, a
polypeptide targeting viral entry by mimicking the HR2 region of HIV
transmembrane protein gp41
74
.
Through well characterized three-dimensional structures of each domain, IN
is divided into three distinct domains with an N-terminal domain, a C-terminal
domain, and a central catalytic domain containing the three highly conserved
catalytic acidic residues: D(64), D(116) and E(152)
152
. It is also thought to associate
into a multimer in solution, at the very least a tetramer
44
. The multimer conformation
is thought to be critical to IN catalytic activity and thus the multimerization becomes
in itself an attractive therapeutic target. Indeed, the most potent IN peptide inhibitors
(peptides reproducing the α1 and α5 helices of the protein) described to date inhibit
IN by preventing the formation of functional oligomers and destabilization of
enzyme secondary structure
97
.
Multimeric forms of peptides are thought to act as multivalent inhibitors due
to the simultaneous occupation of two or more neighboring catalytic sites of the IN
multimer
79
. A range of peptides deriving from the dimerization interface of IN have
also been shown to block IN dimerization
176
. Many of the larger peptide inhibitors
described to date are analogs of a parent peptide, or have been chemically
derivatized to enhance inhibition activity
78, 176
.
Peptides have been gaining attention as a class of therapeutics primarily
because of advancement in delivery systems, resolution of stability issues, structural
modifications and clinical success, as with T-20
80
. In 1995, Puras Lutzke et al. used

152
a combinatorial approach to scan a synthetic hexapeptide combinatorial library of
several thousand compounds to identify a 6-mer peptide HCKFWW that inhibited
IN catalytic activity at an IC
50
of 2 µM
120
. Several peptide inhibitors of IN targeting
catalytic activity as well as oligomerization, although far from entering actual drug
development, have since been reported
32, 36
. These include synthetic peptides
36, 97,
135, 176
from various regions of the IN enzyme as well as those deriving from natural
products
78, 133
. There is also a fast-growing interest in the possibility of designing
inhibitors of IN that target cellular protein-IN and viral protein-IN interactions. To
this end, a recent study used phage display techniques to identify two 7-mer peptides
that possess very high affinity for IN, one of which inhibited only the strand transfer
reaction at an IC
50
value of 70 µM
36
. Peptidomimetic drugs derived from lead
peptide sequences containing structural modifications and derivations to improve
delivery and binding affinity have already been successful in the active site
inhibition of viral PR
124
. At least one study has suggested that conformational
analysis of peptides will allow the development of peptidomimetic and other
inhibitors against IN
96
.
The HIV Pol genome products in past years have all been subject to
immunological studies involving synthetic peptides
76
. The use of synthetic peptides
to map immunodominant epitopes of IN has been documented
109, 148
. Such studies
have helped to elucidate structure-function relationships in the HIV RT protein and
other Pol gene products
48, 122
. Inhibition of IN function and multimerization was
reported from monoclonal antibodies (MAbs) and anti-IN peptide sera
109
. It has also

153
been suggested that peptide fragments reproducing critical functional epitopes would
disrupt protein-protein interactions and inhibit biological activity of the enzyme
96
. A
30-mer synthetic peptide from amino acid residues 147-175 of HIV IN catalytic core
shown to interfere with IN integration activity in vitro was used to raise
monospecific antibodies that blocked IN catalytic activity at low concentrations
96
.
The peptide domain was shown to be a critical functional epitope exposed at the
protein surface, and the antibody-IN complex was unable to bind DNA. While
underlining the importance of the epitope, the existence of IN as a multimer when
bound to DNA was also confirmed
95
.
Here, we report on the inhibition activity of a library of peptides derived from
the HIV-1 HXB2 Pol region. The 20-mer peptides tested against purified wild-type
IN spanned the 1003 amino acid length of the polyprotein, with 10-mer overlaps
between sequential peptides. The most active peptide 56 (IC
50
values of 5 µM for 3’
processing and 2.5 µM for strand transfer) derives from the RT connector region of
the protein. Two other peptides, 34 and 53, showing IC
50
values in the micromolar
range also derive from the RT palm and connector regions. All three are in close
proximity to the active site of the enzyme. Additionally, peptide 68 deriving from the
RNase H region also exhibits inhibition of IN catalytic activity in the low
micromolar range as confirmed by another study
110
.
C.1.2 Results and Discussion
Peptide therapeutics has become a field of interest in the potential search for
HIV-1 IN inhibitors. Interactions of IN with host cell and other viral proteins are

154
vital to the successful completion of host infection and the production of viable
proviruses. Several host proteins have been found to interact with IN, implicated in
such diverse processes as nuclear localization
92
, DNA repair
105, 157
and the
formation of IN-cDNA complexes
46
. IN and viral DNA interactions occur as part of
the PIC in the cytosol, with IN in its oligomeric functional state
114
. Translocation via
the PIC to the nucleus for the final step of integration also occurs in conjunction with
host proteins such as LEDGF/p75
92
. It is a worthwhile effort to develop IN inhibitors
that cause disruption of such functional interactions of IN as this would almost
certainly lead to unsuccessful infection of a host cell. To probe for possible lead
compounds, we tested peptides deriving from the HIV-1 HXB2 Pol polyprotein that
significantly inhibit 3’ processing and strand transfer catalysis by IN at low
micromolar concentrations. Figure C1 shows the corresponding sequences of the five
most active peptides on the polyprotein.

The peptides described here are 20-mer peptides, and could function as parent
macromolecules to smaller inhibitory compounds or at the very least provide a lead
Figure C1. The five most potent peptides derive from different regions of the HIV1 HXB2 Pol
genome. Numbers adjacent to each peptide sequence indicate amino acid spans of the respective
regions on the Pol polypeptide. Numbers in parentheses indicate the number of the peptide.
(Domains not drawn to scale). Numbering according to the HXB2 Numbering Engine
nomenclature, Los Alamos National Laboratory.

155
peptide sequence in the development of derivatized peptide IN inhibitors. With
further investigation, derivatives of those peptides that disrupt either IN-DNA
binding or inhibit its catalytic activity to a great extent could become important
inhibitory tools.
Inhibition of Catalytic Activity of IN by Peptides from the PR Region of Pol
genome. Table C1 shows the sequences and IC
50
values for the two catalytic
activities of IN: 3’ processing and strand transfer at micromolar concentrations for
peptides 5–15 that derive from the PR region of the polyprotein.
Table C1. Inhibition of IN Activities by Peptides Deriving from the PR Region of Pol Polyprotein
com
pd
code sequence
3’ processing
IC
50
(µM)
strand
transfer IC
50
(µM)
region of
Pol
genome
5 4258 PSEAGADRQGTVSFNFPQVT >500 >500 41-60
6 4259 TVSFNFPQVTLWQRPLVTIK >500 >500 51-70
7 4260 LWQRPLVTIKIGGQLKEALL >500 >500 61-80
8 4261 IGGQLKEALLDTGADDTVLE >500 >500 71-90
9 4262 DTGADDTVLEEMSLPGRWKP >500 >500 81-100
10 4263 EMSLPGRWKPKMIGGIGGFI >500 >500 91-110
11 4264 KMIGGIGGFIKVRQYDQILI 358±14 290±8 101-120
12 4265 KVRQYDQILIEICGHKAIGT >500 >500 111-130
13 4266 EICGHKAIGTVLVGPTPVNI >500 >500 121-140
14 4267 VLVGPTPVNIIGRNLLTQIG >500 241±38 131-150
15 4268 IGRNLLTQIGCTLNFPISPI 183±28 70±10 141-160
16 4269 CTLNFPISPIETVPVKLKPG 300±50 267±57 151-170

Inhibition of Catalytic Activity of IN by Peptides from The RT region of the Pol
polyprotein. Peptides 15-59 (Table C2) derive from the RT region of the Pol
polyprotein show a wide range of activity against wild-type purified IN.

156
Table C2. Inhibition of IN Activities by Peptides Deriving from the RT Region of Pol Polyprotein
comp
d
code sequence
3’ processing
IC
50
(µM)
strand
transfer IC
50

(µM)
region of
Pol
genome
15 4268 IGRNLLTQIGCTLNFPISPI 183±28 70±10 141-160
16 4269 CTLNFPISPIETVPVKLKPG 300±50 267±57 151-170
17 4270 ETVPVKLKPGMDGPKVKQWP 283±57 217±28 161-180
18 4271 MDGPKVKQWPLTEEKIKALV 333±28 333±28 171-190
19 4272 LTEEKIKALVEICTEMEKEG 220±26 217±28 181-200
20 4273 EICTEMEKEGKISKIGPENP >500 >500 191-210
21 4274 KISKIGPENPYNTPVFAIKK >500 >500 201-220
22 4275 YNTPVFAIKKKDSTKWRKLV >500 >500 211-230
23 4276 KDSTKWRKLVDFRELNKRTQ >500 >500 221-240
24 4277 DFRELNKRTQDFWEVQLGIP 483±28 172±25 231-250
25 4278 DFWEVQLGIPHPAGLKKKKS 375±25 183±28 241-260
26 4279 HPAGLKKKKSVTVLDVGDAY >500 >500 251-270
27 4280 VTVLDVGDAYFSVPLDEDFR >500 216±28 261-280
28 4281 FSVPLDEDFRKYTAFTIPSI >500 >500 271-290
29 4282 KYTAFTIPSINNETPGIRYQ >500 >500 281-300
30 4283 NNETPGIRYQYNVLPQGWKG >500 >500 291-310
31 4284 YNVLPQGWKGSPAIFQSSMT >500 300±50 301-320
32 4285 SPAIFQSSMTKILEPFRKQN >500 350±0 311-330
33 4286 KILEPFRKQNPDIVIYQYMD 250±50 77±11 321-340
34 4287 PDIVIYQYMDDLYVGSDLEI 5.6±1 10±1 331-350
35 4288 DLYVGSDLEIGQHRTKIEEL 266±28 207±5 341-360
36 4289 GQHRTKIEELRQHLLRWGLT >500 383±28 351-370
37 4290 RQHLLRWGLTTPDKKHQKEP >500 >500 361-380
38 4291 TPDKKHQKEPPFLWMGYELH >500 >500 371-390
39 4292 PFLWMGYELHPDKWTVQPIV 216±28 175±25 381-400
40 4293 PDKWTVQPIVLPEKDSWTVN 240±51 233±28 391-410
41 4294 LPEKDSWTVNDIQKLVGKLN 325±43 353±5 401-420
42 4295 DIQKLVGKLNWASQIYPGIK 270±51 333±28 411-430
43 4296 WASQIYPGIKVRQLCKLLRG 333±28 293±7 421-440
44 4297 VRQLCKLLRGTKALTEVIPL >500 405±7 431-450
45 4298 TKALTEVIPLTEEAELELAE >500 467±57 441-460
46 4299 TEEAELELAENREILKEPVH 236±55 283±28 451-470
47 4300 NREILKEPVHGVYYDPSKDL >500 217±28 461-480
48 4301 GVYYDPSKDLIAEIQKQGQG 233±28 >500 471-490
49 4302 IAEIQKQGQGQWTYQIYQEP 283±28 283±28 481-500
50 4303 QWTYQIYQEPFKNLKTGKYA 258±28 217±28 491-510
51 4304 FKNLKTGKYARMRGAHTNDV >500 >500 501-520
52 4305 RMRGAHTNDVKQLTEAVQKI 467±57 283±28 511-530
53 4306 KQLTEAVQKITTESIVIWGK 6.6±1 4±1 521-540
54 4307 TTESIVIWGKTPKFKLPIQK >500 >500 531-550
55 4308 TPKFKLPIQKETWETWWTEY 217±28 145±32 541-560
56 4309 ETWETWWTEYWQATWIPEWE 5.8±0.3 2.4±0.4 551-570
57 4310 WQATWIPEWEFVNTPPLVKL 267±28 283±28 561-580
58 4311 FVNTPPLVKLWYQLEKEPIV 308±22 313±55 571-590

157
The three most potent peptides 34, 53 and 56 all derive from the RT-RNase H region
of the polyprotein, as shown in Table C2 and Figure C2B.

It has been shown previously that RT and IN physically interact with and inhibit
each other’s catalytic activity
61, 111
. IN has been shown to bind to two discontinuous
regions on RT, the fingers-palm region (amino acids 1 to 242) and carboxy-terminal
half of the connection region. The RT binding region on IN has been mapped to the
carboxy-terminal region from amino acids residues 210 to 288
62
. Deletion analysis
on RT has revealed two IN binding domains: the fingers-palm domain and the
carboxy terminal domain of the connection subdomain
62
. It is interesting to note that
Figure C2. Representative SDS-PAGE image showing inhibition of purified IN by peptide #34,
53, 56 and 65. A) Schematic of IN activity in vitro: a 21-mer oligonucleotide corresponding to
the U5 LTR 5’-end labeled with 32P is reacted with purified IN. The first step, 3’-processing,
involves nucleolytic cleavage of two bases from the 3’-end resulting in a 19-mer oligonucleotide.
the 3’-ends are subsequently covalently joined at several sites to another identical
oligonucleotide that serves as the target DNA. This reaction is known as strand transfer, and the
products formed migrate slower than the original substrate. B) Lane 1, DNA alone; lane 2 and
15, IN alone; lanes 3-14, IN, DNA, and respective peptides at decreasing concentrations of 18, 6,
2 and 0.6 µM. ST: strand transfer.

158
the two most active peptides, 53 and 56 both derive from the RT connection domain.
A third peptide 34 derives from the fingers-palm domain. It is thought that RT
prevents IN self integration into viral DNA before reaching the nucleus, thus
regulating the integration process as part of the pre-integration complex (PIC)
111
. A
similar function has been attributed to an IN interacting protein, BAF (barrier-to-
autointegration factor)
139
. The RT and other protein-protein interactions of IN within
and without the PIC are thought to be essential in early stages of viral replication,
reverse transcription, nuclear localization and integration
24, 92, 111, 163, 164
. Peptide
disruption of the RT-IN complex or any other essential IN-protein interaction could
prevent effective integration into the host genome if self-integration or other
integration prohibitive processes occurred during migration of the PIC to the
nucleus. The exact sequences of RT and IN involved in the interaction have not been
elucidated, although a recent study indicates that inhibitory peptides 33 from the RT
polymerase domain and 68 from the RNase H domain, among others, are involved in
direct interactions with IN
110
. Peptide 33 was found to have IC
50
values less than 5
µM for both strand transfer and 3’-processing
110
. Our own assay of these peptides
did not yield similar results, a difference possibly due to different assay conditions.
Peptide 56 contains six tryptophan residues that possibly contribute to its
potency in inhibiting IN. Tryptophan rich peptides have been known to possess IN
inhibitory activity, as evidenced by indolicidin, a natural antimicrobial peptide and
its analogs
78
. It is thought that this amino acid intercalates into DNA, and could
exert inhibitory activity by sterically disrupting IN-DNA binding. Figure 4 shows the

159
location on the RT protein of peptides 34, 53, and 54 in context of the tertiary
structure of the protein. Although we do not know the conformations these peptides
assume in solution in vitro, it is interesting to note that aspartic acid residues present
in each of the peptides are either within the polymerase active site (peptide 34) or in
close proximity to the RT active site (peptides 53 and 54).
Inhibition of Catalytic Activity of IN by Peptides from The RNase H Region of
the Pol polyprotein. Table C3 shows the IC
50
values for peptides 59-71 deriving
from the RNase H region of the HIV Pol genome.
Table C3. Inhibition of IN Activities by Peptides Deriving from the RNase H of Pol Polyprotein
com
pd
code sequence
3’ processing
IC
50
(µM)
strand
transfer
IC
50
(µM)
region of
Pol
59 4312 WYQLEKEPIVGAETFYVDGA 240±17 171±7 581-600
60 4313 GAETFYVDGAANRETKLGKA >500 117±28 591-610
61 4314 ANRETKLGKAGYVTNRGRQK >500 >500 601-620
62 4315 GYVTNRGRQKVVTLTDTTNQ >500 >500 611-630
63 4316 VVTLTDTTNQKTELQAIYLA 230±15 120±28 621-640
64 4317 KTELQAIYLALQDSGLEVNI 56±2 21±3 631-650
65 4318 LQDSGLEVNIVTDSQYALGI 11.3±1 1.8±0.3 641-660
66 4319 VTDSQYALGIIQAQPDQSES 203±15 233±28 651-670
67 4320 IQAQPDQSESELVNQIIEQL 353±5 341±14 661-680
68 4321 ELVNQIIEQLIKKEKVYLAW 15±2 14±4 671-690
69 4322 IKKEKVYLAWVPAHKGIGGN >500 >500 681-700
70 4323 VPAHKGIGGNEQVDKLVSAG 283±28 333±28 691-710
71 4324 EQVDKLVSAGIRKVLFLDGI 467±57 153±5 701-720

Peptide 65 from the RNase H region of the Pol genome shows low micromolar
activity for both IN catalytic activities, particularly strand transfer. This peptide does
not contain any lysine or tryptophan residues, suggesting that its mechanism of
inhibition does not involve DNA binding. This is confirmed by the Schiff base assay
performed below, which did not indicate the formation of a peptide-IN-DNA
crosslinked complex for this peptide. The particular sequence of 65 deriving from the

160
RNase H region has not previously been identified as possessing IN inhibitory
properties, unlike peptide 68, also deriving from RNase H
110
.
Inhibition of Catalytic Activity of Peptides from The IN Region of the Pol
polyprotein. Table C4 shows the IC
50
values for peptides deriving from the IN
region of the HIV Pol genome.

The activity of the peptides deriving from the IN region of the Pol polyprotein is not
as high as those mentioned above. However, peptides 78, 82, 89, and 100 shows
comp
d
code sequence
3’
processing
IC
50
(µM)
strand
transfer
IC
50
(µM)
region of
Pol genome
71 4324 EQVDKLVSAGIRKVLFLDGI 467±57 153±5 701-720
72 4325 IRKVLFLDGIDKAQDEHEKY 298±2 103±5 711-730
73 4326 DKAQDEHEKYHSNWRAMASD 420±12 183±57 721-740
74 4327 HSNWRAMASDFNLPPVVAKE >500 >500 731-750
75 4328 FNLPPVVAKEIVASCDKCQL 203±11 120±26 741-760
76 4329 IVASCDKCQLKGEAMHGQVD >500 189±19 751-770
77 4330 KGEAMHGQVDCSPGIWQLDC 241±14 117±28 761-780
78 4331 CSPGIWQLDCTHLEGKVILV 267±28 61±2 771-790
79 4332 THLEGKVILVAVHVASGYIE 383±28 258±14 781-800
80 4333 AVHVASGYIEAEVIPAETGQ 417±57 330±20 791-810
81 4334 AEVIPAETGQETAYFLLKLA >500 283±28 801-820
82 4335 ETAYFLLKLAGRWPVKTIHT 208±14 60±5 811-830
83 4336 GRWPVKTIHTDNGSNFTGAT 467±57 410±50 821-840
84 4337 DNGSNFTGATVRAACWWAGI 203±5 90±17 831-850
85 4338 VRAACWWAGIKQEFGIPYNP 467±57 433±57 841-860
86 4339 KQEFGIPYNPQSQGVVESMN 467±57 467±57 851-870
87 4340 QSQGVVESMNKELKKIIGQV >500 >500 861-880
88 4341 KELKKIIGQVRDQAEHLKTA 350±0 183±28 871-890
89 4342 RDQAEHLKTAVQMAVFIHNF 93±6 53±2 881-900
90 4343 VQMAVFIHNFKRKGGIGGYS >500 333±28 891-910
91 4344 KRKGGIGGYSAGERIVDIIA >500 >500 901-920
92 4345 AGERIVDIIATDIQTKELQK 333±28 183±28 911-930
93 4346 TDIQTKELQKQITKIQNFRV >500 >500 921-940
94 4347 QITKIQNFRVYYRDSRNPLW >500 283±28 931-950
95 4348 YYRDSRNPLWKGPAKLLWKG >500 467±57 941-960
96 4349 KGPAKLLWKGEGAVVIQDNS 483±28 417±28 951-970
97 4350 EGAVVIQDNSDIKVVPRRKA >500 333±28 961-980
98 4351 DIKVVPRRKAKIIRDYGKQM 288±20 193±11 971-990
99 4352 KIIRDYGKQMAGDDCVASRQ >500 333±28 981-1000
100 4353 AGDDCVASRQDED 163±23 63±5 991-1003
Table C4. Inhibition of IN activities by peptides deriving from the IN Region of the Pol
polyprotein.

161
strand transfer IC
50
values less than 70 µM. The perceived selectivity towards strand
transfer of some peptides suggests that the mechanisms involving each of these
catalytic processes are not completely identical, and that IN may possess a different
structural conformation for viral and target DNA despite the same critical amino acid
residues being required for both IN catalytic activities
36, 40, 43, 61, 82
. The mechanism of
inhibition of the peptides is not known; it is possible that they may interfere with the
multimerization of the enzyme. A previous study published results of a 30-mer
peptide SQGVVESMNKELK159KIIGQVRDQAEHLKTAY deriving from the
catalytic core region of the IN protein, with IC
50
values in the micromolar range
135
.
Our assays of the two 20-mer peptides containing this sequence, 86 and 87, did not
show a significant inhibition of the wild type IN. It is possible that, in this case the
longer peptide adopts a conformation that inhibits enzyme catalytic function to a
greater extent than two smaller peptides.
Schiff-Base Assays to Determine Peptide-DNA Binding.

Figure C3a. Mechanism of the Schiff base assay. Integrase attaches covalently to the abasic site
via a Schiff base. The imine formed is stabilized by reduction with sodium cyanoborohydride.

162
To gain some insight into whether these peptides inhibit the DNA binding of IN, we
used an assay in which the enzyme is cross-linked to a depurinated oligonucleotide
DNA substrate with an abasic site. IN forms a Schiff base between the abasic site
and its own nearby lysine residue. Stabilization of the imine occurs with reduction to
the amine by sodium cyanoborohydride (Figure C3a). This assay is detailed by
Mazumder, et al
98
. Figure C3b shows the Schiff base formation of five of the more
active peptides in an order-of-addition experiment. In the presence of the divalent
metal cation (Mn
2+
), IN binding to DNA was slightly inhibited by three of the
peptides, although those with the highest inhibitory activity did not show significant
DNA binding.

Of these, two (53 and 89 from the RT protein) had lysine residues which suggests
that the peptides bind to the DNA substrate. Disruption of the IN-DNA crosslinked
complex was not identified by any of the seven peptides tested in this assay.
C.1.3 Conclusion. The active peptides we have described in this paper could
potentially serve as novel probes towards understanding the interactions of IN with
Figure C3b. Schiff Base Crosslinking Experiments. Lane 1, IN alone. Lanes 2-8, peptide #19,
34, 53, 56, 63, 65, 89 at a final concentration of 50 µM. The IN enzyme, peptide and DNA were
added to the reaction mixtures in different combinations and incubated at 30ºC as shown above.

163
other cellular and viral proteins. Since such interactions are known to be essential for
successful integration of viral cDNA into host DNA, an understanding of such
protein-protein and protein-DNA interactions will not only help to characterize the
protein itself, but will also contribute to the development of effective inhibitors to IN
catalytic activity if successful disruption of the interaction can be initiated. Chemical
derivatization or synthetic analogs of these peptides may also function as potent IN
inhibitors of these peptides by the disruption of IN-DNA binding, causing
conformational changes in the protein that may prevent productive integration. It is
foreseeable that with the enhancement of peptide bioavailability and stability and
advances in peptide delivery systems, peptide-derived molecules will become viable
therapeutic options in the future. Although little is known about the inhibition
mechanism of these peptides, additional information from affinity or binding studies
to IN may be utilized towards three-dimensional structure elucidation of the protein.
It would also be of interest to try to characterize possible sites of cellular protein
interaction with IN using antibodies raised to peptides with a high affinity to IN.
Introduction of the specific antibodies into HIV-infected cell lines could provide
more information on the activity of IN in vivo.

164
C.2 Sequence-based design and discovery of peptide inhibitors of HIV-1
integrase: insight into the binding mode of the enzyme
C.2.1 Introduction
HIV-1 integrase (IN) is essential for viral replication. Following reverse transcription
of the RNA into DNA by HIV-1 reverse transcriptase, IN integrates the viral DNA
into the host genome. IN performs its catalysis in two steps: 3’-processing and strand
transfer. Initially, it cleaves the viral DNA ends at the highly conserved CAGT site
to release GT dinucleotide in the cytoplasm.
8, 18, 39
IN subsequently transfer the viral
DNA into the host chromosome in the nucleus. Successful integration requires the
interaction of IN with other viral and host cell proteins including but not limited to,
LEDGF—thought to be involved in nuclear translocation of the pre-integration
complex; BAF, thought to prevent autointegration of viral DNA; Rad18 and Poly
(ADP-ribose) polymerase, which are among a group of repair enzymes thought to
interact with IN during DNA repair and ligation following the integration process
(for a recent review see
142
). The viral reverse transcriptase protein is also thought to
interact with IN in the pre-integration complex.
138

Specific regions of IN are known to be essential for catalytic function.
20

Analogously, there exist specific regions on the enzyme that are required for distinct
protein-protein and protein-DNA interactions that are critical for efficient
integration. Binding and catalytic activity assays of rationally designed peptides
against targeted proteins should provide a unique tool towards elucidating—at a
molecular level—important biochemical sites of the enzyme. In this study, we have

165
defined possible regions of IN involved in catalysis at the amino acid level. In
particular, sequences derived from the α1 helix and the α3 helix of IN appear to play
an important role in binding the enzyme and interrupting its catalytic activity. Our
unpublished data suggests that these peptides are size-dependent, as lengthier
peptides containing the same residues show differences in the inhibition of catalytic
activity. Previous studies have also implicated the α1 helical region in inhibition of
IN activity.

However, we attempt to pinpoint specific amino acid residues within this
region that are perhaps essential to IN catalytic activity and oligomerization.
The success of developing PR inhibitors is based on rational design of
peptidic and peptidomimetic compound with high selectivity. Although numerous
small molecule inhibitors of IN are known, peptide-based approaches targeting IN
have been very limited. The first and most elegant study in this respect was reported
by Plasterk et al. who identified a hexapeptide inhibitor of IN by using a
combinatorial chemical library. They screened a synthetic library of potentially
400,000 hexapeptides and identified a lead peptide with the sequence HCKFWW
that inhibited IN-mediated 3'-processing and integration with an IC
50
of 2 µM. This
peptide also inhibited HIV-2, FIV and MLV integrases, suggesting that a conserved
region of IN is targeted. They suggested that HCKFWW acts at or near the catalytic
site of IN. A more recent study used phage-display techniques to identify a novel
peptide that selectively inhibited the strand transfer activity of IN. In addition, 10-
mer long peptides deriving from the reverse transcriptase region of the HIV-1 Pol
protein also inhibit IN catalytic. Other inhibitory peptides of IN targeting catalytic

166
activity as well as protein oligomerization, although far from entering actual drug
development, have since been reported. Zargarian et al. showed that a peptide
reproducing the α5 dimerization helix sequence of IN inhibits catalytic activity, as
does a peptide with the α1 helix sequence, albeit at a higher micromolar
concentration.
97
Another study screened peptides deriving from the dimerization
interface of IN also shows similar results and indicates that interaction of such
peptides with the enzyme may block its dimerization.
176
Finally, it has been shown
that a 30-mer peptide from residues 147-175 of IN inhibited catalytic activity at low
millimolar concentrations. This study also suggested that conformational analysis of
peptides deriving from IN will allow the development of peptidomimetic and other
inhibitors against IN.
96
Although several larger peptides deriving from other sources
have been reported to inhibit IN in vitro, no other study thus far has attempted to
systematically and rationally design small-peptide inhibitors of IN.
32
Also reported
are peptide inhibitors targeting other aspects of HIV-1 life cycle, including T-20—a
fusion inhibitor—and capsid assembly inhibitor (CAI), targeting the assembly of
capsid particles in vitro.
74, 136

In an effort to take advantage of a growing list of studies defining the role of
particular amino acid residues in catalysis and DNA binding (Table C5, for a review
see
41, 107
), we designed a series of small peptides, each of which contain at least one
amino-acid residue important for IN catalytic activities and/or viral replication. Our
approach is based on the sequence of HIV-1 IN to explore the mechanism of

167
integration, DNA and drug binding, and to provide a new pharmacophore for
antiviral drug design and development.
Table C5. Catalytic and Replication Activities of a Series of Integrase Mutants
Sequence
a
Segment Mutation 3’-pr.
b
Integ.
c
Disint.
d

Repl.
(infect)
e

EEHEKYHSNW
10-19
H12A
H16A
+ + +++
+++
-
-
ASCDKCQLKG
38-47
C40A
C43A
-
+
-
+
+++
+++
-
-
HGQVDCSPGIWQL
DCTH
51-67
H51A
D64A
T66A

-
+

-
+

-
+++
+
-
+
VHVASGY
77-83
S81A
+/- +/- +/- -
PAETGQET
90-97
E92A
++ ++ ++ ++
TAYFLLKLAGRW
97-108

GRWPVKT
106-112
P109A
+/- +/- - +
HTDNGSNF
114-121
D116
N117S
F121A
-
+
+/-
-
+
-
-
+
-

ACWWAGIKQEF
129-139
I135P
K136A

+/-

+/-

+++
-
-
FGIPYNPQSQ
139-148
Y143N
S147I
+++
-
+++
+++
Delaye
d
-
ESMNKELKKI
152-161
E152A
S153A
-
+
-
+
-
+++
-
+++
VRDQAEHLKT
165-175
R166A
-
FIHNFKRK
181-188
F185A
+++ +++ -
GYSAGERIVD
193-202
R199A
+++ +++ +++ -
WKGPAKLLWK
235-244
W235A
L241A
L242A
+++
-
-
+++
-
-
+++
-
+
+/-

VPRRKAKI
260-267
V260E
R262G
R263L
-
++
++
-
++
++
-
+
++

The residues in bold are important for HIV-1 integrase catalytic activity in vitro and effective viral
replication.
b
3’-processing;
c
integration (strand transfer);
d
disintegration;
e
Replication or infectivity
capability. Notations: - (0-10% activity of the wild-type enzyme); + (10-40%); ++ (40-80 %); +++
(80-100%).

With this approach it may be possible to find the Achilles' heel of the enzyme that
might be important for the development of drug resistance, and to discover a novel

168
site that could be used to develop inhibitors. These peptides, in addition to interfering
with HIV-1 IN catalytic activity, may also disrupt HIV-1 infectivity, and thus serve
as possible leads for peptidomimetic prodrugs against the virus. In this study, the
design, synthesis, and evaluation of these small peptides including amino acid
substituted and truncated analogs as IN inhibitors are reported. We have discovered
that certain amino acid residues are key in the inhibitory potential of these peptides,
and may also be implicated in IN catalytic activity. Furthermore, conformational
requirements were also explored with respect to the inversion and retroinversion of
the peptide sequences, providing a better understanding of the mode of IN inhibitory
action by these new agents.
We prepared a series of sequence-based small peptides. As shown in Table C6, our
designed peptides cover the entire functional domains of IN (Figure C4). Two
peptides show selective inhibitory activity against wild-type IN: NL-6,
corresponding to residues 97-108 of IN, and NL-9, corresponding to residues 129-
139 of IN. Figure C4.C shows the location of these peptides on the enzyme. The
structures of these two peptides and their position on the IN crystal structure (PDB
1K6Y) are shown in Figure C5.
C.2.2 Activity of sequence-based peptides. Inhibition of IN specific catalytic
activities were carried out by monitoring both the DNA cleavage step (3’-processing)
and the DNA strand transfer (integration). Our initial sequence-directed peptide
library indicated that the methodology we employed for the design of IN inhibitors
on the basis of the amino acid sequence of IN is encouraging. Among the 16

169
synthesized peptides, two compounds, i.e. NL-6 and NL-9, exhibited promising IN
inhibitory activity with IC
50
values of 21±7 µM and 95±9 µM for 3’-processing; and
2.7±1 µM and 56±5 µM for integration, respectively (Table C6), even though the
other 14 peptides did not show obvious inhibitory activity toward IN.
Figure C4. Three functional motifs of HIV-1 IN. The N-terminus shown in green contains the highly
conserved HHCC motif that binds to zinc. The catalytic core shown in red contains the highly
conserved DDE motif that binds to divalent metals. The least conserved C-terminus shown in blue
binds to DNA. (b) Sequence of the full length HIV-1 integrase. (c) The secondary structure of IN.
The peptides with the best inhibitory activity show sequences deriving from the 1 helix (NL6) and
part of the 3 helix (NL9). (d) The sequences of the lead peptides NL6 and NL9.

170
Table C6. The Sequence and HIV-1 Integrase Inhibitory Activity of the Designed Peptides

Region Code Sequence Fragment
3’
Processing
(IC
50,
μM)
Strand
Transfer
(IC
50,
μM)
NL1 EEHEKYHSNW 10-19 >2000 >2000 N-
terminus NL2 ASCDKCQLKG 38-47 >2000 >2000
NL3 HGQVDCSPGIWQLDCTH 51-67 1000 1000
NL4 VHVASGY 77-83 >2000 >2000
NL5 PAETGQET 90-97 >2000 >2000
NL-6 TAYFLLKLAGRW 97-108 21±7 2.7±1
NL7 GRWPVKT 106-112 >2000 >2000
NL8 HTDNGSNF 114-121 >2000 >2000
NL-9 ACWWAGIKQEF 129-139 95±9 56±5
NL10 FGIPYNPQSQ 139-148 >1000 >1000
NL11 ESMNKELKKI 152-161 >2000 >2000
NL12 VRDQAEHLKT 165-175 >2000 >2000
NL13 FIHNFKRK 181-188 >2000 >2000

Catalytic
core
NL14 GYSAGERIVD 193-202 >2000 >2000
NL15 WKGPAKLLWK 235-244 >1000 >1000 C-
terminus NL16 VPRRKAKI 260-267 >1000 >1000

This finding suggests that the two segments of NL-6 and NL-9 might be selectively
targeting IN. Using the two peptides as lead compounds (Figure C5), further
structural modifications were carried out to provide high-affinity IN inhibitors, and
to furnish information on IN topography concerning the bound conformation of the
inhibitor.

171
Figure C5. The chemical structures of the lead compounds NL-6 and NL-9 (a) and their position on
HIV-1 IN (b). The most potent peptides (red, NL-6, and blue, NL-9) are shown in the context of four
identical tetramers (green, brown, cyan, orange) of the IN N-terminal domain and the catalytic core
domain (PDB 1K6Y).

Alanine Scanned Analogs of NL-6 and NL-9. We synthesized alanine-substituted
mutations of NL-6 and NL-9 to explore the functional importance of the constituent
amino acid residues in the two peptides. The activity data (Table C7) indicate that F
4
,
L
6
, L
8
, and W
12
are the essential residues in NL-6, whose deletion resulted in a

172
significant loss of the IN inhibitory activity. The loss of activity for NL-6 F4A, L6A,
L8A, and W12A were >17-fold for 3’-processing and >120-fold for strand transfer.

Table C7. The HIV-1 Integrase Inhibitory Activity of the Alanine-scanned Analogs of NL-6
and NL-9

Compound Sequence
3‘-Processing
(IC
50,
μM)
Strand Transfer
(IC
50,
μM)
NL-6
TAYFLLKLAGRW
21±7 2.7±1
NL6-T1A
AAYFLLKLAGRW
100±10 47±7
NL6-Y3A
TAAFLLKLAGRW
193±10 119±11
NL6-F4A
TAYALLKLAGRW
>333 >333
NL6-L5A
TAYFALKLAGRW
115±21 51±7
NL6-L6A
TAYFLAKLAGRW
>333 >333
NL6-K7A
TAYFLLALAGRW
113±15 56±7
NL6-L8A
TAYFLLKAAGRW
>333 106±7
NL6-G10A
TAYFLLKLAARW
118±10 19±6
NL6-R11A
TAYFLLKLAGAW
83±15 80±8
NL6-W12A
TAYFLLKLAGRA
>333 >333
NL 9
ACWWAGIKQEF
95±9 56±5
NL9-C2A
AAWWAGIKQEF
277±47 311±19
NL9-W3A
ACAWAGIKQEF
33±6 34±8
NL9-W4A
ACWAAGIKQEF
>333 >333
NL9-G6A
ACWWAAIKQEF
90±10 43±7
NL9-I7A
ACWWAGAKQEF
>333 >333
NL9-K8A
ACWWAGIAQEF
62±13 55±7
NL9-Q9A
ACWWAGIKAEF
>333 >333
NL9-E10A
ACWWAGIKQAF
>333 >333
NL9-F11A
ACWWAGIKQEA
245±13 206±12

For the NL-9 peptide, residues C
2
, W
4
, I
7
, Q
9
, E
10
, and F
11
are each vital for the
inhibition of the IN. These residues might be involved in the direct contact with
binding the surface of the enzyme. However, an interesting result is observed when
substituting the Trp3 with an Ala residue. The 3’-processing and strand transfer
inhibitory activity of the peptide were both increased by 4-fold (IC
50
23±6 µM), and
2.3-fold (IC
50
24±8 µM), respectively relative to the parent peptide NL-9.

173
Substitution of Trp4 with an Ala has the opposite effect: inhibitory activity is
completely abolished for both integration and strand transfer. The same result occurs
with the Trp3 and Trp4 substitutions for other amino acids in the single substituted
analogs of NL-9. Figure C6 shows the results of NL9-W4A and NL9-W3A
substitutions in comparison with the parent peptide.
Figure C6. The isomeric forms of NL-6 and NL-9 show varying activity against wild-type IN. a) The
isomers of NL9 tested against wild type IN show absolute loss in activity. Lane 1: DNA only, lane 2:
wild type IN, lanes 3-5: 333, 111, 37 µM of RNL9, lanes 6-8 333, 111, 37 µM of DNL9, lanes 9-11
333, 111, 37 µM of RDNL9.

Single Substituted Analogs of NL-6 and NL-9. With the identification of essential
amino acids in NL-6 and NL-9 through alanine scanning, we chose Y
3
, K
7
and W
12
of
NL-6, and C
2
, W
3
, W
4
and K
8
of NL-9 to make several additional substitutions to
determine what structural feature of the side chain is favored for the inhibition of IN.
Therefore, during this second stage of the structure-activity study, single substituted
analogs were synthesized and evaluated for their IN inhibitory activity. The
biological data are summarized in Table C8. Our results indicate that both the

174
tyrosine and tryptophan residues in NL-6 are important in maintaining IN inhibitory
activity. When the tyrosine residue is substituted by serine the IC
50
values were
found to be 186±23 µM for 3’ processing and 11±2 µM for integration. Replacement
by a leucine for the tryptophan residue produced IC
50
values of 315±30 µM and
38±2 µM for 3’-processing and integration, respectively. This is a six-fold loss in
catalytic activity. However, when the lysine residue was substituted by isoleucine,
the potency was increased three-fold, suggesting that the side chain in this position
might be directed into a hydrophobic region on the binding surface of the IN. With
respect to NL-9, an ACWW motif proved essential for maintaining the IN activity.
Substitution of cysteine by serine showed an IC
50
values of 294±41 µM and 163±15
µM for 3’-processing and integration respectively, while replacement of one of the
tryptophan residues (Trp4) with glycine resulted in a total loss of the IN inhibitory
activity (Table C8).
Table C8. The HIV-1 Integrase Inhibitory Activity of Single-substituted Analogs of NL6 and NL-9

Compound Sequence
3‘-Processing
(IC
50,
μM)
Strand Transfer
(IC
50,
μM)
NL6 TAYFLLKLAGRW 21±7 2.7±1
NL 6-1 TASFLLKLAGRW 186 ± 23 11 ± 2
NL 6-2 TAYFLLILAGRW 4.1 ± 0.7 3.0 ± 1.0
NL 6-3 TAYFLLKLAGRL 315 ± 30 38 ± 2
NL 9 ACWWAGIKQEF 95 ± 9 56 ± 5
NL 9-1 ASWWAGIKQEF 294 ± 41 163 ± 15
NL 9-2 ACGWAGIKQEF 46 ±5 16 ± 2
NL 9-3 ACWGAGIKQEF > 333 > 333
NL 9-4 ACWWAGIRQEF > 333 > 333

Arginine substitution for lysine also caused remarkable activity loss. These
preliminary results suggest that the basic residues, e.g. Lys and Arg, and the Trp in

175
common between NL-6 and NL-9 might be important for the functional interactions.
Interestingly, similar to the alanine scanning results, substitution of Trp3 with
glycine resulted in a two-fold increase in IN inhibitory activity compared to the
parent peptide.
Truncated Analogs of NL-6 and NL-9. Further structure-activity studies were
conducted on NL-6 and NL-9 to determine the minimal sequence required for
inhibitory activity against IN. As shown in Table C9, NL-6 was truncated into three
hexamers (NL-6-4, NL6-5, NL6-6). Interestingly, the core sequence NL6-5, i.e.
YFLLKL showed comparable activity with respect to the parent peptide NL-6,
which is consistent with the alanine scanning results. The shortened peptide of NL6-
5 containing Y
3
, F
4
, L
5
and L
8
–four key residues out of five—in NL-6, was equally
potent as the dodecamer peptide NL-6.

Table C9. The HIV-1 Integrase Inhibitory Activity of the Truncated Analogs of NL6 and NL-9

Compound Sequence
3‘-Processing
(IC
50,
μM)
Strand Transfer
(IC
50,
μM)
NL6
TAYFLLKLAGRW
21±7 2.7±1
NL 6-4
TAYFLL
500 500
NL 6-5
YFLLKL
20 20
NL 6-6
KLAGRW
> 100 > 100
NL-9
ACWWGAKIQEF
95±9 56±5
NL 9-5
ACWWAG
> 100 60
NL 9-6
WAGIKQ
> 100 > 100
NL 9-7
IKQEF
> 100 > 100

The above studies on the alanine mutation showed that the alanine substitution on the
residues outside of the core sequence of NL-6 resulted in a loss of inhibitory
potency, but the truncated analog without the residues outside of the core sequence

176
still showed activity. We propose that the single substituted analogs might possess
different conformation from the truncated one. Because the side-chain of the
essential residues played an important role in the binding with the protein, the
changed structure of the side-chain might distort the original conformation of the
dodecamer peptide to some extent. For the truncated analog, the retention of the core
sequence might help the retention of active conformation in the absence of the extra
residues. However, NL-9 lost all activity when was truncated into three segments.
Only the N-terminal part of NL-9, i.e. peptide NL9-5, bearing the core motif of
ACWW is slightly active against the strand transfer reaction of IN.
Conformational Modification of NL-6 and NL-9. The dimeric nature of the
catalytic domain of HIV-1 IN suggested that peptides deriving from the dimerization
interface of the enzyme might function as specific inhibitors of the enzyme by
preventing its self-association. Structural data indicate that the interfacial region of
dimeric HIV-1 IN involves strong helix-to-helix contacts α1: α5’ and α5: α1’, where
both hydrophobic and electrostatic interactions contribute to dimer stabilization.
38, 153

Our active peptides NL-6 and NL-9 fall within the sequence of α1 (residues 95-109)
and α3 (residues 123-133), respectively. In order to explore the conformational
requirement, we designed D-peptides composed of all-D amino acid residues of NL-
6 and NL-9 (peptides DNL-6 and DNL-9). The retro isomers of NL-6 and NL-9
(peptide RNL-6 and RNL-9) were also synthesized, with the sequence of amino
acids from the N- to the C-terminus identical to the sequence of NL-6 and NL-9
reading from the C- to N-terminus. The retroinverso isomers of NL-6 and NL-9

177
(peptide DRNL-6 and DRNL-9) possess both the direction of the sequence and the
chirality of the amino acids opposite to those of NL-6 and NL-9. The all-D-, retro-
and retro-inverso peptides belong to the topochemistry, which aims at the change of
whole conformation and configuratiom when the backbone is modified, serving as a
major tool in probing biomolecular topology.
26, 51, 127
There are two pairs of
enantiomers: the natural peptide and the D-peptide, the retro-peptide and the
retroinverso analog. The peptide and its retroinverso isomer are equivalences in
topochemistry (Table C10).
The helical propensity of peptides NL-6 and its mirror image isomer, retro isomer,
retroinverso isomer were measured by circular dichroism (CD) with 50 v/v % of
2,2,2-trifluoroethanol (TFE), an α-helix-stabilizing solvent (Figure C7). The CD
spectra of these peptides showed regular cotton effect. NL-6 and RNL-6 exhibited
positive cotton effect at 192 nm and negative cotton effect at 208 and 222 nm. DNL-
6 and RDNL-6 exhibited negative cotton effect at 192 nm and positive cotton effect
at 208 and 222 nm.
Figure C7. The CD spectra of NL-6 (100 μM) and its isomers in 10 mM PBS, pH 7.2 / 50 % (v/v)
TFE at room temperature.

178
All the CD spectra showed distinct maxima and minima respectively at 208 and 222
nm, consistent with induction of an α-helical conformation. NL-6 and RNL-6
preferentially form right-handed α-helix, while DNL-6 and  RDNL-6 form left-
handed α-helix, consistent with their enantiomeric relationship. Table C10 shows the
calculated percentage of the α-helix forms from the CD data for peptides NL-6,
DNL-6, RNL-6 and RDNL-6 in 50% TFE.
53

Table C10. Calculated content of α-helix of NL-6 and its isomers from CD data based on the method
reported previously.
53

Peptide % α-helix
NL-6

46
RNL-6 46
DNL-6 25
RDNL-6 47

We evaluated the inhibitory activity of the isomers against IN. As shown in Table
C11, all three isomeric forms of NL-9 completely lost inhibitory effect compared to
the parent peptide (Figure C8.A). Conversely, the isomers of NL-6 all show
inhibitory activity to varying extents as compared to the parent peptide. The D-
isomeric form shows a higher micromolar IC
50
value (96±2 µM) for 3’-processing,
but an almost six-fold decrease in IC
50
value for integration (16±4 µM). In a similar
pattern of selectivity for integration, the retro-isomer shows an IC
50
value of 65±8
µM for 3’-processing and 13±1 µM for integration. The retroinverso isomer RDNL-6
has the best IN inhibitory activity with IC
50
values of 3.5±1 µM and 4±1 µM for 3’-

179
processing and integration, respectively showing a six-fold increase in potency for
3’-processing compared to the parent peptide (Figure C8, Table C11).

Table C11. The HIV-1 Integrase Inhibitory Activity of the Mirror Image Isomer, retro isomer and
retro-inverso isomer of NL-6 and NL-9
Compound Sequence
3‘ Processing
(IC
50,
μM)
Strand Transfer
(IC
50,
μM)
1BNL-6
TAYFLLKLAGRW 21±7 2.7±1
DNL- 6 tayfllklagrw
a
65 ± 8 13 ± 1
RNL- 6 WRGALKLLFYAT 96 ± 2 16 ± 4
RDNL- 6 wrgalkllfyat
a
3.5 ± 1 4.0 ± 1

NL-9 ACWWAGIKQEF 95±9 56±5
DNL- 9 acwwagikqef
a
> 1000 > 1000
RNL -9 FEQKIGAWWCA > 1000 > 1000
RDNL- 9 feqkigawwca
a
> 1000 > 1000

a
Represents D-configured amino acid

Figure C8. Inhibitory activity of analogs of NL6 and NL9. a) Alanine-scanned analogs of parent
peptide NL9 show a decrease in potency when Trp4 is substituted with Ala, and an increase in
potency from the parent peptide when Trp3 is substituted. Lane 1: DNA only, lanes 2-5: 333, 111, 37,
12 µM of NL9-4, lanes 6-9: 333, 111, 37, 12 µM of NL9, lanes 11-14: 333, 111, 37, 12 µM of NL9-3,
lane 10: wild type IN only. b) Isomers of NL6 show varying activity when compared to parent peptide
against wild type IN. Lane 1: DNA only, lane 2: wild type IN only, lanes 3-6: 333, 111, 37, 12 µM of
RNL6, lanes 7-10: 333, 111, 37, 12 µM DNL6, lanes 11-16: 333, 111, 37, 12, 4, 1.3 µM of RDNL6,
lane 17: wild type IN only.

Discussion. HIV-1 IN is a 288-amino acid protein, consisting of three functional
domains: the N-terminus, the catalytic core and the C-terminus (Figure C4 A,B).

180
Previously we compiled a list of all reported IN mutants and compared their in vitro
activities as well as their potential infectivity.
107
On the basis of these results we
wanted to understand the role of each individual amino acid in the context of a 6-10
residue long peptide (a distance of < 20Å). Such peptides are not only useful for
mapping DNA-protein interactions or protein-protein interactions, but also can be
used as drug leads. In our initial attempt we designed 16 peptides each of which
contained at least one conserved residue known to reduce the in vitro activity of IN
or block viral replication (Table C5). The 16 peptides were synthesized using Fmoc
chemistry and purified by reverse-phase HPLC. Assays were carried out by
monitoring both the nucleotide cleavage and the DNA strand transfer (integration)
reactions in an in vitro assay specific for IN (see Chapter 6 for Materials and
Methods).
Viruses with amino acid substitutions at conserved residues shown in Table C5 were
replication incompetent. Moreover, several other residues were also identified that
are not highly conserved but dramatically reduce IN activities and viral spread
(Table C5). The simple explanation could be that these mutated enzymes are not
able to recognize and cleave viral DNA. Alternatively, these mutations cause a
decrease or disruption of interaction between IN and other viral and cellular proteins.
In our effort in elucidating residues important in DNA recognition and enzymatic
activity we applied a novel strategy we call ‘sequence walk’ to identify unique
motifs which could potentially be exploited as drug targets.

181
Peptide NL-6 is derived from the IN enzyme residues 97-108. No IN mutants from
this region have thus far been examined for enzymatic activity, HIV-1 infectivity,
and replication. From our data, however, it is interesting that single substitutions of
certain residues results in a complete loss of inhibitory activity compared to the
parent peptide. The NL6-F4A, NL6-L6A and NL6-W12A substitutions are examples
of this event. An interesting and opposite result is obtained with the NL6-K7I
substitution: the peptide becomes more potent in inhibiting both catalytic activities.
It is possible that this particular residue interrupts DNA binding, and further studies
are underway to investigate the implications of the substitutions, and to eliminate the
possibility of non-specific binding. With regard to peptide NL-9, spanning residues
129-139 on the enzyme, K136 is known to be important in DNA binding.
98
As
expected, significant changes in potency of the peptide occur when this amino acid is
substituted. The NL-9-K8A substitution results in an increase in potency while
substitution of the same lysine residue with arginine results in a complete loss of
activity. Binding studies should reveal the mechanism of action of the peptide. It is
plausible that the peptide is binding DNA and preventing the action of IN, or that it
is binding the enzyme itself. The NL-9 C2A and C2S substitutions also show a
significant loss of peptide inhibitory activity. The corresponding C130 mutants have
been shown to have distinct infective dysfunction
177
. This study, as well as
unpublished results from our group, shows that the C130 residue has an important
role in functional integration of the HIV-1 enzyme. Finally, as mentioned earlier, the
ACWW motif seems significant in enzymatic activity as evidenced by the NL-9

182
W3A, W4A, W3G, and W4G substitutions, which show interesting results against
wild-type IN. The implication of these and other amino acid residues in successful
integration of HIV-1 underscores the importance of studies that identify such
residues and investigates possible roles for them.
The α3 and α1 helices of HIV-1 IN, from which the NL-9 and NL-6 peptides derive,
respectively, are on the dimerization interface, as shown in Figure C5 within the
context of the crystal structure of HIV-1 IN
19, 38, 153
. It is possible that both these
peptides define existing ‘hot spots’ on the enzyme that contain amino acid contacts
critical in essential biological interactions of HIV-1 IN, as illustrated by Figure C9.
The targeting of such regions on an enzyme provides the template for rational design
of successful inhibitors either by high-throughput screening, structure-based methods
or in silico studies
108
. For IN, inhibitors to DNA binding have already been
identified
106
. The discovery of inhibitors targeting the disruption of other IN
interactions that are central to its biological activity will be strengthened by studies
such as this, which narrow down areas of interaction to a few important residues.
The use of peptides as probes to elucidating critical residues is thus underscored.
Indeed, short peptides that bind IN to selectively inhibit one of its catalytic
activities—strand transfer—have been identified
36
. It has been proposed that such
peptides could be used as tools in elucidating the entire three-dimensional structure
of IN.

183

Figure C9. The HIV-1 IN binds several biological molecules including other monomers of IN
(multimerization), ligands like small molecule drugs and peptides, various other viral and cellular
proteins, and DNA. Viral proteins like the matrix protein (MA) and numerous cellular enzymes such
as lens epithelium-derived growth factor (LEDGF), IN interacting protein (INI1), and hRad18, a
ubiquitin-conjugating enzyme involved in post-replication/translation DNA repair, are examples of
proteins with which IN interacts. Inhibitors have been described targeting DNA binding of IN, and
studies such as those described in this paper add to the growing body of information towards targeting
IN multimerization and its other protein-protein interactions.

A comparison of the data in Table C10 and C11 suggested that the inhibitory activity
of these peptides are correlated with the α-helical content, and the orientation of key
residue side chains of in the peptide is an important factor as well. The best peptide
is the retroinverso isomer of NL-6, in which the direction of the sequence is reversed
and the chirality of each amino acid residue is inverted. Accordingly, if a
retroinverso peptide is superimposed onto the parent peptide in an antiparallel
fashion, the overall topology of the side chains is maintained, indicating that the
side-chain orientations of a peptide and its retroinverso isomer are similar in
extended conformations. In addition, the retroinverso peptide possesses the highest
percentage of the α-helix conformation, resulting in the best inhibitory activity

184
against HIV-1 IN. The resistence of retroinverso-peptides towards proteolytic
degradation, as compared to the proteolytically susceptible all-L-peptide, suggests
advantages for the development of novel peptidomimetic IN inhibitors as drug
candidates.
Peptide inhibitors of IN have been previously reviewed.
33
While peptides are
somewhat unlikely candidates for clinical drug trials due to various issues of
stability, delivery, and bioavailability, it must be noted that successful peptide
inhibitors to other HIV-1 targets have been described. The fusion inhibitor Fuzeon is
one such example already in clinical use, with others being developed
158
. Fuzeon is
a peptide that mimics the sequence of the HIV-1 gp41 fusion protein
156
. A peptide
inhibitor to virus assembly was also recently identified via phage-display screening
of several peptides with similar sequences that bound to a single reactive site within
the capsid domain of the Gag protein.
136
In short, potential peptide leads to enzyme
inhibition are viable options in drug discovery and cannot be overlooked.
We have developed an innovative approach to providing active IN inhibitors on the
basis of the IN sequence itself. Our preliminary data indicate that this strategy yields
active leads to peptide-based leads by testing only a very few derivatives. This data
also supports our prediction that this study implicates residues essential to HIV-1 IN
catalytic activity (unpublished). As such, this pertinent information can be used in
the rational design of small-molecule as well as peptide or peptidomimetic inhibitors.
The peptides NL-6 and NL-9 are expected to serve as leads for further SAR study to
find novel IN inhibitors with high selectivity and biological efficacy; and also help to

185
further delineate the molecular mechanisms of IN dimerization, protein-protein
interactions, nucleic acid binding, and drug interactions. By extension, this approach
could be applied to IN interacting proteins and other therapeutic targets. Studies are
underway to assess minimal requirement of reported cellular enzymes that bind to IN
and exploit these as novel drug leads.

186
C.3 Design of cell-permeable nanoneedles as HIV-1 Integrase inhibitors
C.3.1 Introduction
Human immunodeficiency virus type 1 (HIV-1) is the causative agent of AIDS, a
global pandemic in need for improved therapeutics. Successful infection of human
immune system cells requires efficient insertion of viral DNA into the host genome,
an event mediated by HIV-1 integrase (IN) in two catalytic events. The first step, 3’-
processing, involves cleavage of a dinucleotide adjacent to a conserved CA on each
terminus of reverse-transcribed DNA, leaving recessed 3’-hydroxyl groups. In the
subsequent strand transfer step, these hydroxyl groups are subject to nucleophilic
attack and then inserted into host DNA
7
. This viral enzyme, having no mammalian
counterpart, is itself an attractive therapeutic target, with small-molecule inhibitors
entering clinical trials only recently
31, 118
. All catalytic events preceding
integration—from viral fusion, to formation and nuclear localization of a large
nucleoprotein complex termed the pre-integration complex (PIC)—as well as
integration itself, are crucially mediated by protein-protein interactions between
relevant viral proteins and host cell co-factors. IN is known to participate in several
of these functional interactions with cellular proteins such as, among others: Gemin2,
and cyclin-dependent kinase inhibitor p21, and the extensively studied lens-
epithelium derived growth factor (LEDGF/p75)
42
. Selectively targeting these vital
interactions as a therapeutic strategy against HIV-1 using peptidomimetic,
biologically stable analogs of active peptides, or small-molecule inhibitors, is a
rapidly developing field of interest that benefits significantly from novel research

187
into biochemical probes or ‘nano-needles’
4
. Given the powerful ability of the
ubiquitous alpha-helical motif in mediating intracellular protein-protein interactions
across extensive biological pathways, such structurally immutable chemical tools
may have general utility in further clarification and/or selective modulation of such
protein-protein interactions within their native cellular environment.
Previous studies from this group have demonstrated inhibitory activity with peptides
interfering with IN catalysis, as well as inhibition of the key LEDGF/p75-IN
interaction necessary for successful integration. A peptide deriving from the reverse
transcriptase region on the HIV-1 HXB2 Pol sequence showed low micromolar
inhibition of IN activity
173
. Inhibition of the LEDGF/p75 interaction was
demonstrated with both wild-type and mutant IN proteins in vitro by a peptide
derived from the LEDGF/p75 protein sequence containing two IN contact ‘hotspots’
5
. Finally, we used a ‘sequence-walk’ strategy to pinpoint areas of catalytic interest
on wild-type IN protein. Evaluation of the inhibition of IN catalysis by a series of
peptides spanning the IN protein sequence resulted in the identification of two
peptides, NL6 and NL9, with low micromolar IC
50
values for inhibition of HIV-1 IN
catalysis. Each peptide was derived from the α1 and α3 helical domains of the IN
protein, respectively. Alanine scanning on these peptides further pinpointed amino
acid residues that were later proven to be critical for IN dimerization and at least one
known HIV-1 IN protein-protein interaction, the LEDGF/p75-IN interaction
3, 6
.
While these studies all provide excellent examples of the potential of peptides in
drug development, the notorious issues of poor cell permeability, secondary structure

188
instability, and in vivo proteolysis severely compromise peptides as valid therapeutic
agents.
The two most potent inhibitory IN-derived alpha-helical peptides from the
latter study are the focus for this report. By examining the two-domain crystal
structures of IN, we found that the central core dimers are bound together via
interactions of various types and the interfacial region of dimeric IN involves the
strong helix-to-helix contacts α1: α5’ and α5: α1’
25, 152
. Since our peptide NL6 just
falls within the sequence of α1 helical domain, we propose that the α-helix
stabilization on NL6 via a hydrocarbon ‘staple’ might enhance the interfacial
interaction and cell permeability. This strategy, first reported by Walensky, et al.,
was used to generate a series of helical, protease-resistant, cell permeable peptides
derived from the amphipathic helical BH3 region found in members of the family of
BCL-2 proteins. A stapled BH3 domain peptide from the BID protein specifically
activated the apoptotic pathway in leukemia cells and human leukemia xenograft
models
149
. More recently, a stapled p53 peptide was shown to directly modulate
transcriptional activity in hDM2-overexpressing cancer cells by reactivation of the
p53-mediated tumor suppressor pathway
12
. Both approaches apply to cancer disease
models. A proof-of-concept cell-penetrating peptide targeting HIV-1 capsid particle
assembly was also recently described which was shown to permeate the cell
membrane and bind to the C-terminal domain of the HIV-1 capsid protein
175
. Cell
permeability enhancement strategies for peptides include the conjugation of
positively-charged protein-transduction domains such as the Tat-peptide conjugation,

189
the Antennapedia homeodomain of Drosophila, and poly-arginine segments
47, 52
.
Cyclization techniques to enforce the alpha-helical nature of many peptides has also
been demonstrated, with polar cross-links often limiting the cell permeability, and
leaving them open to degradation
175
. The hydrocarbon stapling technique elegantly
combines enhanced cell penetration of structurally stable alpha-helical peptides with
an increased resistance to proteolytic degradation. We thus synthesized a series of
NL6 derived peptides each containing hydrocarbon ‘staples’ of different lengths in
attempt to induce and stabilize their helical conformation and consequently enhance
their cell permeability.
This report identifies three alpha-helix stabilized IN-derived peptides with
selective inhibition of HIV-1 replication in infected cells, with low corresponding
cytotoxicity. The corresponding unstapled peptides do not show inhibition of
replication in vivo, although each pair of peptides has similar activity against IN in
our in vitro assay. These peptides also all inhibit the LEDGF/p75-IN interaction in
vitro.
C.3.2 Results
Side-chain bridging is an effective approach to induce and stabilize the α-helix
conformation of the linear peptide, and the all-hydrocarbon bridge enables metabolic
stability and cell membrane permeability. Incorporation of α,α-disubstituted olefinic
amino acids at i and i+4 positions of NL-6 sequence followed by ruthenium-
catalyzed ring-closing metathesis (RCM) reaction generated hydrocarbon stapled
peptides. Based on our previous alanine scanning results, the incorporated position of

190
α,α-disubstituted amino acid was chosen to avoid the essential residues for retaining
effective interaction with IN.. As shown in Figure C11, three types of all
hydrocarbon chain-bridged cyclic peptides were synthesized, with N-terminal, the
middle and C-terminal incorporation, respectively, to determine which position is
favorable for the formation of the α-helix. The CD spectra of the conformationally
constrained peptides were obtained in two solvent systems: CH
3
CN / PBS (40% v/v)
and TFE / PBS (50% v/v) and the final CD spectra are shown in (Figure C12)
68
. The
α-helical content was calculated from CD data based on the method reported
previously (Table C12)
53
.
Fig. C10. Optional positions on NL-6 to incorporate the α,α-disubstituted amino acids for side chain
bridging. (the red letters represent the essential residues for the IN dimerization).

As expected, the flexible peptide NL-6 possessed two times more α-helix content in
CH
3
CN than in TFE, while the conformationally constrained peptides showed slight
difference between the two solvent systems in terms of the α-helix content, which
implies that the hydrocarbon stapled peptide achieved conformational stabilization.
Furthermore, the inserted position of α,α-disubstituted amino acid displayed an
important factor on the α-helical conformation, with the middle position being
optimal for the stabilization of α-helix. The chain length of the hydrocarbon bridge

191
was also found to affect the inducing of α-helix conformation. Among the 3-, 4-, 5-
and 6-methylene olefinic side chains, the pentene side chain was the best length with
respect to inducing and stabilizing the α-helix conformation of peptide NL-6.
Table C12. The α-helical content of the peptides with α,α-disubstituted amino acid incorporation
Content of α-Helix
a

Peptide Sequence
TFE
b
/ PBS
c
CH
3
CN/PBS
c

NL-6 41 22
NLH2 18 24
NLH3 8 ND
d

NLH5 38 42
NLH6 31 34
NLH8 55 69
NLH9 53 38
NLH10 30 18
NLH11 21 22
NLH12 29 30
NLH13 18 14

192
Table C12, continued.
NLH14 7 8
NLH15 26 21
NLH16 24 24
NLH17 24 17

Fig.C11. The CD spectra of NL-6 (100 μM) and NLH series in 10 mM PBS, pH 7.2 / 40 % (v/v)
MeCN at room temperature. Experimental ellipticity data were converted into molar ellipticities and
plotted against wavelength after noise suppression according to the manufacturer’s protocol.

P
a.
P The CD spectra of NL-6 and NLH series were recorded in 10 mM PBS (pH 7.2) / 40 %
MeCN (v/v) or 50% TFE (v/v) at room temperature. Peptides were dissolved at a
concentration of ~100 μM
P
b.
P TFE: 2,2,2-trifluoroethanol
P
c.
P PBS: phosphate buffered solution, 10 mM, pH = 7.2
P
d.
P not determined.

193
Alpha-helical peptide pairs show comparable activity against IN catalysis in
vitro. The sequences of each stapled alpha-helix stabilized peptide and its
corresponding unstapled peptide are shown in Table C12, together with IC
50
values
for 3’-processing and strand transfer catalytic activities. Each pair contains a
covalent hydrocarbon staple linking i and i+4 residues. The position of residue i in
each peptide pair varies, with i in the first pair being the first tyrosine residue in the
sequence TAYFLLKLAGRW. In the second peptide pair the position of residue i
moves to the (i+4) of the first peptide pair, and so on. This change in position of the
hydrocarbon linker has an effect on the inhibitory activity of the peptides. The most
active peptide pair is NLH5 (unstapled) and NLH6 (stapled) with an IC
50
values of 9
± 1 µM for 3’-processing and 6 ± 1 µM for strand transfer. The staple on this pair is
positioned linking the leucine residue (i) in the TAYFLLKLAGRW sequence with
the alanine residue at (i+4). The least active peptide pair are NLH8 (unstapled) and
NLH9 (stapled), with IC
50
values above 100 µM for both catalytic activites contain
the hydrocarbon staple linking the lysine (i) with the arginine residue at (i+4).
Peptides with similar staples but derived from the parent sequence ACWWGIKQEF
are not active in the in vitro assay for inhibition of HIV-1 IN catalysis.
Peptide pairs were evaluated in an in vitro assay for inhibition of the
LEDGF/p75-HIV-1 IN interaction. All peptide pairs exhibited inhibition of the
LEDGF/p75-HIV-1 IN interaction, as shown in Table C13. Similar to inhibition of
IN catalysis, peptide pairs all showed comparable IC
50
values. The most active
peptide was NLH16 (stapled) with an IC
50
value of 5 µM. Its corresponding

194
unstapled peptides NLH15 does not show significant inhibition at concentrations
below 100 µM and NLH17 is moderately active at 23 µM. Unstapled peptide NLH2
shows inhibitory activity of 27 µM while its corresponding stapled partner NLH3 is
inactive. Peptide pair of NLH5 and NLH6 are both moderately active at 15 µM, and
peptides NLH13 and NLH14 are also moderately active (20 µM and 14 µM,
respectively).
Table C13. Inhibition of HIV-1 IN catalytic activities by hydrocarbon-stapled and unstapled alpha-
helical peptides

Peptide Sequence
IC
50
for Inhibition of
LEDGF/p75-HIV-1 IN
interaction
NLH2

26
NLH3

>100
NLH5

15
NLH6

14
NLH13

20
NLH14

14
NLH15

>100
NLH16

5
NLH17

23

195
Stapled alpha-helical peptides inhibit HIV-1 replication in cell-based assays, but
their corresponding unstapled partners do not show inhibition. Three peptide
pairs each were tested for inhibition of HIV-1 replication in MT-4 cells. Cytotoxicity
was also evaluated, and a selectivity index was calculated for each peptide as a ratio
of CC
50
/EC
50
. Table C14 shows the EC
50
, CC
50
and selectivity indices for each of
the peptides evaluated in this assay. From this group of pairs, the peptide showing
the best selectivity index of 6 and thus inhibiting replication with the lowest
corresponding cytotoxicity in MT4 cells is NLH3, containing a hydrocarbon staple
linking the tyrosine residue (i) at the beginning of the sequence with a leucine
residue (i+4). Its corresponding unstapled peptide NLH2, does not show antiviral
activity.
Table C14. Inhibition of HIV-1 replication and native cytotoxicity of conformationally constrained
alpha-helical peptides and their corresponding unrestrained peptide

IC 50 ( μM)
Peptide Sequence
3'-
processing
Strand
transfer
EC 50
( μM)
CC 50
( μM)
Selectivity
Index
NLH2 22±8 15±5 5 5 n/a
NLH3 53±15 19±10 4 25 6
NLH5 9±1 6±1 28 28 n/a
NLH6 9±1 6±1 20 50 3
NLH8 > 100 > 100 NT
a
NT
NLH9 > 100 > 100 NT NT

196
Peptides containing ‘open’ hydrocarbon linkers show cytotoxicity, but the
corresponding peptide containing the ‘closed’ staple inhibits HIV-1 replication
effectively. We synthesized peptides NLH13, NLH15, and NLH17 with the same
parent sequence TAYFLLKLAGRW containing ‘open’ hydrocarbon linkers to
determine whether inhibition of replication is due to retained helicity from the
conformational constraint or due to the hydrocarbon staple itself.
Structures of these peptides are shown in Table C13, with corresponding IC
50
, EC
50

and CC
50
values. All of these peptides show cytotoxicity in MT4 cells.
Table C15. Inhibition of HIV-1 IN catalytic activity and replication by peptides containing ‘open’ or
‘closed’ hydrocarbon linkers

IC
50
( μM)
Peptide Sequence
3'-
processing
Strand
transfer
EC
50

( μM)
CC
50

( μM)
Selectivity
Index
NLH10 > 333 > 111 NT
a
NT
NLH11 61±12 43±5 NT NT
NLH12 48±10 44±4 NT NT
NLH13 74±11 68±5 >18 18 n/a
NLH14 90 83 >40 40 n/a
NLH15 15±2 11±3 11 18 1.6
NLH16 22±3 25±7 20 50 2.5
NLH17 18±2 15±3 10 10 n/a

197
A peptide with the same sequence and modification as NLH15 but with a closed
hydrocarbon staple is NLH16, which has the best overall selectivity index of 8 from
all of the peptides evaluated HIV-1 replication inhibition, with an EC
50
value of 5 ±
2 µM, and a CC
50
value of 41 ± 15 µM. Interestingly, NLH14 has the same
sequence, and contains the hydrocarbon modification in the same position as
peptides NLH13 (‘open’ linker), NLH5 (unstapled), and NLH6 (stapled, closed).
However, the linker itself appears to cause a loss in both in vitro IN inhibition as
well as in vivo HIV-1 replication and cytotoxicity.
Discussion. This study reports the first example of conformationally constrained
peptides derived from the alpha-helical region of HIV-1 IN showing selective
inhibition of HIV-1 replication in MT-4 cells. Each peptide has a corresponding
unconstrained partner that does not exhibit the same HIV-1-induced cytopathic
effects. Peptides with the same sequence that contain ‘open’ hydrocarbon linkers in
similar positions tend to show native cytotoxicity.
Our strategy in the synthesis of these peptides was to extrapolate to a cellular
environment the idea established in a previous study that peptides deriving from the
protein of interest may be used to identify domains on the protein that are essential
for catalytic activity, or protein-protein interactions. Figure C13 summarizes this
venture, from the synthesis of the original peptides, through optimization for activity,
and subsequent choice of peptide for helix stabilization. While this idea has proven
successful in our lab in vitro, we sought to extend it in a cellular environment by
synthesizing hydrocarbon-stapled alpha-helical peptides
84
. Such alpha-helix

198
stabilized peptides mimicking the alpha-helical BH3 domain of the BCL-2 protein,
have previously been shown by Walensky, et al. to have enhanced helicity compared
to their unstapled counterparts
150
. These stapled peptides exhibit stability in a
cellular environment as evidenced by their ability to suppress growth of human
leukemia cells in vivo, by activating apoptotic pathways.
High throughput screening followed by synthetic optimization was essential in
identifying HIV-1 IN inhibitors suitable for clinical use
118
. The complex protein-
protein interactions of HIV-1 IN with cellular cofactors provide innovative
therapeutic targets for the development of small-molecule inhibitors against HIV-1;
however, these targets are plagued by difficulties associated with the shallow,
hydrophobic nature of the protein surfaces involved in such interactions
144
. A recent
study by Brass, et al. identified a host of HIV-dependency factors that are required
for HIV-1 infection, using RNAi and forward genetics techniques
16
. We introduce
the concept of stapled optimized peptides as cell-permeable nano-needles that we
propose will enable the identification of regions of interest on enzymes that have
thus far been identified as interacting protein partners not only of IN, but also to
those of viral proteins throughout the lifecycle, from entry to packaging. The ‘nano-
needles’ may themselves serve as a starting point to the development of novel
inhibitors and be further optimized for HIV-1 infection inhibition with formulation
into nanoparticles and immunoliposomes. Taking advantage of an otherwise ironclad
viral resistance mechanism is hypothetically attractive, but still requires much

199
refinement in terms of realistic inhibitor design and development for which we see
our approach a possible starting platform for such ventures.
Fig. C12. Strategy for design of cell permeable nano-needles as probes towards elucidating regions of
interest within a protein that are essential for efficient catalysis or protein-protein interactions.

Abstract (if available)

Abstract APE1 is an attractive target for the rational design of small-molecule inhibitors in the field of oncological therapeutic research. It is an essential enzyme in the mammalian base excision repair pathway, and works in conjunction with other cellular proteins in the repair of abasic sites within the genome. This enzyme has been implicated in the resistance of tumors to current chemotherapeutic agents. Extensive effort is underway in the identification and development of clinically suitable inhibitors to this essential enzyme. Herein, we present the discovery of a series of inhibitors to APE1, as well as the identification of a novel class of bioisosteric compounds that inhibit the catalytic activity of the enzyme. We have also performed preliminary work in the development and optimization of biochemical models of APE1 expression regulation. This is an ongoing effort to characterize appropriate cellular systems in colon cancer for the development of cellular APE1 inhibitors that can be measured using valid biochemical endpoints. Interesting observations have been made in this regard, for the first time, also implicating the base excision repair pathway in the repair of platinum-agent induced DNA damage. Finally, as validation for APE1 as a therapeutic target, we studied the impact of single nucleotide polymorphisms in two populations of colorectal cancer patients, and found that the D148E polymorphism confers a measure of survival to males possessing it.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Development of novel small molecules targeting mitochondrial and oxidative stress signaling pathways for pancreatic cancer therapy

PDF

HIV-1 integrase and human APE1: two DNA-processing enzymes and two druggable targets

PDF

Pancreatic cancer: a review on biology, genetics and therapeutics

PDF

Design of novel anticancer agents targeting cellular stress response pathways

PDF

Targeting cellular redox modulations for pancreatic cancer treatment

PDF

Discovery of a novel HIV-1 integrase inhibitor binding site: insight into enzyme structure/function and inhibitor design

PDF

Discovery of novel small molecules targeting cancer cell metabolism

PDF

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

PDF

Discovery of novel small-molecule compounds with antiviral and anticancer properties

PDF

Discovery of novel small molecules for ovarian cancer treatment

PDF

Evaluating novel chemotherapeutic strategies in colorectal and gastric cancer: the role of histone deacetylase inhibitors and human epidermal receptor family inhibitors

PDF

Design and discovery of small molecules inhibiting the interaction of cellular LEDGF/p75 and HIV-1 integrase

PDF

Discovery of small molecules for brain cancer treatment

PDF

Assembling NRPS modules in e. coli to establish a platform for rational design of biologically active compounds

PDF

Investigating CD99 as a therapeutic target in acute myeloid leukemia

PDF

Systematic identification of potential therapeutic targets in cancers using heterogeneous data

PDF

Fasting-based differential stress resistance to enhance cancer treatment: a novel strategy to protect normal cells and sensitize cancer cells to chemotherapy

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

UVRAG protects cells from UV-induced DNA damage by regulating global genomic nucleotide excision repair pathway

PDF

Identification of small-molecules targeting CXCR2 function and signaling

Asset Metadata

Creator Zawahir, Faathma (author)

Core Title Targeting human base excision repair as a novel strategy in cancer therapeutics

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Pharmacy / Pharmaceutical Sciences

Publication Date 05/13/2009

Defense Date 01/27/2009

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

Tag APE1,cancer therapeutics,DNA repair,OAI-PMH Harvest

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Haworth, Ian S. (committee chair), Ladner, Robert D. (committee member), Neamati, Nouri (committee member), Wang, Clay C. C. (committee member)

Creator Email zawahir@usc.edu,zzawahir@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2257

Unique identifier UC1314840

Identifier etd-Zawahir-2663 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-248516 (legacy record id),usctheses-m2257 (legacy record id)

Legacy Identifier etd-Zawahir-2663.pdf

Dmrecord 248516

Document Type Dissertation

Rights Zawahir, Faathma

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu