Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Design of novel anticancer agents targeting cellular stress response pathways
(USC Thesis Other)
Design of novel anticancer agents targeting cellular stress response pathways
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
DESIGN OF NOVEL ANTICANCER AGENTS TARGETING
CELLULAR STRESS RESPONSE PATHWAYS
by
Kavya Ramkumar
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PHARMACEUTICAL SCIENCES)
June 2014
Copyright 2014 Kavya Ramkumar
ii
To my loving family
iii
ACKNOWLEDGEMENTS
I am grateful to numerous people who have helped throughout my doctoral
studies. Foremost, I would like to thank my advisor, Prof. Nouri Neamati. His guidance,
support and encouragement through these years helped me through the challenges in
graduate school.
I am very grateful to all the members of my dissertation committee- Profs.
Bangyan Stiles, Roger Duncan and Curtis Okamoto for their time, support and helpful
feedback in my research. I would like to specially thank Dr. Stiles for being an invaluable
support particularly in the last year. I would also like to thank Prof. Amy Lee for her
guidance in my research. I also thank the faculty of the School of Pharmacy for their
support at various points in my program and in my research.
I thank all my friends and colleagues in the PhD program and in the lab,
especially Drs. Helen Ha, Divya Pathania and Yumna Shabaik for their support and
suggestions (and food!). Working those endless days in lab, brainstorming ideas,
troubleshooting assays, getting over failed western blots, celebrating awards, graduations,
and holidays with them made this journey to PhD so fun and memorable. I would like to
specially thank Dr. Shili Xu, Yuting Kuang, Dr. Bikash Debnath, Dr. Soma Samanta and
Shuzo Tamura for their generous support at Ann Arbor when the lab moved to University
of Michigan. I would specially like to thank Dr. Debnath for his collaboration on the
computational aspects included in this dissertation. I would also like to thank Si Li and
iv
Yuting Kuang for their collaboration with the mice studies. My special thanks to Wade
Harper for being so helpful with all the administrative aspects throughout my program.
Most importantly, I owe a lot to my family for all the sacrifices they have made so
I can accomplish this. I cannot not thank my husband enough for his immeasurable love
and support through these grueling years. This would not have been possible without his
constant encouragement and unwavering determination in helping me complete my
studies against all odds and hardships. I am also ever indebted to my parents, my brother
and my in-laws for helping and supporting me to complete this endeavor. Without their
love and blessings, I could not have done this. Lastly, I thank my little one for
brightening up everyday with her beautiful smile.
v
Table of Contents
ACKNOWLEDGEMENTS ................................................................................................. iii
LIST OF FIGURES ........................................................................................................... ix
LIST OF TABLES .............................................................................................................. xi
ABSTRACT ...................................................................................................................... xii
CHAPTER 1: INTRODUCTION ........................................................................................ 1
CHAPTER 2: GLUTATHIONE S-TRANSFERASE OMEGA 1 .......................................... 4
2.1 GLUTATHIONE S TRANSFERASES ............................................................................... 4
2.2 GSTO1 .................................................................................................................... 5
2.2.1 GSTO1 Structure ............................................................................................. 6
2.2.2 GSTO1 catalytic activities ................................................................................ 8
2.2.3 Functions of GSTO1 ........................................................................................ 9
2.2.4 GSTO1 in redox homeostasis and stress response signaling ....................... 10
2.2.5 GSTO1 in diseases ........................................................................................ 12
CHAPTER 3: VALIDATION OF GSTO1 AS A NOVEL ANTICANCER DRUG TARGET 13
3.1 GSTO1 IN CANCER ................................................................................................. 13
3.2 GSTO1 IN DRUG RESISTANCE ................................................................................. 13
3.3 MATERIALS AND METHODS ...................................................................................... 14
3.4 RESULTS ................................................................................................................ 16
3.4.1 GSTO1 is overexpressed in many cancers .................................................... 16
3.4.2 GSTO1 knockdown inhibits cancer cell proliferation ...................................... 20
3.3 DISCUSSION ............................................................................................................ 22
CHAPTER 4: IDENTIFICATION OF NOVEL GSTO1 INHIBITORS ............................... 23
4.1 BACKGROUND ......................................................................................................... 23
4.2 MATERIALS AND METHODS ...................................................................................... 24
4.3 RESULTS ................................................................................................................ 29
4.3.1 Optimization of high throughput activity assay ............................................... 29
4.3.2 Optimization of competitive binding assay ..................................................... 30
4.3.3 Identification of GSTO1 inhibitors .................................................................. 31
Design basis .................................................................................................................................... 31
Pilot screening ................................................................................................................................. 31
SAR studies and Lead identification ................................................................................................ 33
4.3.4 Characterization of lead GSTO1 inhibitors ..................................................... 52
C1-27 shows covalent binding to active site cysteine ..................................................................... 52
C1-27 shows rapid cellular uptake and GSTO1 binding .................................................................. 53
C1-27 shows target occupancy up to 10 h ...................................................................................... 53
GSTO1 inhibition by C1-27 is reversible ......................................................................................... 54
C1-27 shows selectivity in GSTO1 inhibition ................................................................................... 56
vi
4.4 DISCUSSION ............................................................................................................ 57
CHAPTER 5: EVALUATION OF GSTO1 INHIBITORS AS NOVEL ANTICANCER
AGENTS ......................................................................................................................... 59
5.1 BACKGROUND ......................................................................................................... 59
5.2 MATERIALS AND METHODS ...................................................................................... 59
5.3 RESULTS ................................................................................................................ 63
5.3.1 GSTO1 inhibitors block cancer cell proliferation ............................................ 63
5.3.2 GSTO1 inhibitor C1-27 induces apoptotic cell death ..................................... 66
5.3.3 Cell death induced by GSTO1 inhibitor C1-27 is Ras-signaling dependent ... 66
5.3.4 Effect of iron chelator on C1-27 induced cell death ....................................... 68
5.3.5 GSTO1 inhibition enhances cisplatin cytotoxicity ........................................... 69
5.3.6 C1-27 inhibits human colon cancer xenograft growth in vivo ......................... 70
5.4 DISCUSSION ............................................................................................................ 72
CHAPTER 6: EFFECT OF GSTO1 INHIBITORS ON IL-1 β SECRETION ...................... 74
6.1 BACKGROUND ......................................................................................................... 74
6.2 MATERIALS AND METHODS ...................................................................................... 76
6.3 RESULTS ................................................................................................................ 77
6.3.1 GSTO1 inhibitors block IL-1 β secretion from activated THP-1 cells .............. 77
6.3.2 Co-expression of GSTO1 and IL-1 β in cancer ............................................... 79
6.4 DISCUSSION ............................................................................................................ 81
CHAPTER 7: 78 kDa GLUCOSE REGULATED PROTEIN AS A NOVEL ANTICANCER
TARGET ......................................................................................................................... 83
7.1 INTRODUCTION ........................................................................................................ 83
7.1.1 GRP78 Structure ............................................................................................ 85
7.2 GRP78 AS A REGULATOR OF THE UNFOLDED PROTEIN RESPONSE ............................ 87
7.3 GRP78 AS A NOVEL ANTICANCER TARGET ............................................................... 94
7.3.1 GRP78 in cancer proliferation and progression ............................................. 94
7.3.2 GRP78 in drug resistance .............................................................................. 96
7.3.3 GRP78 in breast cancer ................................................................................. 98
7.4 GRP78 INHIBITION ................................................................................................ 101
7.4.1 GRP78 inhibitors in literature ....................................................................... 102
CHAPTER 8: IDENTIFICATION OF NOVEL THIENO[2,3-d]PYRIMIDIN-4(1H)-ONE-
BASED GRP78 INHIBITORS BY PHARMACOPHORE-BASED VIRTUAL SCREENING
...................................................................................................................................... 107
8.1 INTRODUCTION ...................................................................................................... 107
8.2 RESULTS .............................................................................................................. 108
8.2.1 Pharmacophore modeling ............................................................................ 108
8.2.2 Hit identification ............................................................................................ 111
8.2.3 Hit optimization through substructure-similarity search ................................ 113
8.2.5 240 binding to GRP78 .................................................................................. 117
vii
8.2.6 240 inhibits GRP78-mediated chaperone refolding ..................................... 118
8.2.7 240 induces ER stress ................................................................................. 118
8.2.8 Treatment with 240 causes degradation of HSP70 and HSP90 client proteins
.............................................................................................................................. 121
8.2.9 240 induces cancer cell death ...................................................................... 122
8.2.10 240 induces a nonapoptotic cell death. ...................................................... 124
8.2.11 240 shows inhibition of tumor growth in MDA-MB-468 breast xenograft
model. ................................................................................................................... 127
8.3 MATERIALS AND METHODS .................................................................................... 128
8.4 DISCUSSION ...................................................................................................... 136
CHAPTER 9: IDENTIFICATION OF NOVEL N
2
,N
4
-DIPHENYL-1,3,5-TRIAZINE-2,4,6-
TRIAMINE-BASED GRP78 INHIBITORS BY PHARMACOPHORE-BASED VIRTUAL
SCREENING ................................................................................................................. 139
9.1 INTRODUCTION ...................................................................................................... 139
9.2 RESULTS .............................................................................................................. 141
9.2.1 Pharmacophore modeling ............................................................................ 141
9.2.2 Substructure-similarity search ...................................................................... 143
9.2.3 Structure activity relationship ....................................................................... 143
9.2.4 Inhibition of chaperone activity ..................................................................... 147
9.2.5 249 and 781 induce ER stress ..................................................................... 148
9.2.6 781 causes degradation of HSP70 client proteins ....................................... 149
9.2.7 781 induces cancer cell death ...................................................................... 150
9.2.8 781 inhibits tumor growth in a breast cancer xenograft model ..................... 151
Figure 9.8 In vivo efficacy of 781 in MDA-MB-231 breast cancer xenograft. ........ 152
9.3 MATERIALS AND METHODS .................................................................................... 154
9.3.1 Generation of pharmacophore hypothesis and database search ................ 154
9.3.2 Substructure and similarity search ............................................................... 154
9.3.3 Xenograft studies ......................................................................................... 155
9.4 DISCUSSION .......................................................................................................... 156
CHAPTER 10: IN VITRO AND IN VIVO ANTICANCER EFFECTS OF A NOVEL GRP78
INHIBITOR WITH A QUINOLINE SCAFFOLD ............................................................. 157
10.1 RESULTS ............................................................................................................ 157
10.1.1 Identification of quinolone-based compounds through high throughput
screening ............................................................................................................... 157
10.1.2 Structure activity relationship ..................................................................... 159
10.1.2 148 and 170 induce ER stress ................................................................... 161
10.1.3 148 disrupts HSP70/HSP90 chaperone machinery and induces client protein
degradation. .......................................................................................................... 163
10.1.4 148 inhibits cell survival and induces cancer cell death in a panel of breast
cancer cell lines. .................................................................................................... 164
10.1.5 Mechanism of cell death ............................................................................ 166
viii
10.1.6 148 impedes cancer growth in a breast cancer xenograft model ............... 167
10.2 MATERIALS AND METHODS .................................................................................. 171
10.2.1 ATPase assay ............................................................................................ 171
10.2.2 Counter-screening against FGFR kinase ................................................... 171
10.2.3 Xenograft studies ....................................................................................... 172
10.3 DISCUSSION ........................................................................................................ 173
CHAPTER 11: CONCLUSIONS AND FUTURE DIRECTIONS .................................... 175
BIBLIOGRAPHY ........................................................................................................... 180
Appendix ....................................................................................................................... 202
A1. INITIAL GSTO1 HIT, 4A (401) INHIBITS CANCER CELL PROLIFERATION. ................... 202
A2. TIME COURSE OF 4A CYTOTOXICITY. ...................................................................... 203
A3. EFFECT OF 4A ON CELL SIGNALING PATHWAYS. ...................................................... 204
A4. 4A INDUCES ER STRESS. ...................................................................................... 205
A5. 4A ENHANCES CISPLATIN INDUCED CYTOTOXICITY. ................................................. 206
A7. GSTO1 INHIBITOR C2-22 INHIBITS IL-1 β SECRETION FROM ACTIVATED MONOCYTES.
.................................................................................................................................. 208
ix
LIST OF FIGURES
Figure 2.1 Alignment of GSTO1, GSTO2 and Glutaredoxin (GRX) sequences showing
sequence identity. ............................................................................................................. 5
Figure 2.2. Crystal structure of GSTO1 (PDB ID: 1EEM) showing key structural features.
.......................................................................................................................................... 7
Figure 3.1 GSTO1 is overexpressed in several cancers. ............................................... 16
Figure 3.2 GSTO1 knockdown induces cancer cell death. ............................................. 21
Figure 4.1 Structures of GSTO1 inhibitors reported in literature. .................................... 23
Figure 4.2 GSTO1 substrate assay optimization. ........................................................... 29
Figure 4.3. Optimization of GSTO1 competitive binding assay. ..................................... 30
Figure 4.4 Schematic of inhibitor identification strategy .................................................. 33
Figure 4.5 Cluster 1 - N-Phenyl-2
0
-chloroacetamide ...................................................... 34
Figure 4.6 Cluster 2 - N-Heterocycle-2°-chloroacetamide ............................................. 38
Figure 4.7 Cluster 3 - N-Linker-2
0
-chloroacetamide ....................................................... 41
Figure 4.8 Cluster 4 - Non-cyclized 3
0
-chloroacetamide ................................................. 44
Figure 4.9 Cluster 5 - Cyclized 3
0
-chloroacetamide ........................................................ 46
Figure 4.10 Identification and characterization of C1-27 as a potent GSTO1 inhibitor. .. 50
Figure 4.11 C1-27 binds to GSTO1 and promotes protein destabilization. .................... 51
Figure 4.12 Binding pose of C1-27 determined by crystallographic studies. .................. 52
Figure 4.13 C1-27 rapidly binds to GSTO1. .................................................................... 53
Figure 4.14 Duration of GSTO1 inhibition by C1-27. ...................................................... 54
Figure 4.15 Reversibility of GSTO1 inhibition by C1-27. ................................................ 55
Figure 4.16 Selectivity of GSTO1 inhibition. ................................................................... 56
Figure 5.1 Inhibition of cancer cell proliferation by GSTO1 inhibitors. ............................ 64
Figure 5.2 C1-27 induces an apoptotic cell death. .......................................................... 66
Figure 5.3 Cell death induced by C1-27 is Ras-signaling dependent. ............................ 68
Figure 5.4 Enhancement of C1-27-mediated cell death by DFO. ................................... 69
Figure 5.5 C1-27 enhances cisplatin-induced cytotoxicity. ............................................. 70
Figure 5.6. GSTO1 inhibitor C1-27 inhibits tumor growth in vivo. ................................... 71
Figure 6.2 GSTO1 inhibitors suppress IL-1 β secretion. .................................................. 78
Figure 6.3. IL-1 β expression strongly correlates with GSTO1 expression. ..................... 80
x
Figure 7.1 GRP78 domain structure ............................................................................... 86
Figure 7.2 UPR as an emerging drug target ................................................................... 88
Figure 7.3 GRP78 expression in a panel of breast cancer cell lines. ............................. 98
Figure 7.4 GRP78 knockdown reduces breast cancer cell viability ................................ 99
Figure 8.1 Five-feature Pharmacophore model and mapping of GRP78 inhibitors. ..... 110
Figure 8.2 Hit identification and optimization. ............................................................... 112
Figure 8.3 240 binds to GRP78. ................................................................................... 118
Figure 8.3 Inhibition of GRP78 chaperone function by 240. ......................................... 118
Figure 8.4 240 treatment induces ER stress in breast cancer cells. ............................. 120
Figure 8.5 240 treatment induces degradation of HSP70/HSP90 client proteins. ........ 122
Figure 8.6 240 preferentially induces cancer cell death. ............................................... 123
Figure 8.7 240 induces a nonapoptotic, autophagic and necroptotic cell death. .......... 126
Figure 8.8 240 shows tumor growth inhibition up to day 28. ......................................... 127
Figure 9.1 Pharmacophore mapping onto GRP78 inhibitors (scaffold B). .................... 142
Figure 9.2 249 and related analogue, 781 inhibit GRP78 ATPase activity. .................. 146
Figure 9.3 Inhibition of GRP78 chaperone activity by 249. ........................................... 147
Figure 9.4 ER stress induction and UPR activation by 249 and 781. ........................... 148
Figure 9.5 781 depletes HSP90 client proteins without HSP70 induction. .................. 149
Figure 9.6 Cytotoxicity profile of 781 in cancer versus nontransformed cells. .............. 150
Figure 9.9 781 treatment did not cause any significant systemic toxicity. .................... 153
Figure 10.1 Identification of GRP78 ATPase inhibitor. ................................................. 158
Figure 10.2 148 and 170 induce ER stress in breast cancer cells. ............................... 162
Figure 10.3 Effect of 148 treatment on abundance of HSP70/HSP90 client proteins. .. 164
Figure 10.4 Reduction of colony forming ability of MCF7 cells following 148 exposure.
...................................................................................................................................... 165
Figure 10.5 Mechanism of cell death induced by 148. .................................................. 167
Figure 10.6 GRP78 inhibitor 148 inhibits tumor growth in vivo. .................................... 169
Figure 10.7 148 treatment did not cause any significant systemic toxicity. .................. 170
Figure 11.1 First-in-class GSTO1 inhibitors target key hallmarks of cancer. ................ 178
xi
LIST OF TABLES
Table 2.1 GSTO catalytic activities ................................................................................... 9
Table 3.1. GSTO1 expression in Oncomine gene expression studies ........................... 18
Table 4.1 Results of pilot screening ................................................................................ 32
Table 4.2. Screening data for Cluster-1 - N-Phenyl-2
0
-chloroacetamide ....................... 36
Table 4.3 Screening data for Cluster-2 - N-Heterocycle-2
0
-chloroacetamide ................. 39
Table 4.4 Screening data for Cluster-3 - N-Linker-2
0
-chloroacetamide .......................... 42
Table 4.5 Screening data for Cluster-4 - Non-cyclized 3
0
-chloroacetamide ................... 45
Table 4.6 Screening data for Cluster-5 - Cyclized 3
0
-chloroacetamide .......................... 47
Table 4.7: Inhibitory activities of selected GSTO inhibitors ............................................. 49
Table 5.1: Cytotoxicity of selected GSTO inhibitors in a panel of cancer cell lines ........ 65
Table 5.2. Effect of C1-27 treatment in a panel of cancer cell lines. ............................... 67
Table 7.1 HSP70 family of chaperones. BLASTp results of some of the HSP70 proteins
against human HSPA5 sequence. .................................................................................. 84
Table 7.2 Recent advances in inhibitors targeting the UPR pathway ............................. 90
Table 7.3. GRP78 in tumor progression – Summary of knockout studies ...................... 95
Table 7.4 GRP78 and chemoresistance in cancer ......................................................... 97
Table 7.5 Compounds targeting GRP78 ....................................................................... 104
Table 8.1. Inhibitory activity of compounds ................................................................... 115
Table 8.2. Inhibitory activity of compounds ................................................................... 116
Table 8.3. Inhibitory activity of compounds ................................................................... 116
Table 9.1 Biochemical activities and SAR of 249 and analogues ................................. 144
Table 9.2. Biochemical activities and SAR of 249 and analogues ................................ 145
Table 9.3 Biochemical activities and SAR of 430 and analogues ................................. 146
Table 9.4 Cytotoxicity of 781 in a panel of human breast cancer cell lines .................. 150
Table 10.1 Inhibitory activities of active compounds 148 and 170 ................................ 158
Table 10.2 Biochemical activities and SAR of 148 and analogues ............................... 159
Table 10.3 Biochemical activities and SAR of 148 and analogues ............................... 160
Table 10.4 Biochemical activities and SAR of 148 and analogues ............................... 161
Table 10.5 Cytotoxicity of 148 in a panel of human breast cancer cell lines ................ 165
xii
ABSTRACT
Cancer cells exploit cellular stress response mechanisms, such as the unfolded
protein response and oxidative stress response, to cope with unfavorable growth
conditions and to survive through cytotoxic treatments. Cellular stress response pathways
are therefore attractive targets for developing novel anticancer agents. We have identified
novel small-molecule compounds that target two players of the cellular stress response
pathways – Glutathione S transferase omega 1 (GSTO1) and the 78 kDa glucose-
regulated protein (GRP78).
GSTO1 is an atypical GST isoform, with an active-site cysteine, that is involved
in oxidative stress response. GSTO1 is overexpressed in several cancers and has been
implicated in drug resistance. We show that silencing GSTO1 significantly impairs
cancer cell survival and proliferation validating GSTO1 as a new target in oncology.
Through extensive screening and hit optimization, we have identified a series of
chloroacetamide-containing small-molecule compounds that are potent GSTO1 inhibitors
with activity in the nanomolar range. Crystal structures and biochemical studies revealed
a unique reversible covalent binding to the active site cysteine. Potent GSTO1 inhibitors
suppressed the growth of cancer cells, enhanced the cytotoxic effects of cisplatin, and
inhibited tumor growth without apparent systemic toxicity. GSTO1 inhibitors also
blocked IL-1β secretion from activated monocytes. Our findings demonstrate the
therapeutic utility of selective GSTO1 inhibitors as anticancer and anti-inflammatory
agents.
xiii
We also have identified small-molecule inhibitors of the 78 kDa glucose regulated
protein (GRP78) chaperone - a key mediator of the unfolded protein response. As a
molecular chaperone belonging to the HSP70 family, GRP78 regulates protein quality
control and relieves proteotoxic stress. In addition to providing a growth advantage to
cancer cells, GRP78 induction can lead to drug resistance. Suppressing GRP78 levels
using siRNA has been shown to slow cancer cell progression and overcome drug
resistance. Another approach to inactivating GRP78 is by inhibiting its ATPase activity
that has not received much attention. Here, we have identified novel classes of small-
molecule inhibitors of GRP78 ATPase activity through a ligand-based drug design
approach. Based on the initial hits, we further optimized several analogues with more
potent activity for GRP78 inhibition. The lead compounds induced CHOP significantly
under conditions of ER stress and activated autophagy. These compounds induced cancer
cell death in a panel of breast cancer cell lines and increased sensitivity to
chemotherapeutic agents, further supporting the role for GRP78 in cancer survival and
drug resistance. The lead compounds also caused degradation of HSP70 client proteins.
In vivo studies in a breast cancer xenograft mice model further demonstrated efficacy of
these compounds in impeding tumor growth alone and in combination with standard
chemotherapeutic agent, doxorubicin. Taken together, compounds identified in this
dissertation provide novel approaches to target cellular stress pathways in cancer and will
serve as initial leads for further optimization into more potent compounds.
1
CHAPTER 1: INTRODUCTION
Cancer cells are often characterized by tumorigenic traits such as sustained
proliferation, resisting cell death, replicative immortality, etc., which are further
facilitated by tumor-promoting conditions in the tumor microenvironment. These
hallmarks of cancer are outlined in detail in the seminal review by Hanahan and
Weinberg (Hanahan and Weinberg, 2000, 2011). Many of these cancer traits arise from
alterations at the genetic level such as chromosomal instability, mutations, amplification
or overexpression of oncogenes or from loss of tumor suppressors and checkpoints etc.
Oncogenes often drive the cancer phenotype and are considered ideal targets for
molecular therapeutics because of the dependence of cancer cells on them (oncogene
addiction) (Weinstein and Joe, 2008).
However, as a result of oncogenic transformation, cancer cells are constantly
under elevated cellular stress to meet the demands of tumor progression. Therefore, in
addition to the classical hallmarks that are necessary for tumor initiation and progression,
cancers are also characterized by the prevalence of cellular stress phenotypes such as
oxidative stress, proteotoxic stress, metabolic stress, mitotic stress that arise as a result of
sustained tumor progression (Luo et al., 2009). Consequently, cancer cells depend on
normal cellular stress response processes, which are not inherently oncogenic, for their
survival (non-oncogene addiction) (Luo et al., 2009; Solimini et al., 2007). For example,
cancer cells show increased levels of reactive oxygen species, arising from increased
2
metabolic needs, mitochondrial damage, oncogenic stimulation etc. This results in
oxidative damage to proteins, DNA and lipids (Fiaschi and Chiarugi, 2012). To alleviate
oxidative stress, cancer cells exploit several ROS-activated cell signaling pathways and
the antioxidant system (Engel and Evens, 2006; Hayes and McLellan, 1999; Martindale
and Holbrook, 2002). Another example is the proteotoxic stress response. The protein
quality control and proteostasis in a cell are largely regulated by two players – the
chaperones (protein folding machinery) and the proteasome (protein degradation
machinery). Cancer cells have an increased need for protein synthesis, proper folding and
clearance to sustain rapid proliferation and tumor growth. Together with mutations and
oxidative damage to proteins, cancer cells are under constant proteotoxic stress and are
dependent on the cellular protein quality control system (Dejeans et al., 2014; Workman
et al., 2007).
The increased dependence of cancer cells on cellular stress response pathways can
be exploited for therapy in two ways – by inhibiting the stress response pathway so the
cancer cells can no longer cope with oncogenic stress (Stress sensitization) or by
overwhelming the stress response pathway with exacerbated oncogenic stress (Stress
overload) (Luo et al., 2009). In the following chapters, we will describe our drug design
efforts to target two of these cellular stress response pathways to develop novel
anticancer agents. In the first approach, we designed inhibitors of glutathione S
transferase omega 1 (GSTO1), an emerging player in the oxidative stress response
pathway, to sensitize cancer cells to redox stress. In the second approach, we identified
inhibitors of the chaperone proteins, 78 kDa glucose regulated protein (GRP78) and 70
3
kDa heat shock protein (HSP70), disrupting the protein folding machinery, to sensitize
and overload the cancer cells with proteotoxic stress. We report that these novel
inhibitors of GSTO1 and GRP78 hamper cancer cell survival and impede tumor growth,
further validating these proteins and their cellular stress response as promising drug
targets.
4
CHAPTER 2: GLUTATHIONE S-TRANSFERASE OMEGA
1
2.1 Glutathione S transferases
Glutathione S transferases (GSTs) are a diverse family of cytosolic, mitochondrial
and microsomal enzymes that are primarily involved in phase II metabolism. The
mammalian cytosolic GSTs are classified into 7 different classes – alpha, mu, pi, theta,
sigma, zeta and omega, which are evolutionarily distinct. They carry out xenobiotic
detoxification through glutathione conjugation. There is some degree of overlap in
substrates metabolized by the different GST classes while some are highly specific. In
general, most GST isozymes can metabolize 1-chloro-2,4-dinitrobenze (CDNB) (Hayes
et al., 2005).
In addition to xenobiotic metabolism, GSTs play a role in the synthesis and
metabolism of endogenous compounds, redox homeostasis, and cellular signaling (Hayes
et al., 2005; Pajaud et al., 2012). GSTs such as GSTP1, GSTA1, and GSTM1 have also
been shown to regulate cell proliferation and apoptosis through protein-protein
interactions with MAP kinases such as JNK and ASK1 (Board and Menon, 2013;
Laborde, 2010). GST overexpression and polymorphisms are seen in several cancers as
well as in other diseases, and have been implicated in resistance to chemotherapy
(McIlwain et al., 2006; Townsend and Tew, 2003). As a result, GSTs are an attractive
drug target. Several small-molecule inhibitors against GST isoforms have been designed
as anticancer agents. These include compounds that inhibit GST activity, disrupt GST-
5
protein interactions, or are GST-activated pro-drugs or glutathione analogs (Mahajan and
Atkins, 2005; Townsend and Tew, 2003).
2.2 GSTO1
In 2000, Board et al. reported a unique GST isoform through a sequence-based
search of the human EST database (Board et al., 2000). This isoform, GST omega
(GSTO) exhibited a low sequence similarity (< 20%) with the other GST isoforms and,
unlike the other GST isoforms, had a cysteine residue at its active site. Similar sequences
in mice and rat were reported earlier but thought to be member of GST theta class or
dehydroascorbate reductase enzymes (Ishikawa et al., 1998; Kodym et al., 1999). In
humans, two GSTO genes encoding GSTO1 and GSTO2 enzymes have been identified.
GSTO1 (241 amino acids) shares 64% sequence identity with GSTO2 (243 amino acids)
(Fig. 2.1).
Figure 2.1 Alignment of GSTO1, GSTO2 and Glutaredoxin (GRX) sequences
showing sequence identity.
57
57
51
117
117
69
176
177
86
231
232
106
241
243
GSTO1
GSTO2
GRX
GSTO1
GSTO2
GRX
GSTO1
GSTO2
GRX
GSTO1
GSTO2
GRX
GSTO1
GSTO2
GRX
32
6
2.2.1 GSTO1 Structure
Although GSTO1 shares only 20% sequence identity with other GST isozymes,
its crystal structure (PDB ID: 1EEM) shows that it still adopts the canonical GST fold -
an N-terminus thioredoxin like domain with a central β-sheet flanked by 2 α-helices and
a 3
10
helix, and a C-terminus composed of 7 α-helices (Board et al., 2000). In addition,
GSTO1 has some unique structural attributes. It has a proline-rich 19-residue N-terminal
extension that forms a unique interface with the last 2 α-helices of the C-terminus
domain. The N-terminal extension also contains a PXXP motif, which could be a binding
site for SH3 domain proteins. Finally, at its active site, GSTO1 has a cysteine residue
(C32) instead of the characteristic tyrosine or serine as in other GST isoforms. Typically,
the hydroxyl group in tyrosine or serine stabilizes the thiolate ion in glutathione molecule
through hydrogen bonding interactions mediated through water molecules. In GSTO1,
C32 forms a disulfide bond with glutathione instead. Thus, as a combined result, GSTO1
has low glutathione conjugating activity and instead can efficiently catalyze thiol
transferase reactions. The resulting mixed disulfide with glutathione is resolved by
another glutathione molecule with the release of oxidized glutathione (Brock et al.,
2013).
Much like other GSTs, GSTO1 too has a glutathione binding site (G-site) and an
adjacent hydrophobic/substrate-binding site (H-site). The H-site in GSTO1 is large, well-
defined and lined with some polar residues, reducing its hydrophobicity (Board et al.,
2000; Whitbread et al., 2005). Under native conditions, GSTO1 can form a homodimers
of 56 kDa, characteristic of cytosolic GSTs, with an atypically open and large subunit
7
interface. It has been suggested that such a cleft together with a polar H-site could serve
as a binding site for an endogenous substrate or ligand, which could be partly polar in
nature, possibly another protein. Recently, the existence of a novel ligand binding pocket
deep in the dimeric interface has been reported (Brock et al., 2013).
GSTO2 shares 64% sequence identity with GSTO1. The active site cysteine is
conserved in GSTO2 as well, which also has a higher overall cysteine content with 11
cysteine residues.
Figure 2.2. Crystal structure of GSTO1 (PDB ID: 1EEM) showing key structural
features.
8
2.2.2 GSTO1 catalytic activities
GSTO1 has a cysteine at its active site, instead of the typical tyrosine or serine.
Consequently, instead of glutathione conjugation, it catalyzes thioltransferase reactions
similar to glutaredoxins. In addition, GSTO1 has dehydroascorbate reductase, S-phenacyl
glutathione reductase and monomethylarsonate (V) (MMAV) reductase activities (Board,
2011; Board and Anders, 2007; Schmuck et al., 2005; Whitbread et al., 2005), making it
a unique member of the GST family. GSTO1 and GSTO2 share 64% sequence identity,
but differ in their enzyme activities. GSTO1 catalyzes S-phenacyl glutathione reduction
and monomethylarsonate (V) reduction (Board, 2011; Board and Anders, 2007; Schmuck
et al., 2005; Whitbread et al., 2005) with minimal dehydroascorbate reductase activity.
On the other hand, GSTO2 has a prominent dehydroascorbate reductase activity
(Schmuck et al., 2005). Both GSTO1 and GSTO2 have little activity with most GST
substrates. Table 2.1 below summarizes and compares the activity of GSTO1 and other
isoforms with representative substrates ((Board, 2011; Board and Anders, 2007; Board et
al., 2008; Board et al., 2000; Schmuck et al., 2005). GSTO1 has minimal activity with
common GST substrates. For example, GSTO1 has no activity on CDNB (Table 2.1). On
the other hand, upon mutation of the active site cysteine to alanine, its activity with
CDNB increases (Whitbread et al., 2005).
9
Table 2.1 GSTO catalytic activities
Substrate Specific activity (µ µmol/min/mg)
GSTO1 GSTO2
1-chloro-2,4-dinitrobenzene
(GSH conjugation)
0.18 ± 0.006
Thiol transferase 2.92 ± 0.12 1.5 ± 0.62
Dehydroascorbate reductase 0.16 ± 0.005 13.8 ± 0.29
S-phenacyl glutathione reductase 11.1 ± 0.1 N.D.
S-4-nitro-phenacyl glutathione 168.6 ± 6.3 N.D.
MMAV reductase 0.33 ± 0.037 0.42 ± 0.07
Ethacrynic acid N.D.
*
Cumene hyperoxide N.D.
trans-octenal 0.03 ± 0.006
trans-nonenal 0.03 ± 0.003
* N.D., Non detectable
2.2.3 Functions of GSTO1
GSTO1 has been reported to modulate ryanodine receptor (RyR), sarcoplasmic
reticulum Ca
2+
ion channels. GSTO1 showed inhibition of RyR2 in cardiac muscle. This
inhibition was shown to be dependent on enzyme activity. On the other hand, a
potentiation of RyR2 and RyR1 in skeletal muscle was independent of enzyme activity
(Dulhunty et al., 2001). It is interesting to note here that GSTO1 shares sequence
similarities with CLIC1, a nuclear chloride channel with a GST fold. It has been
suggested that GSTO1 could play a role in regulating intracellular Ca
2+
ion levels,
especially during apoptosis induced by Ca
2+
mobilization, although it remains to be
proven.
10
Recent studies have also demonstrated a novel role for GSTO1 in the processing
and release of IL-1β (Coll and O'Neill, 2011; Laliberte et al., 2003). It remains to be
elucidated if GSTO1’s effect on IL-1β release is a direct one, or by modifying protein
glutathionylation status of any inflammasome component, or through modulating ion
channels. A role in protection against oxidative cellular stress is also emerging (discussed
below).
2.2.4 GSTO1 in redox homeostasis and stress response signaling
GSTO1 is ubiquitously expressed in most tissues with high levels in heart, liver
and skeletal muscle, contrasting with a more tissue-specific expression seen with other
GSTs (Board et al., 2000). While the exact cellular functions of GSTO1 are not yet
understood, a novel role for GSTO1 in cellular stress response is emerging. In lower
organisms, GSTO1 overexpression conferred resistance to oxidative stress, while GSTO1
knockdown increased sensitivity to oxidants (Burmeister et al., 2008). Nuclear
translocation of GSTO1 during heat shock stress has also been reported (Kodym et al.,
1999). Interestingly, this has also been observed in patients with Barrett’s esophagus
progressing to esophageal cancer (Piaggi et al., 2009). The significance of this
translocation is yet to be elucidated.
The structural and biochemical properties of GSTO1 also hint at a potential role
in redox homeostasis. The N-terminal domain of GSTO1 has a thioredoxin-like fold.
Though it lacks the conserved CXXC motif commonly found in thioredoxins and
glutaredoxins, its active site cysteine (Cys-32) is oriented exactly like the proximal
11
cysteine in the CXXC motif (Board et al., 2000). Furthermore, like glutaredoxins,
GSTO1 catalyzes a thioltransferase reaction, an enzyme activity not shared by the other
GST isoforms. Glutaredoxins can also reduce mixed disulfides via a monothiol
mechanism involving the vicinal cysteine and glutathione (Fernandes and Holmgren,
2004). In addition, such monothiol glutaredoxins, with a single cysteine in a CFGS motif,
have been associated with intracellular iron sensing and trafficking via a bridging
glutathione- containing Fe/S cluster (Herrero and de la Torre-Ruiz, 2007; Muhlenhoff et
al., 2010). Therefore, we expect that GSTO1 may similarly catalyze formation or
reduction of protein-mixed disulfides, playing a role in thiol redox homeostasis. Indeed,
studies have shown that GSTO1 can glutathionylate mitochondrial F0/F1 ATPase β
subunit and suppress ER stress in a Drosophila model of Parkinson’s disease (Kim et al.,
2012). Recently, a significant role for GSTO1 in protein deglutathionylation has been
described (Menon and Board, 2013) and β-actin was identified as one of the cellular
substrates for GSTO1-catalyzed deglutathionylation.
Based on these studies and the biochemical properties ascribed to GSTO1, it
appears that there could be a strong role for GSTO1 in cellular stress response and that
inhibition of GSTO1 could disrupt such thiol homeostatic functions and cause redox
stress and cancer cell death. Indeed, recently, GSTO1 has been identified to be one of the
potential targets of piperlongumine, a small-molecule that causes death of cancer cells
selectively by disrupting redox stress response (Raj et al., 2011).
12
2.2.5 GSTO1 in diseases
Several association and genetic linkage studies have reported that GSTO1 resides in an
Alzheimer’s disease and Parkinson’s disease susceptibility linkage region on
chromosome 10 controlling age-at-onset (Li et al., 2003; Li et al., 2006). Polymorphisms
in GSTO1 and GSTO2 have also been associated with lower brain levels and altered risk
in these neurodegenerative diseases, although different studies in different patient
populations yielded differing results (Allen et al., 2012; Capurso et al., 2010; Kolsch et
al., 2004; Nishimura et al., 2005; Nishimura et al., 2004; Ozturk et al., 2005; Piacentini et
al., 2012). Similar, a limited association of GSTO1 and GSTO2 with age-at-onset in
amyotrophic lateral sclerosis has been reported (van de Giessen et al., 2008).
13
CHAPTER 3: VALIDATION OF GSTO1 AS A NOVEL
ANTICANCER DRUG TARGET
3.1 GSTO1 in cancer
Given the emerging role of GSTO1 in oxidative stress response and redox
homeostasis, GSTO1 appears to be important for cancer cell survival. Previously,
proteomic profiling studies have reported an overexpression of GSTO1 in invasive breast
cancer cell lines (Adam et al., 2002). GSTO1 A140D polymorphism has been associated
with susceptibility to acute lymphoblastic leukemia (Pongstaporn et al., 2009), bladder
cancer (Djukic et al., 2013) and head and neck cancer (Sanguansin et al., 2012).
However, the role of GSTO1 in cancer cells is yet to be elucidated.
3.2 GSTO1 in drug resistance
Several lines of evidence also suggest a role for GSTO1 in drug resistance.
GSTO1 overexpression has been reported in a mouse lymphoma cell line that is resistant
to cisplatin, etoposide, doxorubicin and radiation treatment. These resistant cells have
increased thiol content and do not undergo apoptosis, but are sensitive to arsenic trioxide
(Giri et al., 2005; Kodym et al., 2001; Story and Meyn, 1999). Similar overexpression has
been observed in platinum-resistant ovarian cancer cell lines (Yan et al., 2007). Further, it
has been shown that in cells overexpressing GSTO1, pro-apoptotic JNK signaling in
response to cisplatin treatment is inhibited, resulting in decreased apoptosis and
resistance (Piaggi et al., 2010). Although, GSTO1 overexpression is found in drug-
14
resistant cell lines, it does not appear to contribute directly to the drug resistance
(Schmuck et al., 2008). Moreover, enhanced drug detoxification by GSTO1 does not
appear to be a contributing factor for the drug resistance since GSTO does not catalyze
the typical GST reactions.
Therefore, we sought to examine the importance of GSTO1 in cancer cells and
validate GSTO1 as a useful anticancer drug target.
3.3 Materials and Methods
3.3.1 Bioinformatics analysis. GSTO1 gene expression in normal and cancer tissues was
analyzed using Oncomine cancer microarray database (www.oncomine.org) across
various cancer subtypes using a cancer versus normal differential analysis. Within the
same cancer sub-type, meta-analysis was done to compare different studies and assess
overall significance of GSTO1 expression. Data sets used in the study are summarized in
Table 3.1. Cancer subtypes with more than one study, p < 0.05 and fold change >1.5
were used as inclusion criteria for further analysis. GSTO1 mRNA expression in different
cancer cell lines was assessed using BioGPS (www.biogps.org).
3.3.2 siRNA experiments. GSTO1 siRNA (Trilencer-27) and scrambled control siRNA
were purchased from Origene (Rockville, MD). HCT116, H460, MDA-MB-435, MCF-7
cells were transfected with GSTO1 siRNA (siGSTO1-1, siGSTO1-2, siGSTO1-3, 10 nM)
or scrambled control siRNA (siSCRAM, 10 nM) using Lipofectamine
®
RNAiMAX
15
(Invitrogen) following manufacturer’s instructions. GSTO1 knockdown at 24 and 48 h
post-transfection was confirmed by in-gel fluorescence using CMFDA.
3.3.3 Cell viability assays. Cancer cells were seeded onto 96-well microtitre plates in a
100 µL volume of media. 10 µL of GSTO1 siRNA (10 nM), scrambled control (10 nM)
or siRNA buffer in transfection reagent mixture containing RNAiMAX and Opti-MEM
media was added to the cells. After 72 h post-transfection, cell viability was assessed by a
3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. MTT
solution (3 mg/mL; 20 µL) was added to each well and cells were incubated for 3 h at 37
°C. After incubation, media from each well was removed and the dark blue formazan
crystals formed by live cells were dissolved in DMSO (150 µL/well). The absorbance
intensity was measured at 570 nm on a microplate reader (Molecular Devices, Sunnyvale,
CA, USA). Two independent experiments with each siRNA tested in 6 replicate wells
were performed for each cell line.
3.3.4 Clonogenic survival assay. HCT116 cells were seeded at a density of 200
cells/well in a 6-well microtiter plate. After overnight attachment, cells were treated with
different concentrations of GSTO1 siRNA. Colonies were allowed to grow in medium
containing siRNA for 7 days and stained with 0.5% crystal violet.
16
3.4 Results
3.4.1 GSTO1 is overexpressed in many cancers
Figure 3.1 GSTO1 is overexpressed in several cancers.
(a) GSTO1 overexpression in select cancers analyzed using the Oncomine database. Box
plots show representative studies with normalized GSTO1 expression in cancer versus
normal tissue. (b) GSTO1 overexpression in various cancer cell lines analyzed using
BioGPS database.
a
b
17
We evaluated the expression of GSTO1 in various tumor types using publicly
available gene expression datasets from the Oncomine database. GSTO1 was
significantly overexpressed in colorectal, head and neck, breast (Fig. 3.1a) and
esophageal cancers, melanomas and lymphomas (Table 3.1). Similarly, we observed an
upregulation of GSTO1 mRNA in different cancer cell lines (HCT116, MDA-MB-435,
MDA-MB-231, H460) by analyzing transcript data in BioGPS (Fig. 3.1b). These were
further confirmed at protein level (Fig. 3.2a).
18
Table 3.1. GSTO1 expression in Oncomine gene expression studies
Cancer Cancer subtype Data set Year
Number of
samples
Fold change in
expression
p-value
Head and
neck
Tongue squamous cell
carcinoma
Talbot Lung 2005 93 1.969 1.23E-7
Estilo Head-Neck 2009 58 2.015 1.86E-6
Ye Head-Neck 2008 38 1.458 0.008
Head and neck squamous cell
carcinoma
Cromer Head-Neck 2004 38 1.722 0.008
Colorectal
Colon adenocarcinoma
Kaiser Colon 2007 105 1.457 0.001
TCGA Colorectal 2011 237 1.456 3.03E-7
Colon mucinous
adenocarcinoma
Kaiser Colon 2007 105 1.521 5.97E-4
TCGA Colorectal 2011 237 1.635 3.18E-6
Cecum adenocarcinoma
Kaiser Colon 2007 105 1.547 9.10E-4
TCGA Colorectal 2011 237 1.365 2.61E-4
Colorectal carcinoma
Skrzypczak Colorectal 2010 105 1.575 7.76E-6
Hong Colorectal 2010 82 1.408 9.05E-5
Rectal adenocarcinoma
Kaiser Colon 2007 105 1.496 0.002
TCGA Colorectal 2011 237 1.331 1.14E-4
Esophageal
Esophageal adenocarcinoma
Wang Esophagus 2006 52 2.146 7.18E-8
Kimchi Esophagus 2005 24 1.608 0.011
Hao Esophagus 2006 48 2.148 0.033
Esophageal squamous cell
carcinoma
Su Esophagus 2 2011 106 1.764 6.13E-15
Hu Esophagus 2010 34 1.521 2.47E-4
Lymphoma Diffuse large B-cell lymphoma Rosenwald Multi-cancer 2001 102 2.125 4.13E-4
19
Rosenwald Lymphoma 2002 293 1.482 4.57E-4
Brune Lymphoma 2008 67 3.815 1.83E-7
Compagno Lymphoma 2009 136 1.881 3.21E-14
Basso Lymphoma 2005 336 1.251 0.011
Activated B-cell like Diffuse B-
cell lymphoma
Alizadeh Lymphoma 2000 120 1.813 8.44E-6
Compagno Lymphoma 2009 136 1.390 0.003
Germinal center B-cell like
Diffuse B-cell lymphoma
Alizadeh Lymphoma 2000 120 2.148 1.85E-7
Compagno Lymphoma 2009 136 1.918 1.07E-7
Melanoma Cutaneous melanoma
Riker Melanoma 2008 87 3.831 5.24E-5
Talantov Melanoma 2005 70 4.084 2.81E-7
Criteria: Number of studies > 1, p-value<0.05. Green = Significant overexpression of GSTO1 and IL-1 β
20
3.4.2 GSTO1 knockdown inhibits cancer cell proliferation
To determine the significance of overexpression in cancer cells we specifically
silenced GSTO1 using siRNA. Knockdown of GSTO1 in HCT116 cells was confirmed
by in-gel Western using 5-Chloromethylfluorescein diacetate (CMFDA) that potently and
irreversibly binds to GSTO1 (Son et al., 2010) (Fig. 3.2a). We found that treatment with
GSTO1-specific siRNA decreased viability of cell lines with high GSTO1 expression
(HCT116, H460 and MDA-MB-435) cancer cells (Fig. 3.2b). On the other hand, cell
lines with low (MCF7) or no (T47D) GSTO1 expression showed little to no effect on cell
viability. Formation of colonies was also significantly decreased in GSTO1-silenced
HCT116 cells (Fig. 3.2c). We excluded the possibility of any off-target or concentration-
dependent effects by testing different siRNA concentrations. Taken together, the results
from the gene expression analysis and the siRNA studies strongly suggest that GSTO1 is
important for cancer cell survival and can be harnessed as a novel anticancer target.
21
Figure 3.2 GSTO1 knockdown induces cancer cell death.
(a) GSTO1 expression in different cancer cell lines. (b) siRNA-mediated knockdown of
GSTO1 (lower band). HCT116 cells were transfected with GSTO1-specific siRNAs
(siGSTO1-1, siGSTO1-2, siGSTO1-3, 30 nM) or scrambled control siRNA (siSCRAM)
for 48 and 72h. Protein expression was confirmed using CMFDA as a GSTO1 binding
dye. (c) GSTO1 is important for cell viability. Cell lines (HCT116, H460, MDA-MB-
435, MCF7, T47D) were transfected with 3 different GSTO1-specific siRNAs or
siSCRAM (10 nM). Cell viability was measured 72 h post transfection using MTT assay
and compared against siSCRAM-transfected cells. Percent decrease in cell viability is
shown. Data are mean ±
SEM of replicate wells. (d) GSTO1 knockdown inhibits colony
formation of HCT116 cells.
22
3.3 Discussion
The results of this study show that silencing GSTO1 inhibits viability of cancer cells,
especially in cell lines with high GSTO1 expression, indicating that GSTO1 has crucial
roles in cancer cell survival. We further show that GSTO1 is overexpressed in various
cancer types such as colon, head and neck, and esophageal cancers, providing a rationale
for targeting GSTO1 for developing anticancer agents.
23
CHAPTER 4: IDENTIFICATION OF NOVEL GSTO1
INHIBITORS
4.1 Background
Few small-molecules and peptides bearing electrophilic moieties have previously
been reported as GSTO1 inhibitors, acting through the active site cysteine (Bachovchin et
al., 2009; Pace et al., 2012; Sampayo-Reyes and Zakharyan, 2006; Son et al., 2010;
Tsuboi et al., 2011) (Fig. 4.1). In addition, tocopherol esters have also been reported as
uncompetitive inhibitors of GSTO1, thought to bind in a novel ligand binding pocket
(Brock et al., 2013; Sampayo-Reyes and Zakharyan, 2006).
Figure 4.1 Structures of GSTO1 inhibitors reported in literature.
24
Many of these compounds contain electrophilic moieties such as α-
chloroacetamides or phenol groups. Compounds such as 8 and KT53 have good potency
and show irreversible GSTO1 inhibition in vitro and in situ (Bachovchin et al., 2009;
Tsuboi et al., 2011). While Tsuboi et al. reported that cancer cells treated with KT53
shows increased sensitivity to cisplatin, an in-depth characterization of their effects in
cancer cells is lacking. We expand on this previous study and sought to design small-
molecule inhibitors of GSTO1 to study its cellular functions and therapeutic utility in
cancer. This chapter describes the screening assays, inhibitor design and optimization
effort and characterization of lead GSTO1 inhibitors.
4.2 Materials and Methods
4.2.1 Compounds for high throughput screening. Compounds for pilot screening were
synthesized by Dr. Yarovenko, N.D. Zelinsky Institute of Organic Chemistry, Russian
Academy of Sciences, Russia. For the similarity search and expanded screening, a
diverse library of small molecule compounds (~1,000,000) from Asinex, Enamine and in-
house collection was used. All compounds were stored as 50 mM stock in
dimethylsulfoxide (DMSO) at -80 °C. The lead compound C1-27 was procured from
Enamine (97% purity) and further characterized using mass spectrometry (ESI-MS (m/z):
[M]
+
calcd. for C
10
H
12
Cl
2
N
2
O
3
S, 309.99; found, 309.16). Substrate for GSTO1 enzymatic
assay, S-(4-nitrophenacyl)glutathione (4-NPG) was synthesized as previously described
(Board et al., 2008). Product purity was confirmed by mass spectrometry (ESI-MS (m/z):
25
[M]
+
calcd. for C
18
H
23
N
4
O
9
S, 471.12; found, 471.04) and Ellman’s reagent (unreacted
residual glutathione < 1%).
4.2.2 Expression and purification of GSTO1 for screening. Human GSTO1-1 plasmid
was a kind gift from Dr. Philip Board, John Curtin School of Medical Research,
Australian National University. Recombinant GSTO1-1 was expressed in E.Coli. M15
(Rep4) cells (Qiagen) and purified as described previously (Board et al., 2000; Whitbread
et al., 2003). Briefly, GSTO1-1 pQE30 plasmid was transformed into M15 cells, which
were grown in LB broth supplemented with 100 µg/mL Ampicillin and 25 µg/mL
Kanamycin at 37 °C to an OD
600
of approximately 0.8 and induced with 0.1 mM
isopropyl thio-β-D-galactoside for 4 h. Bacterial cells were harvested by centrifugation at
4000g for 20 min at 4 °C and the cell pellet was resuspended in buffer (50 mM NaH
2
PO
4
,
300 mM NaCl, pH 6.0). Cells were lysed by sonication and the cleared supernatant was
purified using Ni-NTA resin. The purified enzyme was dialyzed against 20 mM Tris-
HCl, 60 mM NaCl, pH 8.0. Protein purity was assessed by SDS-PAGE.
4.2.3 GSTO1 enzyme activity assay. GSTO1 enzyme activity was measured by
monitoring the reduction of S-(4-nitrophenacyl)glutathione (4-NPG) to 4-
nitroacetophenone by GSTO1 (Board et al., 2008). Briefly, in a 200 µL reaction volume,
5 µg/mL recombinant GSTO1 in reaction buffer (100 mM Tris (pH 8.0), 1.5 mM EDTA,
1 mM dithiothreitol) was incubated with DMSO or different concentrations of inhibitors
for 30 min at 37 °C. 4-NPG (1 mM) was added to the reaction and decrease in
26
absorbance at 305 nm was recorded on an Envision plate reader (Perkin Elmer) using a
320 nm UV filter.
4.2.4 Reversibility of inhibition. GSTO1 (1 µg) was incubated with inhibitor at a
concentration of 10 k
i
in a 5 µL reaction volume for 30 min at 37 °C and then diluted to
200 µL with reaction buffer (100 mM Tris (pH 8.0), 1.5 mM EDTA, 1 mM dithiothreitol)
to attain final concentrations of 5 µg/mL GSTO1 and 0.25 k
i
inhibitor. 4-NPG (1 mM)
was added and decrease in absorbance was monitored. % Recovery of enzyme activity
was determined using a DMSO control.
4.2.5 In-gel fluorescence binding assay using CMFDA. Primary screening of GSTO1
inhibitors was based on competitive inhibition of CMFDA binding to endogenous
GSTO1. Briefly, HCT116
p53+/+
cells (4 x 10
4
/well) were seeded in a 12-well plate. After
overnight attachment, cells were treated with test compounds at 10 µM for 2 h at 37 °C
followed by addition of 500 nM CMFDA for 1 h. Cells were then washed with PBS and
lysed using Cell Lytic M buffer (Sigma). 15 µg of extracted whole-cell protein was
incubated with laemmli sample buffer at 90 °C and resolved on a 15% polyacrylamide
gel. Gels were immediately scanned on a Typhoon variable mode imager (GE
Healthcare). Quantification of fluorescent band intensity was performed using Image
Quant 5.2 software. Compounds that showed at least 50% inhibition of CMFDA binding
to GSTO1 were selected for dose-response determinations. Similarly, in vitro binding
assay with recombinant GSTO1 was performed using 1 µM GSTO1 in reaction buffer
27
(100 mM Tris (pH 8.0), 1.5 mM EDTA, 1 µM dithiothreitol) incubated with compounds
for 30 min at 37 °C and CMFDA (500 nM) for 30 min. The reaction was quenched by
incubating with laemmli sample buffer for 5 min at 90 °C. The samples were resolved on
a 10% polyacrylamide gel and scanned on a Typhoon variable mode imager. Protein
loading was assessed by Sypro Ruby staining.
4.2.7 Thermal shift assay. Florescence based thermal shift assay was carried out on
ThermoFluor 384 analyzer (Johnson and Johnson, New Brunswick, NJ). Recombinant
GSTO1 (0.3 mg/ml), 1-anilinonaphthalene- 8-sulfonic acid (1,8 ANS, 0.3mM) and 50
mM tris pH 8.0 (buffer) were used for the experiments. C1-27 was tested at a range of
concentrations (9.375 – 75 µM, 2-fold dilutions). 1.5% DMSO in buffer was used as a
control. The assay was carried out using standard protocol (Pantoliano et al., 2001).
Briefly, 5µl protein-dye solution and an equal volume of the test compound solutions
were added in duplicate into a 384-well microplate (Thermo Scientific, AB1384K). 3 µl
of silicone oil (Sigma) was added to the each well to avoid evaporation. Reference
control well contains 1.5% DMSO (no test compound). Plate was heated at a temperature
range of 25 to 90
0
C with a temperature increment of 1
0
C/min. Fluorescence was
measured by fiber optics and fluorescence emission was detected by measuring light
intensity using CCD camera. The midpoint (T
m
value) from the melting curve was
determined by fitting the fluorescence data to a modified Boltzmann model using
Thermofluor software.
28
4.2.6 Kinetics of target association. Time course of inhibitor binding was performed as
follows. HCT116 cells were treated with C1-27 (100 nM) for indicated durations,
followed by drug washout. CMFDA (500 nM) was added for 1 h and the cells were lysed
and processed for the in-gel fluorescence binding assay as detailed above, to assess
inhibition of CMFDA binding to GSTO1. The fluorescence intensity of bands were
quantified using Image Quant and plotted as mean ± SD of three independent
experiments.
4.2.7 Receptor occupancy and target dissociation studies. Receptor occupancy and
duration of inhibition experiment was performed based on a previously reported method
(Serafimova et al., 2012). HCT116 cells were treated with C1-27 (100 nM) for 2 h,
followed by drug washout. At indicated times after washout, CMFDA (500 nM) was
added for 1 h and the cells were lysed and processed for in-gel fluorescence binding
assay, to assess recovery of CMFDA binding to GSTO1. The fluorescence intensity of
bands were quantified using Image Quant and plotted as mean ± SD of three independent
experiments.
29
4.3 Results
4.3.1 Optimization of high throughput activity assay
We purified recombinant human GSTO1 expressed in E. Coli. M15 cells. The
purity of yield and concentration was assessed by SDS-PAGE (See Appendix). For
activity screening, we adapted a GSTO1 substrate assay that quantifies reduction of 4-
nitro-phenacylglutathione (4-NPG), a GSTO1-specific substrate (Board et al., 2008), by
recombinant human GSTO1 into a high throughput format. Z’ factor value for the assay
was 0.71 (Fig. 4.2).
Figure 4.2 GSTO1 substrate assay optimization.
(a) Schematic representation of the reaction mechanism for GSTO1-catalyzed reduction
of 4-nitrophenacyl glutathione (4NPG). (b) Linearity of signal (R
2
= 0.99). (c) Kinetics of
substrate reduction. (d) Assay performance characteristics in high throughput format
(average Z’ factor = 0.73, average S/N ratio = 16.5).
30
4.3.2 Optimization of competitive binding assay
Using CMFDA, which has been reported to be a GSTO1-specific inhibitor (Son et
al., 2010), we optimized a gel-based binding assay that measures competitive inhibition
of CMFDA binding to recombinant GSTO1 or endogenous GSTO1 in a soluble proteome
(Fig. 4.3).
Figure 4.3. Optimization of GSTO1 competitive binding assay.
(a) Titration of recombinant GSTO1 concentration using CMFDA (500 nM) (b) Time
course of CMFDA (500 nM) binding to recombinant GSTO1 (1 µM). (c) Effect of DTT
concentration in reaction buffer on CMFDA binding to recombinant GSTO1. (d) Time
course of CMFDA (500 nM) binding to endogenous GSTO1 (12 µg of soluble
proteome). Protein loading assessed using Sypro ruby. Fluorescent gels were scanned on
a Typhoon variable mode imager using Fluorescein 535 nm filter.
31
4.3.3 Identification of GSTO1 inhibitors
Design basis
To identify small molecule inhibitors of GSTO1 activity, we employed a 2-
chloro-N-phenylacetamide scaffold as the starting point to target the active site cysteine
residue and identify novel GSTO1 inhibitors. The α-chloroacetamide is a privileged
scaffold that has been commonly used to target cysteine residues on proteins. For
example, compounds containing tertiary α-chloroacetamide moieties have been reported
to show GSTO1 (Bachovchin et al., 2009; Tsuboi et al., 2011) and protein disulfide
isomerase inhibition (Hoffstrom et al., 2010).
Pilot screening
Using the 2-chloro-N-phenylacetamide scaffold, we designed a small set of novel
small molecules to evaluate their effect on GSTO1 and explore structure-activity
relationship. 4a, one of the initial hits, inhibited GSTO1 enzyme activity with an IC
50
value of 3.1 µM. We observed that addition of electron withdrawing substituents on the
phenyl ring increased the electrophilicity of the chloroacetamide carbon and improved
potency of inhibition (Table 4.1).
32
Table 4.1 Results of pilot screening
Cpd Structure
Substrate assay
IC
50
(µM)
2a
F
H
N
O
Cl
F
3.5
2b
F
H
N
O
Cl
F
NO
2
1.7
2c
0.71
4a
3.1
4b
3.9
4c
80.8
6
>100
F
F
H
N
NO
2
O
Cl
F
F
H
N
O
Cl
NH
O
Cl
F
H
N
O
Cl
NH
O
Cl
F
H
N
O
Cl
NH
O
Cl
H
N
HN
O
O
Cl
Cl
33
SAR studies and Lead identification
Based on the results of pilot screening, we next performed a structural similarity
search across our in-house compound collection as well as commercial small-molecule
libraries (~1,000,000 compounds) and identified 141 additional hits (Fig. 2A). We
screened these hits using the high throughput activity assay at a single concentration of
10 µM. Their GSTO1 inhibitory activity was further confirmed by the competitive
binding assay. The 141 hits were screened at 10 µM in the competitive binding assay
against recombinant and endogenous GSTO1. The general approach for inhibitor
identification is outlined in Fig. 4.4.
Figure 4.4 Schematic of inhibitor identification strategy
% Inhibition of CMFDA
binding to rec GSTO1
% Inhibition of CMFDA
binding to endo GSTO1
% Inhibition of GSTO1
enzyme activity
1° Screening: GSTO1 substrate assay
2° Screening: Binding assay – Rec GSTO1
3° Screening: Binding assay – Endo GSTO1
Actives and dose response assays (43)
Identify candidate compounds (141)
Similarity search of compound library (1,000,000)
Pilot screening of synthesized analogues
34
Based on their structures, the hit compounds were grouped into 5 different
clusters (Fig. 4.5-4.9). Among the different clusters, Cluster-1, based on a N-Phenyl-2
0
-
chloroacetamide scaffold, yielded some of the most potent active compounds. Replacing
the phenyl ring with a heterocyclic moiety was also well-tolerated and resulted in potent
hits (Cluster-2). Cluster 3 consisting of N-Linker-2
0
-chloroacetamides was mostly
inactive, suggesting that inclusion of additional linker groups between the
chloroacetamide moiety and the aromatic center decreased its electrophilic nature and
negatively affected potency. Similarly, replacing the 2
0
-chloroacetamide with a 3
0
-
chloroacetamide derivative such as in Cluster-4 also proved unfavorable and resulted in
loss of potency. On the other hand, cyclization of the 3
0
-chloroacetamide restored activity
(Cluster-5).
From this preliminary screening, we selected 43 compounds that showed at least
50% inhibition in all three assays for further dose response studies (Tables 4.2-4.6).
Inhibitory profile of select potent compounds is shown in Table 4.7.
35
Figure 4.5 Cluster 1 - N-Phenyl-2
0
-chloroacetamide
N
H
O
Cl
C1-1
NO
2
N
H
O
Cl
Cl
C1-2
O
2
N
N
H
O
Br
Br
Cl
C1-3
NO
2
N
H
O
Cl
NH
F C1-4
N
H
O
S
Cl
C1-5
N
H
O
O
F
F Cl
C1-6
N
H
O
O
Br
Cl
C1-7
N
H
O
O
O
O
O
Cl
C1-8
N
H
O
O
O
O
Cl
C1-9
HOOC
N
H
O
Cl
O
Cl
C1-10
N
H
O
O
O
Cl
Cl
C1-11
N
H
O
O
O
Cl
Cl
C1-12
N
H
NH
O
O
O
Cl
C1-13
N
H
N
O
Cl
Cl
C1-14
N
N
H
O
Cl
C1-15
N
N
H
O
Cl
C1-16
N
N
H
O
Cl
C1-19
N
N
N
H
O
Cl
C1-20
CF
3
N
N
H
O
Cl
C1-17
N
N
H
O
O
Cl
C1-18
N
N
O
O
Cl
H
C1-21
N
N
H
O
O
Cl
C1-22
N N
N
N
H
O
Cl
C1-23
S
O
O
N
H
O
F
F
Cl
C1-24
S
N
O
O
N
H
O
Cl
C1-25
S
N
H
N
O O
O
Cl
C1-26
S
N
O O
N
H
O
Cl
Cl
C1-27
S
N
O O
N
H
O
Cl
C1-28
S
N
N
H
O O
O
O
Cl
Cl
Cl
Cl
C1-29
S
N
O O
N
H
O
O
Cl
C1-30
S
N
N
H
O O
O
O
Cl
C1-31
S
N
N
H
O O
N
O
O
Cl
C1-32
36
Table 4.2. Screening data for Cluster-1 - N-Phenyl-2
0
-chloroacetamide
GSTO binding assay
(% Inhibition at 10 µM)
GSTO substrate assay % Inhibition of
HCT116p53+/+ cell
proliferation at 10 µM
MW HBA HBD LogP
Recombinant Endogenous % I at 10 µM IC
50
(µM)
C1-1 17.7 16.2 14, 21.3
253.77 2 1 3.89
C1-2 54.1 66.2 100.9 1.22 7.4
249.05 4 1 2.42
C1-3 0.9 3.8 -4.9
372.41 4 1 2.77
C1-4 81.5 31.8 92.5 80.6
323.71 5 2 3.97
C1-5 52.9 22.9 96.3 -3.8, -2.6
241.74 2 1 2.85
C1-6 45.6 87.6 22.7
235.62 3 1 1.95
C1-7 75.6 56.6 77.7 0.162 29.6
340.61 3 1 4.12
C1-8 61.2 56.5 100.9 5.68
285.69 6 1 2.13
C1-9 56.5 39.3 12.5 -1.2
255.66 5 1 1.15
C1-10 49.1 49.7 81.7 19.7
278.09 5 2 2.42
C1-11 55.2 58.6 77.7 1.16 63.4, 65.6
276.12 4 1 2.13
C1-12 26.5 40.1 10.4 85
334.16 4 1 2.98
C1-13 35.2 61.1 62.2 14.3, 6.6
292.72 5 2 1.68
C1-14 -15.6 6.7 16.5 88.0
247.13 3 1 2.65
C1-15 43.8 78.6 91.3
254.76 3 1 2.89
C1-16 49.2 46.7 87.8 41.9
252.75 3 1 2.81
C1-17 -4.1 -9.5 88.5 94.8, 91
306.72 3 1 3.55
C1-18 6.4 54.4 3.7 41.5, 41.7
252.70 4 1 1.46
C1-19 38.3 65.7 61.5 79.2, 85.4
266.77 3 1 3.34
C1-20 -8.5 -15.1 66.9 43.7, 39
281.79 4 1 2.13
C1-21 69.9 58.2 -8.4
254.72 4 1 1.56
C1-22 25.1 49.0 41.4 80.8, 86
252.70 4 1 1.29
C1-23 56.6 63.0 89.8 4.79 36.5
304.78 4 1 1.77
37
C1-24 68.4 62.1 77 0.148 89.5, 91.8
283.68 4 1 1.9
C1-25 70.3 9.8 11.3, 21.7
344.86 5 1 2.63
C1-26 51.4 48.9 31.1 13.2
304.80 5 1 1.33
C1-27 69.8 88.3 0.036 93.8
311.19 5 1 1.62
C1-28 71.5 57.2 87.6 0.064 94.4
338.81 5 1 2.31
C1-29 70.6 61.0 102.3 0.030 42.5
470.16 6 1 3.61
C1-30 75.9 80.8 75 0.079 45, 20.7
360.86 6 1 1.93
C1-31 60.0 77.0 0.221 88.7, 83.5
360.86 6 1 2.33
C1-32 4.4 48.2 84.2
389.90 7 1 1.61
38
Figure 4.6 Cluster 2 - N-Heterocycle-2°-chloroacetamide
N
N
H
O
Cl
C2-1
N
N
H
O
Cl
C2-2
N N
H
N
O
F
Cl
C2-3
N
N
N
H
O
Cl
C2-4
N
N
N
H
O
Cl
C2-5
N
N
N
H
N
O
Cl
C2-6
N N
H
N
O
Cl
Cl
C2-7
S N
H
COOEt
O
Cl
C2-8
S N
H
COOEt
Cl
O
Cl
C2-9
S N
H
COOMe
O
Cl
Cl
C2-10
S
EtOOC
N
H
O
Cl
C2-11
S N
H
COOEt
O
O
O
Cl
C2-12
S
N
H
COOEt
O
O
O
Cl
C2-13
S N
H
NH
COOEt
O
O
Cl
C2-14
S
N
H
N
Cl
O
O
O
C2-15
S N
H
COOEt
O
Cl
C2-16
S N
H
COOEt
N
O
Cl
C2-17
S N
H
CONH
2
Cl
O
C2-18
N
N
NH
S N
H
O
O
Cl
O
C2-19
S
N
H
O
Cl
C2-20
S N
H
O
Cl
EtO
N
C2-21
N
O
2
N
N
H
S
O
Cl
C2-22
N
N
H
S
O
Cl
C2-23
N
N
H
S
O
HO
Cl
O
C2-24
N
N
H
H
N
S
O
Cl
C2-25
N
N
H
S
O
Cl
Cl
C2-26
N
S
N
H
O
O
O
C
l
C2-27
O N
H
N
O
Cl
C2-28
N
HN
N
H
O
O
O
Cl
C2-29
N
N
S
N
H
O
O
Cl
C2-30
O
39
Table 4.3 Screening data for Cluster-2 - N-Heterocycle-2
0
-chloroacetamide
GSTO binding assay
(% Inhibition at 10 µM)
GSTO substrate assay
% Inhibition of
HCT116p53+/+ cell
proliferation at 10 µM
MW HBA HBD LogP
Recombinant Endogenous % I at 10 µM IC
50
(µM)
C2-1 -0.8 45.0 54.8 65.6, 62.7
204.70 3 1 0.85
C2-2 6.5 66.7 90.5 1.1
266.77 3 1 1.98
C2-3 37.8 47.0 61.5 49
305.74 3 1 2.95
C2-4 35.8 51.3 0.5 81.4, 84.4
277.76 3 1 2.29
C2-5 64.1 69.4 4.4 -2.7, -8.6
275.74 3 1 2.12
C2-6 36.8 25.2 84.5 -17, -9.7
260.68 4 1 1.43
C2-7 56.0 55.3 79.7 0.598 85.7, 91
284.15 3 1 2.39
C2-8 39.2 51.7 3.8 4.5, 0.8
289.78 4 1 3.37
C2-9 50.0 59.2 13.9 16.3
358.25 4 1 4.24
C2-10 48.4 51.6 88.4 0.062 9.1
344.22 4 1 3.89
C2-11 58.2 19.9 -1.5
323.80 4 1 3.88
C2-12 74.9 59.8 3.4
383.85 6 1 3.38
C2-13 48.1 41.3 25.9 38.1
395.86 6 1 3.78
C2-14 57.9 58.2 -7.0 6.3
408.91 6 2 4.01
C2-15 63.2 74.2 87.8 0.360
378.88 5 1 2.62
C2-16 54.7 53.0 73.0 0.066
315.82 4 1 3.86
C2-17 51.5 24.8 77.2 85.2, 94.5
330.84 5 1 2.7
C2-18 72.8 78.7 0.119 82.1
286.78 4 2 2.39
C2-19 51.8 58.6 1.2 52.7, 40.5
458.97 5 2 3.14
C2-20 68.3 39.6 91.1
333.84 3 1 4.1
C2-21 78.6 52.3 99.5 0.350
426.97 4 1 5.73
C2-22 80.3 84.9 0.136 75.7
297.72 5 1 2.32
C2-23 76.0 26.4
294.80 3 1 3.3
40
C2-24 82.4 62.1 34.0 22.9, 19.2
298.75 5 2 2.4
C2-25 75.7 3.7 0.5, 4.3
291.76 3 2 2.58
C2-26 72.9 80.0 0.125 93.8
301.20 3 1 3.18
C2-27 69.0 59.6 85.1 8.669 42.6, 38.2
284.72 5 1 1.88
C2-28 59.5 -2.8 88.5
364.83 4 1 4.45
C2-29 13.5 65.0 42.7 8.5, 29.1
295.73 6 2 0.88
C2-30 50.5 41.1 47.4 12.0
271.73 4 1 1.58
41
Figure 4.7 Cluster 3 - N-Linker-2
0
-chloroacetamide
N
N
H
O
O
Cl
C3-1
S
N
O
O
N
H
O
Cl
C3-2
N
+
-
O
O
N
H
O
OH
Cl
C3-3
N
H
O
O O
O Cl
C3-4
O
N
O
O
Cl
H
C3-5
N
H
O
Cl
C3-6
N
H
O
O
O
Cl
C3-7
N
H
O
O
O
Cl
C3-8
O
O
N
H
O
Cl
C3-9
N N
N
H
O
Cl
C3-10
O
O
N
H
F F
O
Cl
C3-11
N
H
N
O
Cl
C3-12
O
N
H
Cl
O
O
C3-13
O
N
H
S
Cl
C3-14
O
N
H
S
Cl
C3-15
N
O
N
H
Cl
C3-16
H
N
N
H
O Cl
Cl Cl
Cl
C3-17
N
H
O O
Cl
C3-18
N
H
N
H
O O
O
O
Cl
C3-19
H
N
N
H
O
O
Cl
C3-20
H
N
N
H
O
O
Br
Cl
C3-21
H
N
N
H
O
O
O
O
O
Cl
C3-22
S
H
N
N
H
O
O
Cl
C3-23
N
H
H
N
O
O
Cl
C3-24
N
H
H
N
O
O
F
F
Cl
C3-25
N
H
S
N
H
S
O
O
O
Cl
C3-26
N
H
N
H
N
H
S O
Cl Cl
Cl
Cl
C3-27
N
H
S
O
Cl
C3-28
O
N
H
O
Cl
C3-29
N
N
N
H
O
Cl
C3-30
N
N
H
O
Cl
C3-31
N
S
O
O
O
N
H
F
Cl
C3-32
N
H
O
Cl
C3-33
O
N
H
Cl
C3-34
42
Table 4.4 Screening data for Cluster-3 - N-Linker-2
0
-chloroacetamide
GSTO binding assay
(% Inhibition at 10 µM)
GSTO substrate assay
% Inhibition of
HCT116p53+/+ cell
proliferation at 10 µM
MW HBA HBD LogP
Recombinant Endogenous % I at 10 µM IC
50
(µM)
C3-1 6.0 7.3 11.1
282.77 4 1 1.32
C3-2 63.3 63.6 86.4 1.48 55.3, 49.6
318.82 5 1 1.36
C3-3 47.9 52.0 93.2
244.64 5 2 1.61
C3-4 -2.6 31.6 16, 17.5
273.72 5 1 1.19
C3-5 14.7 31.8 85.1 64.1, 63.6
283.76 4 1 2.13
C3-6 53.7 51.3 56.2 7.88 13.6, 34.7
267.80 2 1 3.64
C3-7 12.3 12.6 12.5 73.6
241.68 4 1 1.48
C3-8 -4.9 33.4 16.5 91.2, 77.9
283.76 4 1 2.06
C3-9 30.7 53.9 80.7 24.9
241.68 4 1 1.21
C3-10 29.4 48.4 78.4 96.5, 87.4
263.73 3 1 1.29
C3-11 10.1 45.6 9.8 8.1, 16.1
293.70 4 1 2.15
C3-12 22.0 9.7 83.2 5
240.73 3 1 1.55
C3-13 25.1 62.6 1, 16.8
257.72 4 1 1.93
C3-14 38.8 56.4 78.4 28.5
243.76 2 1 2.66
C3-15 35.7 86.9 5.1
257.78 2 1 2.82
C3-16 23.7 45.4 83.1 28.7
240.73 3 1 2.03
C3-17 57.6 39.3 61.9 7.4
287.96 3 2 2.89
C3-18 0.3 37.5 0.6
247.68 3 1 2.57
C3-19 59.2 65.2 99.5 0.508
300.74 6 2 2.01
C3-20 3.9 32.9 42.7 29.1
268.74 4 2 1.98
C3-21 58.4 27.6 81.3 9.9
291.54 4 2 1.55
C3-22 -2.4 -12.1 10.4
302.72 7 2 0.68
C3-23 -4.1 -11.7 -3.5 3.7
272.75 4 2 1.81
43
C3-24 29.2 58.8 72.3
268.74 4 2 1.61
C3-25 36.2 57.1 82.4 64.5, 67.8
262.64 4 2 1.29
C3-26 68.2 42.2 69.5
382.89 5 2 3.38
C3-27 63.4 63.8 54.2 1.65
389.13 4 3 3.53
C3-28 15.1 44.1 42.0 48.5, 48.7
231.75 2 1 2.21
C3-29 46.7 88.9 73.0 89.7
237.69 3 1 2.15
C3-30 10.6 14.5 83.1
291.78 3 1 2.01
C3-31 -13.7 23.0 0.3
272.82 3 1 2.83
C3-32 23.6 53.1 12.5
342.78 5 1 2.02
C3-33 16.6 55.3 36.3 24.7, 38.5
255.79 2 1 3.52
C3-34 50.9 41.7 78.6 -2.4
255.79 2 1 3.88
44
Figure 4.8 Cluster 4 - Non-cyclized 3
0
-chloroacetamide
N
F
3
C
O
Cl
Cl
C4-1
CF
3
N
N
O
Cl
C4-2
N
O
O NH
2
Cl
C4-3
S
N O
O
O
Cl
C4-4
N
N S
O
Cl
Cl
Cl
C4-5
N
O
Cl
O
C4-6
H
N
N
O
O Cl
Cl
Cl
C4-7
O
N
O
O
Cl
C4-8
N
O
Cl
Cl O
C4-9
N
N
N
O
Cl
C4-10
N
O
O
O
Cl
C4-11
N
N
O
O O
Cl
C4-12
S
N
O
O
H
N
O
O
Cl
C4-13
N
O
O H
2
N
Cl
C4-14
N
O
O
Cl
C4-15
N
N
S
O
Cl
C4-16
N
O
O
O O
Cl
C4-17
N NH
N
O
O H
2
N
O
Cl O
C4-18
N HN
N
O
O NH
2
O
Cl
C4-19
P
H
N
N
N
N
O
O
O
O
O
Br
Cl
C4-20
H
N
N
O
O
HO
Cl
C4-21
45
Table 4.5 Screening data for Cluster-4 - Non-cyclized 3
0
-chloroacetamide
GSTO binding assay
(% Inhibition at 10 µM)
GSTO substrate assay
% Inhibition of
HCT116p53+/+ cell
proliferation at 10 µM
MW HBA HBD LogP
Recombinant Endogenous % I at 10 µM IC
50
(µM)
C4-1 40.4 58.5 73.0 3.80 76.2, 81.4
286.08 2 0 2.97
C4-2 53.2 52.8 78.6 1.07 84.1, 93.6
290.67 3 0 2.29
C4-3 -1.7 32.3 70.6, 71.9
240.69 4 1 0.72
C4-4 5.9 17.0 -0.8 -11.3, 5.3
239.72 4 0 -0.42
C4-5 -8.2 15.9 5.0 9.9, 16.9
361.68 3 0 4.52
C4-6 55.2 49.5 29.3 44.1, 68.5
277.75 3 0 3.27
C4-7 -36.4 53.0 1.8 81.8
309.58 4 1 2.21
C4-8 59.5 53.0 97.2 0.237 70.7
255.70 4 0 1.42
C4-9 66.7 43.5 20.7 -6.8, 5.4
262.14 3 0 2.39
C4-10 66.4 53.1 99.5 1.95 82.1
263.73 3 0 1.7
C4-11 60.0 51.8 103.4 2.36 2.8
255.70 4 0 1.72
C4-12 -4.7 -29.3 81.7 91.1, 92.6
234.68 5 0 -0.76
C4-13 5.1 7.3 27.2 62.3, 67.3
296.77 6 1 -1.03
C4-14 45.2 56.8 89.1 70.3, 75
254.72 4 1 0.69
C4-15 65.8 34.0 98.2 -6.8, -5.5
263.73 3 0 2.17
C4-16 66.3 30.0 32.6 86.3, 89.8
330.84 3 0 3.8
C4-17 69.3 70.2 0.902 29
373.84 5 0 3.42
C4-18 4.5 18.7 14.2
366.81 6 2 0.65
C4-19 2.1 -20.7 11.8 67.3
330.82 5 2 1.54
C4-20 9.3 27.6 -7.3 10.4, 6.3
509.73 9 1 0.67
C4-21 9.8 20.6 23.3 50.6, 54.9
428.96 5 2 3.03
46
Figure 4.9 Cluster 5 - Cyclized 3
0
-chloroacetamide
N
O
Cl
N
N
N
O
Cl
C5-1
C5-2
N
N
O
F
Cl
C5-3
N
N
O
Cl
HO
C5-4
N
N
O
Cl
O
C5-5
F
3
C
N N
N
O
Cl
Cl
C5-6
N
N
O
F Cl
C5-7
N
N
N
O
Cl
C5-8
N
N
O
Cl
C5-9
N
S
N
O
Br
Cl
C5-10
N
N
N
S
O
Cl
C5-11
S
N
N
O
O
O
Cl
C5-12
S
N
N
N
O
O
O
Cl
C5-13
S
N
N
O
O
O
F
F
Cl
C5-14
S
N
N
O
O
O
O
O
Cl
C5-15
S
N
N
O
O
O
Cl
C5-16
N
O
O
O Cl
C5-17
S
N
N
O
O
O
Cl
C5-18
N
N
O
Cl
C5-19
S
N
O
O
H
2
N O
Cl
C5-20
N
N
O
O
O
Cl
O
C5-21
N
N
S
O
O
Cl
C5-22
N
N
O
Cl
Cl
Cl
C5-23
N
O
O
O NH
2
Cl
C5-24
O
O
47
Table 4.6 Screening data for Cluster-5 - Cyclized 3
0
-chloroacetamide
GSTO binding assay
(% Inhibition at 10 µM)
GSTO substrate assay
% Inhibition of
HCT116p53+/+ cell
proliferation at 10 µM
MW HBA HBD LogP
Recombinant Endogenous % I at 10 µM IC
50
(µM)
C5-1 67.5 83.5 0.048 93.4
251.76 2 0 3.13
C5-2 52.5 66.2 58.2 0.559 40.2, 61.1
243.74 4 0 0.61
C5-3 9.4 61.5 59.8 28.2
256.71 3 0 1.76
C5-4 13.5 33.4 41.0 81.9, 84.4
254.72 4 1 0.92
C5-5 44.2 51.5 78.6 -16.3
268.74 4 0 1.67
C5-6 70.2 94.6 0.191 56.2
342.15 4 0 2.46
C5-7 65.1 53.1 72.8 0.338 -2.0, -4.5
270.74 3 0 1.9
C5-8 62.3 75.7 93.8 0.293 82.6, 87.3
277.76 4 0 1.07
C5-9 62.3 52.8 83.7 0.381 68.4, 79.3
302.81 3 0 2.95
C5-10 59.8 55.8 68.3 0.411 71.5, 90.4
337.67 3 0 2.39
C5-11 57.7 62.9 5.7 19.6
283.78 4 0 0.87
C5-12 66.0 47.1 84.9 7.3
330.84 5 0 1.06
C5-13 61.1 66.2 98.1 0.156 81.1, 55.9
327.79 6 0 0.26
C5-14 61.2 60.9 91.8 0.496 28.2, 11.4
338.76 5 0 1.21
C5-15 62.4 63.7 0.262 82.7, 87.3
360.82 7 0 0.34
C5-16 82.7 61.7 2.4 48.3, 46.5
328.82 5 0 0.71
C5-17 9.4 69.5
269.73 4 0 1.81
C5-18 69.7 72.7 0.164 91.6, 91.4
316.81 5 0 1.37
C5-19 72.2 40.0 84.6, 75.2
240.65 4 0 1.8
48
C5-20 68.1 61.0 2.29 80.5, 89.8
288.75 5 1 0.8
C5-21 65.0 28.9 40.0 78.2, 83.6
388.85 6 0 3.87
C5-22 68.6 22.7 1.0 41.2
294.76 4 0 2.49
C5-23 67.2 16.6 42.2
433.77 3 0 6.01
C5-24 54.9 69.2 83.7 4.11 79.8, 87.3 254.67 5 1 0.15
49
Table 4.7: Inhibitory activities of selected GSTO inhibitors
Cpd Structure Substrate
assay
Competitive Binding assay Selectivity
Endogenous GSTO
IC
50
(nM)
a
IC
50
(nM) at 20 µM
4a
3139 ± 266 3900 Yes
C1-24
148 ± 34 N.T.
b
N.T.
C1-27
31 ± 6 21 No
C1-28
64 ± 16 8 No
C1-29
30 ± 5 44 No
C1-31
221 ± 77 7 Yes
C2-10
62 ± 10 190 No
C2-16
66 ± 13 520 No
C2-18
119 ± 27 75 No
C2-22
136 ± 18 N.T. N.T.
C2-26
125 ± 39 N.T. N.T.
a
IC
50
values are reported as mean ± SEM calculated from at least 3 independent experiments.
b
N.T. – Not tested
F
F
H
N
O
Cl
NH
O
Cl
S
O
O
H
N
O
F
F
Cl
S
N
O
O
NH
O
Cl Cl
S N
O
O
HN
O
Cl
S
N H
N
O
O
O
O
Cl
Cl
Cl
Cl
S
N
H
N
O
O
O
O
Cl
S
N
H
COOMe
O
Cl
Cl
S
NH
COOEt
O Cl
S
H
N
O
O
H
2
N
Cl
N
NO
2
N
H
S
O
Cl
N
NH
S
O
Cl
Cl
50
Interestingly, some of the top hits from this large unbiased screen belonged to the
N-(3-sulfamoyl)phenyl chloroacetamide class of compounds. Among them, C1-27 was
the most potent and selective GSTO1 inhibitor (Fig. 4.10a). C1-27 potently inhibited
GSTO1 enzyme activity with an IC
50
value of 31 nM (Fig. 4.10b,c). It potently competes
with the CMFDA probe for binding to recombinant protein as well as endogenous
GSTO1 in the milieu of soluble proteome (Fig. 4.10d,e).
Figure 4.10 Identification and characterization of C1-27 as a potent GSTO1
inhibitor.
(a) Chemical structure of lead inhibitor C1-27. (b) Time course of GSTO1-catalyzed 4-
NPG reduction in the presence of a range of C1-27 concentrations. (c) Plot depicts
concentration dependent inhibition of GSTO1 activity by C1-27, calculated from
absorbance values at T = 30 min. (d) C1-27 potently competes with CMFDA for binding
to recombinant GSTO1. Recombinant GSTO1 was incubated with indicated
concentrations of C1-27 for 30 min, followed by addition of CMFDA (500 nM) for 30
min and resolved by SDS-PAGE. (e) C1-27 inhibits CMFDA binding to endogenous
GSTO1. HCT116 cells were incubated with indicated concentrations of C1-27 for 2 h,
followed by addition of CMFDA (500 nM) for 1 h. Soluble proteome was resolved by
SDS-PAGE. Competitive inhibition of CMFDA binding to recombinant or endogenous
GSTO1 was assessed by in-gel fluorescence scanning.
51
We further performed ligand binding studies using a thermal shift assay. Protein
unfolding curves in presence and absence of C1-27 is shown in Fig. 4.11a. We observed
a negative shift in the thermal denaturation curve to lower melting temperature upon
inhibitor binding (Fig. 4.11b). Such negative shifts are seen with ligands that bind
through a covalent mechanism. The binding energy and unfavorable strains generated
through covalent bond formation are thought to promote protein destabilization.
Figure 4.11 C1-27 binds to GSTO1 and promotes protein destabilization.
(a) Thermal denaturation curves for GSTO1 in the presence of C1-27 showing a decrease
in melting temperature. (b) Plot of C1-27 concentration versus melting temperature T
m
showing a decrease in melting temperature with increasing ligand concentrations.
C1-27
75µM
37.5µM
18.75µM
9.37µM
1.5%DMSO
Temperature
Fluorescence intensity
& & & && && & & & & &
Tm = 65.16 0.17
Tm = 67.49 0.3
Tm = 68.26 0.23
Tm = 68.97 0.16
Tm = 70.76 0.19
-5.5 -5.0 -4.5 -4.0
64
66
68
70
Log Concentration (M)
Tm (°C)
a) b)
52
4.3.4 Characterization of lead GSTO1 inhibitors
C1-27 shows covalent binding to active site cysteine
Crystallographic studies were performed by our collaborator, Dr. Jeanne Stuckey,
Life Institute, University of Michigan, to determine the strucutres of GSTO1 in complex
with our potent inhibitors. Co-crystal structures of GSTO1 with C1-27 (Fig. 4.12a), C1-
31 and C4-10 revealed a covalent interaction between the chloroacetamide moiety and
active site cysteine (C32) in accordance with our initial premise. In addition, several
hydrophilic and hydrophobic interactions with amino acid residues lining the G-site and
H-site were also identified (Fig. 4.12b). C1-27 forms extensive interactions with these
amino acid residues, which explains its tight binding and potent GSTO1 inhibition.
Figure 4.12 Binding pose of C1-27 determined by crystallographic studies.
(a) Crystal structure of GSTO1-bound C1-27 reveals covalent interaction with active site
C32. (b) Shifts in side chain orientations due to C1-27 binding depicted here.
53
C1-27 shows rapid cellular uptake and GSTO1 binding
To understand the kinetics of inhibitor binding to GSTO1, we analyzed the cellular
uptake of C1-27 and its association with endogenous GSTO1. We observed that C1-27
was rapidly taken up by cells with more than 50% GSTO1 binding occurring within 5
min (Fig. 4.13).
Figure 4.13 C1-27 rapidly binds to GSTO1.
HCT116 cells were incubated with C1-27 (100 nM) for indicated durations, followed by
washout and incubation with CMFDA (500 nM). Binding to endogenous GSTO1 (lower
band) was assessed by the competitive binding assay.
C1-27 shows target occupancy up to 10 h
To further examine the nature of GSTO1-inhibitor complexes, we followed the
time course of C1-27 dissociation from GSTO1. CMFDA labeling of GSTO1 was
inhibited by C1-27 up to 10h and recovered completely by 48h, indicating sustained
54
target occupancy (Fig. 4.14a). On the other hand, 4a showed a prolonged target
occupancy, even at 24 h (Fig. 4.14b).
Figure 4.14 Duration of GSTO1 inhibition by C1-27.
HCT116 cells were incubated with (a) C1-27 (100 nM) or (b) 4a (5 µM) for 2 h,
followed by washout. At different time points after washout, CMFDA binding to
endogenous GSTO1 (lower band) was assessed by the competitive binding assay.
GSTO1 inhibition by C1-27 is reversible
We analyzed the dissociation of C1-27 from GSTO1 using a pre-incubation and
dilution paradigm to further understand the mechanism of inhibition (Fig. 4.15a). 80% of
GSTO1 activity was recovered after pre-incubation with C1-27 at 200 nM followed by a
large dilution to 5 nM, indicating a rapid dissociation (Fig. 4.15c). These results suggest
that C1-27 forms a tight albeit reversible association with the active site cysteine (C32).
4a, on the other hand, exhibited an irreversible mode of inhibition (Fig. 4.15b). Other
55
structurally similar analogs of C1-27 also showed a rapid dissociation, indicating a
reversible covalent binding (Fig. 4.15d,e). Figure 4.15f summarizes the inhibition
modality of selected GSTO1 inhibitors.
Figure 4.15 Reversibility of GSTO1 inhibition by C1-27.
(a) Schematic of the experiment to analyze reversibility of inhibition. Recombinant
GSTO1 was incubated with inhibitor for 30 min at 10 k
i
followed by a rapid large
dilution to 10 nM 0.25 k
i
. Substrate assay was performed and % recovery of GSTO1
enzyme activity was compared to enzyme inhibition without dilution. (b) Compound 4a
shows an irreversible mode of GSTO1 enzyme inhibition. On the other hand, compounds
a
c d
f
!! !!
!!
!
!!
!!
Pre-incubation Large dilution
!!
!!
!!
!!
!!
Cpd Dissociation Inhibition modality
4a Slow Irreversible
C1-27 Rapid Reversible
C1-28 Rapid Reversible
C1-29 Rapid Reversible
C1-31 Rapid Reversible
C4-10 Slow Irreversible
C5-1 Rapid Reversible
b
e
56
C1-27, C1-28 and C1-29 (c-e) show a reversible mode of inhibition. (f) Table
summarizes the reversibility of GSTO1 inhibition by selected compounds.
C1-27 shows selectivity in GSTO1 inhibition
Since compounds with reactive electrophilic groups are often found to be non-
specific thiol-binders, we tested the selectivity of GSTO1 inhibition of our compounds in
the fluorescence-based binding assay using a colon cancer cell line proteome (HCT116).
While the fluorescent probe, CMFDA, labeled several other cysteine containing proteins
in the proteome, our compounds showed selective inhibition of the GSTO1 band up to 1
µM. C1-27 showed selective inhibition of GSTO1 starting at 1.4 nM. Even
concentrations up to 10 µM did not inhibit other proteins, suggesting selectivity for
GSTO1 (Fig. 4.16). Crystallography studies also show that C1-27 selectively targets only
the active site cysteine and not other non-catalytic cysteines in GSTO1.
Figure 4.16 Selectivity of GSTO1 inhibition.
(a) C1-27 selectively inhibits GSTO1. Soluble proteome from HCT116 cells was
incubated with indicated concentrations of C1-27 for 2 h, followed by addition of
CMFDA (500 nM). CMFDA binds to GSTO1 as well as other cysteine-bearing proteins
in the proteome. C1-27 specifically competes with CMFDA for binding to the GSTO1
57
band without affecting other proteins. (b) HCT116 soluble proteome was incubated with
indicated GSTO1 inhibitors at 20 µM and at 1 µM for 2 h, followed by addition of 500
nM CMFDA for 1 h. Selectivity of GSTO1 inhibition versus inhibition of CMFDA
binding to various other cysteine-bearing proteins in the proteome was analyzed.
Based on the in vitro inhibition and selectivity profile, we selected C1-27 for
further experiments to study the cellular effects of GSTO1 inhibition.
4.4 Discussion
GSTO1 is a novel GST isoform with a cysteine residue in the active site. It is
overexpressed in several cancers as well as in drug-resistant cell lines, although its
cellular functions are not fully understood. To gain a better insight into the role of
GSTO1 in cancer cells, we sought to identify and characterize in-depth novel GSTO1
inhibitors. Through extensive structure-activity relationships, screening, biochemical
assays, and X-ray crystallography, we have identified several potent and selective
GSTO1 inhibitors. This is also the first reported crystal structure of GSTO1 in complex
with a small molecule inhibitor. Crystal structures revealed a covalent association with
the active site cysteine (C32) and additional hydrophilic and hydrophobic interactions in
the H-site. Mechanistic studies with the lead inhibitor C1-27 further demonstrated the
reversible nature of this interaction. Such reversible covalent inhibitors offer potent
inhibition, prolonged duration of action and target selectivity, with minimal irreversible
off-target effects. These potent and selective GSTO1 inhibitors will serve as valuable
tools to investigate its role in cancer and other pathologies as well as uncover additional
functions in the cell. Studies are in progress to design clinically viable candidates through
58
structure-based drug design approaches using the co-crystal structures of our lead
inhibitors.
59
CHAPTER 5: EVALUATION OF GSTO1 INHIBITORS AS
NOVEL ANTICANCER AGENTS
5.1 Background
GSTO1 overexpression has been reported in many cancers and drug-resistant cell
lines. Results from our siRNA studies (Chapter 3) demonstrated the importance of
GSTO1 for the growth and proliferation of cancer cells. Previous studies with small-
molecule GSTO1 inhibitors have not sufficiently investigated their effect on cancer cells.
We, therefore, sought to evaluate in-depth the effects of our potent small-molecule
GSTO1 inhibitors (described in chapter 4) on cancer cell survival.
5.2 Materials and Methods
5.2.1 Cell lines. Colon cancer cell line HCT116 was generously provided by Dr. Bert
Vogelstein, Johns Hopkins Medical Institutions, Baltimore, MD. Pancreatic cancer cell
line Panc-1 was obtained from Dr. Alan Epstein, University of Southern California. All
other cell lines were purchased from the American Type Culture Collection (Manassas,
VA). Cell lines were cultured in RPMI 1640 or Dulbecco’s minimal essential media
supplemented with 10% fetal bovine serum at 37 °C in a humidified atmosphere of 5%
CO
2
.
60
5.2.2 Reagents. The following antibodies and reagents were purchased from Cell
Signaling Technology: phospho ERK, total ERK, phospho JNK, total IL-1β and cleaved
IL-1 β, MEK inhibitor (PD98089), PI3K inhibitor (LY294002) and JNK inhibitor
(SP600125). Antibodies to total JNK, Bcl-2 and β-tubulin were purchased from Santa
Cruz Biotechnology. Cisplatin was purchased from Sigma. CellTracker Green (CMFDA -
5-Chloromethylfluorescein diacetate) was purchased from Invitrogen. Erastin and
Piperlongumine were purchased from Sigma and LKT labs, respectively.
5.2.3 Cell viability assays. Cell proliferation was assessed by a 3-(4,5-dimethylthiazol-2-
yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Cancer cells were seeded in 96-well
microtitre plates and after overnight attachment, treated with GSTO inhibitors. After 72
h, MTT solution (3 mg/mL; 20 µL) was added to each well and cells were incubated for 3
h at 37 °C. After incubation, media from each well was removed and the dark blue
formazan crystals formed by live cells were dissolved in DMSO (150 µL/well). The
absorbance intensity was measured at 570 nm on a microplate reader (Molecular Devices,
Sunnyvale, CA, USA). Cell viability after 24 h treatment was assessed using ApoTox-
Glo triplex assay (Promega) according to manufacturer’s protocol. At least 3 independent
dose response experiments with each concentration tested in triplicate were performed for
each cell line.
5.2.4 Clonogenic survival assay. HCT116 cells were seeded at a density of 200
cells/well in a 6-well microtiter plate. After overnight attachment, cells were treated with
61
different concentrations of GSTO1 inhibitors for 24 h. Colonies were allowed to grow in
drug-free medium for 7 days and stained with 0.5% crystal violet. For drug combination
studies, HCT116 cells (300/well) were plated in a 12-well plate. C1-27, cisplatin (1 µM,
dissolved in 0.9% NaCl), PD98089 (10 µM), LY294002 (10 µM), SP600125 (1 µM) or
desferoxamine mesylate (10 µM) were added alone or in combination and colonies were
allowed to grow for 7-10 days.
5.2.5 Cell cycle analysis. Cells were seeded at a density of 0.5 x 10
6
/well in a 6-well
plate. After overnight attachment, cells were treated with C1-27 (5 µM) for 24 h and
harvested using trypsin. Cells were then washed twice with PBS containing 10% serum
and fixed with 70% ethanol overnight at -20 °C. Fixed cells were washed twice in PBS
containing 10% serum and stained with 50 µg/ml propidium iodide solution containing
100 µg/mL RNase A for 1 h at 37 °C. Cell cycle was analyzed on a BD LSR II flow
cytometer (BD Biosciences) and data was analyzed using ModFit software.
5.2.6 Caspase 3/7 assay. HCT116 cells were treated with indicated concentrations of C1-
27 for 24 h. Caspase activation was measured using ApoTox-Glo triplex assay reagent
according to manufacturer’s protocol.
5.2.7 Western blotting. HCT116 cells were seeded in a 6-well plate and treated with
GSTO1 inhibitors for 24 h. Cells were washed with ice cold PBS and lysed with a triple
detergent lysis buffer supplemented with protease and phosphatase inhibitor cocktail.
62
Lysate was sonicated and centrifuged at 12,000 rpm for 10 min at 4 °C to remove cell
debris. Total protein concentration in the supernatant was measured using BCA protein
assay kit (Pierce Biotechnology). 40 µg of protein lysate was incubated with laemmli
sample buffer for 5 min at 90 °C, resolved on a 10% polyacrylamide gel and
electroblotted on a PVDF membrane. Membranes were blocked in 5% non-fat milk in
Tris-buffered saline with 0.1% Tween-20 (TBST) for 1 h at RT and incubated overnight
with primary antibodies. Subsequently, the membranes were washed with TBST and
incubated with appropriate HRP-linked secondary antibodies for 2 h at RT followed by
further washing with TBST. The immunoblots were visualized using ECL Western
blotting substrate (Pierce Biotechnology) on a Chemidoc XRS imager (Bio Rad).
5.2.8 In vivo efficacy studies. HCT116 cells (1x10
6
) in exponential phase were injected
s.c. into the left flank of 8-10 weeks old female nude mice (25-30 g; Simonsen
laboratories, Gilroy, CA). All animal experiments were done in accordance with
protocols approved by the Institutional Animal Care and Use Committee. The
perpendicular diameters of the tumors were measured 3 times weekly using standard
calipers and tumor volumes were calculated using the formula: 0.5 x D x d
2
, where D and
d were the longest and shortest perpendicular diameters, respectively. Tumors were
allowed to grow to a volume of 50 mm
3
and mice were treated via i.p. administration
with either vehicle (n=5) or C1-27 dissolved in peanut oil (n=3; 20 mg/kg/day for the
first 2 weeks with a 5 d on and 2 d off treatment schedule, the dose was then increased to
25 mg/kg/day for the next 23 d and further escalated by 5 mg/kg/day to a final dose of 45
63
mg/kg for the remaining duration of treatment). Tumor volumes and body weights were
measured 3 times weekly to monitor tumor burden and weight loss during treatment. At
the end of the experiment, animals were euthanized and tumor, kidney and liver were
collected, fixed, and paraffin embedded for histology.
5.2.9 Statistical Analysis. Data are presented as mean of at least 3 independent
experiments and statistical analysis was done using Prism 6.0 (Graphpad software).
Groups were analyzed by either t-test or 1-way ANOVA followed by Dunnett’s post hoc
test. p values at α=0.05 are reported (* p<0.05, ** p<0.01, *** p<0.001, **** p<0.0001).
5.3 Results
5.3.1 GSTO1 inhibitors block cancer cell proliferation
Results from our siRNA studies indicate that GSTO1 is important for the growth
and proliferation of cancer cells. We, therefore, evaluated the effect of inhibiting GSTO1
activity using small-molecule compounds on cancer cell survival. Treatment of HCT116
cells with C1-27 decreased cell viability in a dose-dependent manner (Fig. 5.1a and
Table 5.1). We further observed significant cytotoxicity of the top GSTO1 inhibitors in a
panel of different cancer cell lines (Fig. 5.1b and Table 5.1). Treatment with C1-27 also
inhibited the clonogenic survival of HCT116 cells at sub-micromolar concentrations (Fig.
5.2c). These findings are consistent with the results of siRNA inhibition.
64
Furthermore, at sub-cytotoxic concentrations, the size of the colonies was
reduced, indicating an effect on cell proliferation. To confirm this, we analyzed the effect
of C1-27 on cell cycle progression. HCT116 cells were treated with 5 µM of C1-27 for
24h and cell cycle distribution was evaluated by flow cytometry. C1-27 treatment
resulted in an increase in the S-phase population, with a concomitant decrease in the
G0/G1 population (Fig. 5.2d).
Figure 5.1 Inhibition of cancer cell proliferation by GSTO1 inhibitors.
C1-27 inhibits cancer cell proliferation in different colon cancer cells, assessed after 24 h
(a) and 72 h treatment (b). HCT116 cells were more sensitive to C1-27 treatment. (c) C1-
27 inhibits clonogenicity of HCT116 cells. Representative experiment of at least three
independent experiments is shown here. (d) Cell cycle analysis of C1-27 vs. DMSO-
treated HCT116 cells. HCT116 cells, treated with 5 µM C1-27 for 24 h, show a decrease
a b
c d
DMSO 0.12 µM 0.36 µM
3.3 µM 10 µM 1 µM
HCT116
65
in G0/G1 and an increase in S-phase populations. Percent of PI-labeled cells in each cell
cycle phase is shown as mean ±
SD of three independent experiments.
Table 5.1: Cytotoxicity of selected GSTO inhibitors in a panel of cancer cell lines
a
GI
50
is the drug concentration that causes a 50% reduction in cell proliferation.
GI
50
values are reported as mean ± SEM calculated from at least 3 independent
experiments.
b
N.T. – Not tested
Cpd
GI
50
a
(µ µM)
HT29 HCT116 H460 NCI
ADR Res
4a 5.6 ± 0.5 2.4 ± 0.2 3.5 2.15
C1-24 5.1 ± 1.3 3.3 ± 1.1 N.T.
b
N.T.
C1-27 4.3 ± 0.6 1.2 ± 0.6 3.25 3.675
C1-28 2.7 ± 0.5 2.2 ± 0.7 4.5 3.8
C1-29 15.5 ± 0.2 13.7 ± 1.3 N.T. N.T.
C1-31 8.7 ± 1.2 9.1 ± 1.1 >20 >20
C2-10 18.1 ± 1.6 18.8 ± 1.6 N.T. N.T.
C2-16 19.8 >20 N.T. N.T.
C2-18 9.5 ± 2.8 6.9 ± 1.9 N.T. N.T.
C2-22 10.9 ± 1.6 7.4 ± 1.9 >20 N.T.
C2-26 14.2 6.2 ± 1.8 >20 14.6
66
5.3.2 GSTO1 inhibitor C1-27 induces apoptotic cell death
Inhibition of cell proliferation was accompanied by an activation of caspases and
reduction of anti-apoptotic BCL2 expression (Fig. 5.2a and 5.2b).
Figure 5.2 C1-27 induces an apoptotic cell death.
(a) Induction of apoptosis by C1-27 treatment for 24 h, as assessed by a Caspase-Glo
assay. Data are mean ±
SD of triplicate wells. (b) Western blotting of HCT116 cells
further shows a suppression of anti-apoptotic protein Bcl-2 expression in response to C1-
27 treatment for 24 h.
5.3.3 Cell death induced by GSTO1 inhibitor C1-27 is Ras-signaling dependent
Among the three colon cancer cell lines that we tested, HCT116 was most
sensitive to C1-27 treatment. Interestingly, HCT116 has a mutant KRAS (G13D). We
tested other cancer cell lines with KRAS mutations and found a similarly increased
sensitivity (>3-fold) to C1-27 when compared to cells containing wild-type KRAS
(Table 5.2). Ras-selective compounds such as erastin and piperlongumine were also
tested for comparison. The KRAS-selectivity of C1-27 was comparable to that of
piperlongumine. We next studied the effect of C1-27 on Ras signaling pathway.
a b
DMSO
0.5
1
2.5
5
10
Bcl2
25
β-Tubulin 55
C1-27 (µM)
67
Treatment with C1-27 induced a sustained phosphorylation of ERK, while JNK
phosphorylation occurred at higher doses (Fig. 5.3a,b).
Table 5.2. Effect of C1-27 treatment in a panel of cancer cell lines.
Cancer type Cell line
GI
50
(µM)
a
RAS
status
C1-27 Erastin Piperlongumine
Colon cancer
HT29 4.3 ± 0.6 42.8 ± 3.8 5.1 ± 0.5
Wild
type
HCT116 1.2 ± 0.6 15.9 ± 0.7 2.6 ± 0.2
KRAS
Mutant
Pancreatic
cancer
BxPC3 5.5 ± 1.0 > 50 6.8 ± 1.4
Wild
type
Panc-1 1.8 ± 0.1 0.5 ± 0.2 4.9 ± 0.8
KRAS
Mutant
Lung cancer
A549 11.7 >50 16.9
KRAS
Mutant
H1299 1.9 1.2 2.9
NRAS
mutant
a
GI
50
is the drug concentration that causes a 50% reduction in cell proliferation. GI
50
values are reported as mean ± SEM calculated from at least 3 independent experiments.
To further investigate whether the cell death induced by C1-27 was Ras-
dependent, we treated HCT116 cells with inhibitors of the Ras signaling pathway. C1-27-
induced cell death was suppressed by the MEK inhibitor PD98089 but not by the JNK
inhibitor SP600125. PI3K inhibitor (LY294002) also rescued C1-27-induced cell death
(Fig. 5c). These results suggest a putative role for Ras-activated signaling pathways in
C1-27-induced cell death.
68
Figure 5.3 Cell death induced by C1-27 is Ras-signaling dependent.
Western blotting of HCT116 cells an induction of (a) pERK and (b) pJNK in response to
C1-27 treatment for 24 h. (c) Cell death mediated by C1-27 is MEK and PI3K-dependent
and JNK-independent. HCT116 cells were treated with C1-27 alone or in combination
with PD98089 (10 µM), LY294002 (10 µM) or SP600125 (1 µM) and colonies were
grown for 7-10 days. Percent colony survival depicted are mean ±
SD of 3 independent
experiments.
5.3.4 Effect of iron chelator on C1-27 induced cell death
Compounds containing a α-chloroacetamide moiety have been reported to show
an oxidative, KRAS-selective, MEK-dependent cell death, which was non-apoptotic in
nature (Weiwer et al., 2012; Yang and Stockwell, 2008). Interestingly, this cell death was
also iron-dependent - iron chelator, desferroxamine mesylate (DFO) inhibited cell death.
Based on some of the similarities we observed in C1-27 induced cell death, we next
examined the role of iron in C1-27 cytotoxicity. We treated HCT116 cells with C1-27 in
combination with DFO. Contrary to the previous findings with other Ras-selective
a c
DMSO
0.5
1
2.5
5
10
pJNK
JNK
β-Tubulin
C1-27 (µM)
55
40
55
40
55
0.5
1
2.5
5
10
pERK
β-Tubulin
C1-27 (µM)
55
44
DMSO
b
69
chloroacetamide compounds, treatment with DFO surprisingly enhanced C1-27
cytotoxicity (Fig. 5.4).
Figure 5.4 Enhancement of C1-27-mediated cell death by DFO.
HCT116 cells were treated with C1-27 (1 µM) and DFO (10 µM) for 24 h.
Representative colony formation assay is shown on the left. Percent colony survival after
treatment is shown on the right. Data shown are mean ±
SD of 3 independent
experiments.
5.3.5 GSTO1 inhibition enhances cisplatin cytotoxicity
GSTO1 overexpression has been implicated in drug resistance to platinum
compounds (Piaggi et al., 2010; Yan et al., 2007). Therefore, we tested the effects of
GSTO1 inhibitors in combination with cisplatin in HCT116 cells. C1-27 was able to
significantly enhance cisplatin-induced cytotoxicity in a clonogenic assay, even at non-
cytotoxic concentrations (Fig. 5.5).
DFO
(10 µM)
DMSO C1-27 (1µM)
Vehicle
70
Figure 5.5 C1-27 enhances cisplatin-induced cytotoxicity.
HCT116 cells were treated with indicated concentrations of C1-27 and cisplatin for 24 h.
Representative colony formation assay is shown on the left. Percent colony survival was
assessed. Data are mean ±
SD of 3 independent experiments (Right).
5.3.6 C1-27 inhibits human colon cancer xenograft growth in vivo
We next assessed the in vivo efficacy of C1-27 in a human colon cancer xenograft
model. C1-27 (20-45 mg/kg) was administered as a single agent to nude mice bearing
HCT116 xenografts. After 5 weeks of treatment, tumor growth was significantly
inhibited in C1-27-treated mice compared to the vehicle-treated group (P < 0.001) (Fig.
5.6a,b). H&E staining of the tumor sections from C1-27-treated mice showed extensive
regions of necrosis compared to control (Fig. 5.6c). C1-27 treatment was generally well
tolerated by mice up to 45 mg/kg, with no obvious signs of toxicity. To further assess any
treatment-related toxicity, kidney and liver tissue sections from control and C1-27-treated
mice were examined by histology. H&E staining showed no gross morphological
differences (Fig. 5.6d). The average body weight of the control or C1-27-treated mice
also did not vary significantly throughout the duration of the study (Fig. 5.6e).
C1-27 (µM) - 0.5 1
+ Cisplatin (1 µM)
71
Figure 5.6. GSTO1 inhibitor C1-27 inhibits tumor growth in vivo.
HCT116 cells were injected s.c. into the flanks of nude mice. Tumor-bearing mice were
treated with C1-27 or vehicle. (a) Tumor progression in mice treated with vehicle (n=5)
or C1-27 (n=3). Data shown are mean tumor volumes (Error bars = SEM). Treatment
72
schedule is described in the methods section. (b) Significant reduction in tumor volumes
on days 36 and 41 of C1-27 treatment. Mean
± SEM shown. p < 0.05 by two-tailed test.
(c) C1-27-treated tumors show extensive areas of necrosis, as seen by histological
analysis of tumor sections from C1-27 and vehicle-treated mice with hematoxylin/eosin
staining. Representative image of 4 independent sections is shown. (d) No gross signs of
systemic toxicity were observed. Liver and kidney sections from C1-27 and vehicle-
treated mice were assessed by hematoxylin/eosin staining. (e) Comparison of body
weights of C1-27 and vehicle-treated mice over the duration of treatment. (f) C1-27 binds
to GSTO1 in vivo. Tumor tissue homogenates of C1-27 and vehicle-treated mice were
incubated with CMFDA. GSTO1 binding was assessed as described previously.
These results indicate that C1-27 reduced tumor growth without any obvious
adverse effects to normal tissues. We further examined C1-27 binding to GSTO1 in vivo
using tumor tissue homogenates. CMFDA binding to GSTO1 was inhibited in C1-27-
treated tumors, but not in the control group (Fig. 5.6f). These results further support the
fact that C1-27 selectively targets GSTO1 in tumors in whole animals without apparent
toxicity.
5.4 Discussion
The lead GSTO1 inhibitor, C1-27, reduced cell viability in a panel of cancer cell
lines. We further observed that C1-27 caused an accumulation of cells in the S-phase of
cell cycle and induced apoptotic cell death. We determined in vivo efficacy of C1-27 in a
xenograft model of colon cancer. C1-27 inhibited tumor growth without causing
significant toxicity to organs. We also demonstrated that C1-27 targeted GSTO1 in the
tumor mass.
73
Interestingly, we observed that KRAS mutant cancer cell lines were more
sensitive to GSTO1 inhibition than cells with wild-type KRAS. The cell death induced by
C1-27 also appeared to be Ras-signaling dependent. Several oncogenic RAS-selective
compounds, such as piperlongumine, lanperisone, erastin and RSL3, have been reported
recently (Dixon et al., 2012; Raj et al., 2011; Shaw et al., 2011; Yang and Stockwell,
2008), although the characteristics of cell death elicited by them differ. For example,
RSL3, which also contains a tertiary α-chloroacetamide, induces a unique iron-
dependent, oxidative, non-apoptotic cell death, whereas piperlongumine induces an
oxidative apoptotic cell death (Raj et al., 2011). Interestingly, GSTO1 has been identified
to be one of the potential targets of piperlongumine. Furthermore, we observed that the
KRAS-selectivity of C1-27 was comparable to piperlongumine. However, the
mechanism of cell death by C1-27 does not appear to be iron-dependent and involves
activation of caspase 3 and 7 with decreased BCl2 expression. Further studies are needed
to understand the role, if any, of GSTO1 in KRAS-selective cytotoxicity. The possibility
of KRAS-selective cytotoxicity as a feature common to chloroacetamides independent of
GSTO1 should also be examined.
Our studies also show that GSTO1 inhibitors increase the sensitivity of cancer
cells to cisplatin. Given, that GSTO1 overexpression has been implicated in
chemotherapeutic drug resistance, GSTO1 inhibitors will be useful in re-sensitizing drug
resistant cells. In sum, our studies with C1-27, a potent GSTO1 inhibitor, validate
GSTO1 as an important target in oncology and demonstrate the therapeutic utility of
GSTO1 inhibitors.
74
CHAPTER 6: EFFECT OF GSTO1 INHIBITORS ON IL-1 β β
SECRETION
6.1 Background
Recent studies have proposed a novel
role for GSTO1 in the regulation of IL-
1β processing and release. IL-1β is a pro-
inflammatory cytokine and a master regulator
of the inflammation cascade secreted
primarily by monocytes and macrophages.
Activation of IL-1β is regulated at two levels.
A first inflammatory stimulus such as
lipopolysaccharide (LPS) induces the
transcription of 31 kDa pro-IL-1β through the
activation of Toll-like receptors and NF-κB.
Pro-IL-1β is biologically inactive and requires
processing into mature 17 kDa IL-1β by
caspase-1. A second secretion signal such as
extracellular ATP is required to trigger caspase-1 activation as well as processing and
release of mature IL-1β.
PYD CARD
PYD CARD
Inflammasome activation
NLRs
or AIM2
ASC
Pro-Caspase-1
Pro-IL-1β β" "
Mature-IL-1β β" "
Processed
Caspase-1
PYD
CARD
ASC
oligomerization
PYD CARD
PYD CARD
PYD CARD
CARD
PYD
ATP
GS TO%
Figure 6.1 Inflammasome activation
cascade
75
The activation of pro-caspase-1 into caspase-1 is facilitated by a large assembly of
multiprotein complexes called inflammasomes. Depending on the constituent proteins,
inflammasomes can be classified into NLRP1
1
, NLRP3, NLRC4, NLRP6 and AIM2
2
inflammasomes. Multiple inflammatory stimuli and other factors interact and activate
different inflammasome complexes. The NLR proteins interact with other adaptor
proteins such as ASC (Apoptosis-associated Speck-like protein containing a CARD
3
)
through the PYD
4
death-fold domain (Fig. 6.1). ASC in turn recruits pro-caspase-1 to the
inflammasome through its CARD domain and facilitates its activation into caspase-1
through self-cleavage (Franchi et al., 2009; Latz et al., 2013; Mariathasan and Monack,
2007). Activated caspase-1 then cleaves pro-IL-1β into mature IL-1β. Recent studies
have identified GSTO1 as the cellular target of Cytokine Release Inhibitory Drugs
(CRIDs) (Laliberte et al., 2003). The diarylsufonylurea-based CRIDs inhibit IL-
1β processing and secretion, inhibit caspase activation and also preserve plasma
membrane latency. Through affinity chromatography and radiolabeling, CRIDs were
shown to bind to the cysteine residue in GSTO1. More recently, GSTO1 has been shown
to colocalize with ASC (Coll and O'Neill, 2011), though its precise role in the
inflammasome assembly and pro-IL-1β processing is not yet clear.
1
NLR – NOD-like receptor containing NACHT, Leucine rich repeats (LRR), PYD and
2
AIM2 – Absent in melanoma 2
3
CARD – Caspase activation and recruitment domain
4
PYD – Pyrin domain
76
6.2 Materials and Methods
6.2.1 IL-1β β ELISA. THP-1 cells (2.5 x 10
4
/well) were seeded in a 96-well microtiter
plate and treated with 10 nM phorbol myristate acetate overnight. Attached THP-1 cells
were stimulated with 200 ng/mL LPS (Sigma) for 4 h and treated with DMSO or test
compounds in serum free media for 1 h followed by addition of 5 mM ATP (Sigma) for 3
h. Supernatant media was analyzed for IL-1β content by ELISA (BioLegend) according
to manufacturer’s protocol.
6.2.2 Western blotting. THP-1 cells (1 x 10
6
/well) were seeded in a 12-well plate and
differentiated with 10 nM phorbol myristate acetate overnight. Attached THP-1 cells
were stimulated with 100 ng/mL lipopolysaccharide (LPS, Sigma) for 3 h and treated
with DMSO, GSTO1 inhibitors or Glyburide (Enzo Life sciences) in serum free media
for 1.5 h followed by addition of 5 mM ATP (Sigma) for 30 min. Supernatant media was
collected, precipitated using TCA and dissolved using Tris pH 8.0. The sample was then
incubated with laemmli sample buffer for 5 min at 90 °C, resolved on a 10%
polyacrylamide gel and electroblotted on a PVDF membrane. Membranes were blocked
in 5% non-fat milk in Tris-buffered saline with 0.1% Tween-20 (TBST) for 1 h at RT and
incubated overnight with primary antibodies. Subsequently, the membranes were washed
with TBST and incubated with appropriate HRP-linked secondary antibodies for 2 h at
RT followed by further washing with TBST. The immunoblots were visualized using
ECL Western blotting substrate (Pierce Biotechnology) on a Chemidoc XRS imager (Bio
Rad).
77
6.2.3 Bioinformatics analysis. IL-1β gene expression in normal and cancer tissues was
analyzed using Oncomine cancer microarray database (www.oncomine.org) across
various cancer subtypes using a cancer versus normal differential analysis. For
correlation analysis, normalized median-centered values for GSTO1 and IL-1β
expression in each microarray study was plotted and Pearson’s correlation coefficient
was computed using Prism 6.0. As a negative control, correlation between GSTP1 and
IL-1β expression in the same datasets was also analyzed.
6.3 Results
6.3.1 GSTO1 inhibitors block IL-1β β secretion from activated THP-1 cells
LPS-primed monocytic THP-1 cells were treated with potent lead compounds and
IL-1β secretion in response to ATP stimulation was measured by ELISA and Western
blotting. Several of the potent GSTO1 inhibitors inhibited the secretion of mature IL-1β
(Fig. 6.2 a,b). C1-27 inhibited ATP-induced IL-1β release with an IC
50
of 3.5 µM (Fig.
6.2 c). Interestingly, we also found that pre-treatment of THP-1 cells with C1-27
inhibited the induction of pro-IL-1β by LPS (Fig. 6.2 d). This suggests that in addition to
inhibiting IL-1β processing and release post LPS induction, C1-27 also has an effect at
transcriptional level preventing pro-IL1β induction by LPS.
78
Figure 6.2 GSTO1 inhibitors suppress IL-1β β secretion.
(a) ATP-induced IL-1β secretion from LPS-stimulated THP-1 cells was analyzed by
ELISA. THP-1 cells were stimulated with LPS (200 ng/mL) for 4 h and treated with 20
µM of indicated GSTO1 inhibitors in serum-free media for 1 h followed by addition of 5
mM ATP for 3 h. Cytokine secretion level is expressed as a percentage of that released
from control cells treated with DMSO. Data are mean readings of duplicate wells. There
was less than 15% decrease in cell viability during the 4 h treatment with most of the
compounds. (b) Inhibition of IL-1β secretion by GSTO1 inhibitors was further confirmed
by Western blotting. Supernatant media from LPS-primed THP-1 cells treated with
GSTO1 inhibitors (20 µM) or Glyburide (GLY, 50 µM) for 1.5 h and ATP (5 mM) for 30
min was concentrated and precipitated using TCA and immunoblotted for mature IL-1β
(17 kDa). (c) Dose-dependent inhibition of IL-1β release by C1-27. LPS-primed THP-1
cells were treated with indicated concentrations of C1-27 or GLY (50 µM) for 1.5 h,
followed by stimulation with ATP (5 mM) for 30 min. Concentrated supernatant was
analyzed by western blotting using anti IL-1β antibody. Representative blot from one of
three independent experiments shown. (d) C1-27 inhibits pro-IL-1β production at higher
concentrations. THP-1 cells were treated with indicated concentrations of C1-27 or GLY
(50 µM) for 30 min, followed by LPS stimulation for 3 h. Expression of pro-IL-1β in
response to LPS was analyzed by western blotting. Representative blot from one of two
independent experiments shown.
DMSO
C1-27 (µM)
DMSO
DMSO
0.1
1
5
10
20
GLY
LPS
ATP
Supernatant
IL-1β"
p17
Cell lysate
Pro-IL-1β p31
C1-27 (µM)
DMSO
20
10
5
1
0.1
GLY
LPS
β-Tubulin p45
DMSO
DMSO
DMSO
4a
C1-27
C1-29
C5-1
C2-10
C2-16
C1-28
C2-22
GLY
IL-1β"
Supernatant "
LPS"
ATP "
Coomasie
a b
c d
79
6.3.2 Co-expression of GSTO1 and IL-1β β in cancer
Aberrant IL-1β expression has been observed in several cancer types (Deans et
al., 2006; Elaraj et al., 2006; Qin et al., 2011; Russell et al., 2011). Such an increase in
IL-1β levels promotes cancer cell migration, angiogenesis and metastasis (Voronov et al.,
2003; Yano et al., 2003). Furthermore, tumor-associated macrophages also act as a source
of in the tumor microenvironment (Pollard, 2004). We analyzed the expression of
GSTO1 and IL-1β in several cancers using Oncomine and observed a significant positive
correlation between GSTO1 overexpression and IL-1β overexpression in many invasive
cancer types (Fig. 6.3).
80
Figure 6.3. IL-1β β expression strongly correlates with GSTO1 expression.
(a-f) Representative plots of several gene expression datasets in Oncomine analyzed
shown. Each point represents the expression in a single patient of GSTO1 (x-axis) and
IL-1β (y-axis). Pearson’s correlation coefficient R is indicated. (g,h) Lack of correlation
between GSTP1 and IL-1β indicating specificity of co-expression.
81
6.4 Discussion
Recent studies have proposed an exciting role for GSTO1 in the regulation of IL-
1β processing and secretion, although the exact mechanism of this regulation is not clear
(Coll and O'Neill, 2011; Laliberte et al., 2003). Our results provide further proof-of-
concept that GSTO1 inhibition using small-molecule inhibitors blocks IL-1β release from
activated monocytic cells. This is the first report of α-chloracetamide containing
compounds as IL-1β release inhibitors. Inhibition of IL-1β secretion by GSTO1
inhibitors has exciting implications on tumor progression and tumor-immune cells
crosstalk in the microenvironment. Several studies have reported elevated IL-1β levels in
cancers (Deans et al., 2006; Elaraj et al., 2006; Qin et al., 2011; Russell et al., 2011).
Such overexpression of IL-1β in cancer cells promotes metastasis via enhanced
expression of invasion and angiogenesis-related factors (Yano et al., 2003). In addition to
producing IL-1β directly, cancer cells also induce macrophages in the tumor
microenvironment to secrete IL-1β, which further promotes angiogenesis and metastasis
(Pollard, 2004; Vidal-Vanaclocha et al., 2000; Voronov et al., 2003). For example, colon
cancer cells have been found to induce IL-1β secretion from monocytic THP-1 cells,
which in turn promotes cell growth and migration via a Wnt-signaling dependent
pathway (Honda et al., 2011; Kaler et al., 2009a; Kaler et al., 2010; Kaler et al., 2009b).
Interestingly, we found a significant positive correlation between GSTO1 overexpression
and IL-1β overexpression in many invasive cancer types. Therefore, targeting the
inflammasome using GSTO1 inhibitors, to inhibit tumor cell-derived and tumor-
82
associated macrophage-derived IL-1β secretion, could be a potential and effective
approach to mitigate IL-1β-mediated pro-oncogenic effects (Zitvogel et al., 2012). In
addition, GSTO1 inhibitors could have potential applications in inflammatory diseases.
83
CHAPTER 7: 78 kDa GLUCOSE REGULATED PROTEIN
AS A NOVEL ANTICANCER TARGET
7.1 Introduction
The endoplasmic reticulum (ER) is the primary site for the synthesis of secretory
and transmembrane proteins in cells. Nascent unfolded polypeptides entering the ER are
bound by the various ER chaperone proteins, such as the 78 kDa glucose regulated
protein (GRP78), to facilitate proper folding and post-translational modifications
necessary for its maturation into a functional protein.
GRP78 belongs to the HSP70 family of chaperone proteins, which has at least 14
homologous members identified so far (Stricher et al., 2013). HSP70 (HSPA1/2) and
Hsc70 are the main inducible and constitutive forms of HSP70, respectively. Hsc70 is
abundant and ubiquitously expressed in all cells. HSP70 is undetectable in cells under
normal conditions and greatly upregulated under conditions of cellular stress. The other
proteins are variants or tissue-specific isoforms of HSP70 that differ in localization or
expression patterns. Most of the HSP70 members are cytosolic with the exception of ER-
resident GRP78 (HSPA5, BiP) and mitochondrial GRP75 (HSPA9). Being an ER
resident protein, GRP78 has an N-terminal ER signal sequence and a C-terminal KDEL
sequence for ER retention, targeting it to the ER lumen. Unlike HSP70, GRP78 synthesis
is not affected by heat shock. Instead, GRP78 expression is regulated by glucose, nutrient
starvation and misfolded proteins (Kozutsumi et al., 1988; Lee, 1981). Despite these
differences, there is considerable structural and functional similarity between GRP78 and
84
the other HSP70 family proteins. For example, GRP78 shares 66% sequence identity and
82% similarity with Hsc70 (HSPA8). A BLAST search across the entire protein database
from all organisms retrieved several homologues of GRP78 from different species with
>90% sequence identity and similarity (e.g. hamster BiP: 98% sequence identity, Mouse
BiP: 98% sequence identity). This suggests that GRP78 homologues from various species
share a common ancestry.
Table 7.1 HSP70 family of chaperones. BLASTp results of some of the HSP70
proteins against human HSPA5 sequence.
Protein
Accession
code
Description Length
%
Identity
1
%
Similarity
2
Score
E-
value
3
HSPA1 P08107
Inducible isoform
of HSP70
641
393/611
(64%)
496/611
(81%)
809 0.0
HSPA2 P08107
Inducible isoform
of HSP70
641
393/611
(64%)
497/611
(81%)
785 0.0
HSP70t P34931
Testis-specific
isoform
641
398/618
(64%)
498/618
(80%)
812 0.0
Hsc70B P54652
Minor
constitutively
expressed form
639
398/613
(64%)
494/613
(80%)
810 0.0
HSP70B P17066
Putative
pseudogene
protein
643
384/608
(63%)
489/608
(80%)
795 0.0
Hsc70 P11142
Major
constitutively
expressed isoform
646
406/613
(66%)
503/613
(82%)
826 0.0
mtHSP70 P38646
Mitochondrial
isoform
679
323/637
(50%)
436/637
(68%)
592
2e-168
STCH P48723
Microsomal
isoform
471
170/449
(37%)
262/449
(58%)
290 8e-78
HSP90-
Alpha
*
P07900
HSP 86, Renal
carcinoma antigen
NY-REN-38
732
10/30
(33%)
15/30
(50%)
21.2 0.19
HSP90-
beta
*
P08238 HSP 90 724
10/30
(33%)
14/30
(46%)
19.2 0.71
ACTH
*
P68133
Alpha skeletal
muscle actin
377 6/23 (26%)
13/23
(56%)
16.9 1.8
1 Percent identity - measures how many amino acids or bases are identical in an alignment of two
sequences
85
2 Percent similarity - the ratio between the alignment score (using some substitution matrix and gap costs)
and the worse of the two self-alignment scores for the two sequences
3 E-value between 0.1 and 10 is generally represents significant hits (though E value < 0.1 is considered
most significant).
* A comparison of GRP78 with HSP90 and ATPase containing actin proteins
7.1.1 GRP78 Structure
GRP78 contains a highly conserved N-terminal nucleotide binding domain (NBD)
and a C-terminal substrate-binding domain (SBD), characteristic of the HSP70 family.
ATP binding and hydrolysis in the NBD causes conformational changes in the protein
that are transmitted through an inter-domain region to the SBD, influencing the affinity
and kinetics of peptide binding and release. In turn, substrate binding at SBD affects the
rate of ATP hydrolysis. Thus, the rates of ATP hydrolysis and peptide release depends on
the nucleotide occupying the catalytic site on the N-terminal domain (Flynn et al., 1989),
peptide binding on the C-terminal domain (Blond-Elguindi et al., 1993; Flynn et al.,
1989), and also co-chaperones binding (Shen and Hendershot, 2005; Shen et al., 2002).
Mutagenesis studies have shown that disrupting GRP78 ATPase activity inhibits the
release of bound peptides, which results in the accumulation of unfolded or misfolded
proteins in the ER (Gaut and Hendershot, 1993; Hendershot et al., 1995). The crystal
structure of the isolated ATPase domain has been resolved (Fig. 7.1a) and it closely
resembles the HSP70 structure. Structure of full length GRP78 is yet to be elucidated. A
representation of the full length GRP78 based on bacterial DnaK homologue is shown in
Fig. 7.1b. Residues involved in ATPase activity and interdomain communication are
highlighted.
86
Figure 7.1 GRP78 domain structure
(a) Crystal structure of GRP78 ATPase domain (b) Domain architecture of GRP78 based
on full length DnaK.
87
7.2 GRP78 as a regulator of the unfolded protein response
ER is an important site for protein quality control (PQC). Newly synthesized
unfolded polypeptides entering the ER are bound by the various ER chaperone proteins to
facilitate proper folding. Any improperly folded proteins are cleared by an ER-associated
degradation (ERAD) machinery. There is a dynamic homeostatic balance between the
flux of newly synthesized proteins entering the ER, the protein folding capacity of the ER
and the degradation of misfolded proteins by the ERAD. Conditions such as nutrient
depletion, intracellular calcium level fluctuation, or redox alteration cause perturbations
to the PQC system and result in the accumulation of unfolded or misfolded proteins in the
ER. In response to ER stress, an unfolded protein response (UPR) is activated. The UPR
controls acts to either reduce the ER stress and restore the ER homeostasis in case of mild
ER stress (pro-survival response), or in case of severe ER stress, activate cell death
signaling (pro-apoptotic response) (Walter and Ron, 2011). As an ER chaperone that
regulates protein folding, GRP78 is an integral component of the UPR. It binds to
misfolded proteins, prevents their aggregation, and facilitates their refolding. In addition,
GRP78 acts as the molecular switch that transduces ER stress signals to activate
downstream survival pathways.
Under normal growth conditions, GRP78 remains bound to three ER
transmembrane proteins: PKR-related ER kinase (PERK), inositol requiring enzyme
(IRE-1) and activated transcription factor-6 (ATF6). These act as major transducers of
ER stress and remain inactive when bound to GRP78 on the luminal side. During ER
stress, GRP78 dissociates from these transmembrane proteins and thus activates them
88
(Lee, 2001). In turn, each of these proteins further activates various downstream
signaling pathways aimed at decreasing the protein folding load and increasing protein
folding capacity. When the stress is severe and the pro-survival UPR signaling cannot
restore ER homeostasis, the UPR transducers activate CHOP, caspase-7 and JNK, leading
to apoptotic cell death (Lai et al., 2007; Szegezdi et al., 2006). Figure 7.2 provides an
overview of the UPR signaling cascade, highlighting the key players.
Figure 7.2 UPR as an emerging drug target
In the recent years, the UPR has emerged as an attractive target for anticancer
drug development (Hetz et al., 2013; Li et al., 2011). Table 7.2 summarizes some of the
89
recent progress in the development of UPR-targeted drugs. Most of the drug discovery
efforts have been focused towards inhibiting the IRE1 arm of UPR. Aside from UPR
targets, there has been significant progress in developing drugs targeting the ERAD and
proteasome pathway. Bortezomib (or Velcade), a potent proteasome inhibitor has been
approved for clinical use in the treatment of multiple myeloma. However, despite its role
in regulating the UPR and implications in cancer and drug resistance, very few GRP78
inhibitors have been identified till date. This opens up a new avenue for drug design and
development, targeting GRP78. However, as initial drug discovery efforts have revealed,
targeting GRP78 using small-molecules could prove challenging.
90
Table 7.2 Recent advances in inhibitors targeting the UPR pathway
Drug Structure Target Mechanism of action Clinical
development
Reference
GSK2656157
PERK
Inhibits PERK
autophosphorylation
Preclinical
(Pancreatic,
multiple
models)
(Atkins et
al., 2013;
Axten et
al., 2013)
ISRIB
PP1C Inhibits eIF2 α phosphatase
Salubrinal
GADD34
Inhibits eIF2 α
dephosphorylation
(Boyce et
al., 2005)
Guanabenz
GADD34
Binds GADD34 and
inhibits GADD34-PP1C
assembly
(Tsaytler
et al.,
2011)
N
N
N
N
O
N
NH
2
Cl
O
N
H
O
H
N
O
O
Cl
NH
O
N
H
Cl
3
C NH
S
N
Cl
Cl
N
N NH
2
NH
2
91
Sunitinib
IRE1
ATP competitive kinase
inhibitor, stabilizes active
confirmation. Inhibits
IRE1a activity but
activates RNase domain
Phase II
(Multiple
myeloma),
Approved
(Renal cell
carcinoma)
(Wang et
al.,
2012),(Ali
et al.,
2011)
APY24
IRE1
ATP competitive kinase
inhibitor, stabilizes active
confirmation. Inhibits
IRE1a activity but
activates RNase domain
(Wang et
al., 2012)
Compound 3
IRE1
Kinase inhibitor, stabilizes
inactive conformation
(Wang et
al., 2012)
Irestatin 9389
IRE1
Preclinical
(multiple
myeloma)
(Feldman
and
Koong,
2007)
N
H
F
O
N
H
NH
O
N
N
N
H
N
H
N
HN
N
N
H
N
N
N
N
HN
O
HN
CF
3
NH
2
N
CF
3
N
S
H
N
O
S
92
STF-083010
IRE1 IRE1 α RNase domain
Preclinical
(multiple
myeloma)
(Papandre
ou et al.,
2011)
4µ8C
IRE1 IRE1 α RNase domain Preclinical
(Cross et
al., 2012)
3-Methoxy-6-
bromosalicyla
ldehyde
IRE1 IRE1 α RNase domain
(Papandre
ou et al.,
2011)
Toyocamycin
IRE1 IRE1 α RNase domain
Preclinical
(multiple
myeloma)
(Ri et al.,
2012)
MKC-3946
IRE1
IRE1 α RNase domain
Preclinical
(multiple
myeloma)
(Mimura
et al.,
2012)
OH
N
S
O
O
S
O
O
O
CH
3
HO
CHO
HO Br
O
N N
N
O
HO
HO
OH
N
NH
2
HO
O
S
N
O
N
93
Table adapted from reviews in (Hetz et al., 2013; Li et al., 2011).
AEBSF
ATF6
Inhibits nuclear
translocation of cleaved
ATF6
(Okada et
al., 2003)
H
2
N
S
O
O
F
94
7.3 GRP78 as a novel anticancer target
7.3.1 GRP78 in cancer proliferation and progression
The microenvironment of solid tumors is characterized by hypoxia, low pH and
glucose depletion. To survive such unfavorable stressful conditions, cancer cells
induce glucose-regulated pathways such as those mediated by GRP78. Indeed, GRP78 is
found to be induced in a wide range of cancers such as breast, gastric, prostate and
hepatocellular carcinomas (Lee, 2007). Several in vivo and retrospective biopsy studies
have implicated GRP78 in tumor proliferation and survival (Dong et al., 2008; Li and
Lee, 2006). GRP78 was found to be required for tumor progression in a mouse
fibrosarcoma model (Jamora et al., 1996). GRP78 plays an important role in
tumorigenesis and proliferation in PTEN-null induced prostate cancer. Reducing the
GRP78 levels mitigated the tumor caused by PTEN loss and also inhibited Akt activation
(Fu et al., 2008). These studies support an important role for GRP78 in the early stages of
tumorigenesis, conferring anti-apoptotic effects and an aggressive cancer phenotype.
Interestingly, GRP78 has been found to exert a protective role against PTEN-null induced
liver cancer (Chen et al., 2013). In contrast to the results seen in PTEN-null prostate
cancer, liver-specific knockdown of GRP78 aggravated the progression of PTEN-null
induced liver cancer. Hence, there may be a possibility of tissue-specific and tumor-
specific function of GRP78 in either promoting or preventing cancer progression. Table
7.3 summarizes some of these studies using knockout mice models. Similarly, silencing
GRP78 using siRNA has also been shown to inhibit cancer cell proliferation and promote
cell death.
95
Table 7.3. GRP78 in tumor progression – Summary of knockout studies
Cell/Cancer type Model Key findings Reference
Embryonic KO • Embryonic peri-implantation lethality
• Increased apoptosis of inner cell mass
(Luo et al., 2006)
Embryonic Heterozygous GRP78
• Viable
• Increase in GRP94, PDI but not CHOP and XBP1
• Normal organ development
• Normal antibody production
(Luo et al., 2006)
Prostate cancer GRP78 and PTEN CKO in
prostate epithelium
• GRP78
-/-
suppresses PTEN-null prostate tumorigenesis
• GRP78
-/-
inhibits AKT activation in PTEN-null prostate
(Fu et al., 2008)
Mammary tumor Heterozygous GRP78 in
MMTV-endogenous mammary
tumor
• Inhibition of tumor growth and increase in apoptosis
• Decrease in microvessel density in tumor
(Dong et al.,
2008)
Endothelial cell CKO in host endothelial cell Decrease in tumor angiogenesis and metastasis (Dong et al.,
2011)
Whole body/Tumor
microenvironment
Heterozygous GRP78 (host) Inhibition of neovascularization during early tumor
formation
(Dong et al.,
2011)
Purkinjee cell CKO in Purkinjee cells • Purkinjee cell degeneration, motor coordination defect,
cerebellar atrophy
• Increase in UPR signaling and apoptosis.
• Significant reduction in cytosolic ubiquitinated proteins
• ER expansion
(Wang et al.,
2010)
Liver cancer GRP78 and PTEN CKO in
liver
• PTEN-null mediated liver steaosis and hepatocellular and
cholangial carcinoma accelerated by GRP78
-/-
• JNK activation, β-catenin inhibition and PDGFR α up-
regulation
(Chen et al.,
2013)
Hematopoietic system CKO in hematopoietic system • Decrease in hematopoietic stem cell, lymphoid cells and
myeloid progenitor cell populations
• Increased apoptosis
• Constitutive activation of UPR signaling in bone marrow
(Wey et al.,
2012a)
Leukemogenesis GRP78 and PTEN CKO in
hematopoietic system
• Restores hematopoietic stem cells
• Suppresses leukaemic cell expansion
• Akt activation is inhibited
(Wey et al.,
2012b)
96
In addition to its pro-survival role in promoting cancer cell survival, GRP78 also
exerts a direct anti-apoptotic role in preventing cell death. GRP78 forms an inhibitory
complex with BIK, preventing its activation during estrogen-starvation induced
apoptosis. Further studies have revealed that GRP78 also forms a complex with BCL2
and upon overexpression, prevents BCL2 sequestration by BIK. Suppression of GRP78
levels using siRNA restored the sensitivity of breast cancer cells to estrogen- starvation
treatment (Fu et al., 2007; Zhou et al., 2011). Similarly, GRP78 physically interacts with
pro-caspase-7 through its ATPase domain and forms an inhibitory complex preventing its
activation and release during apoptosis (Reddy et al., 2003). Mutations in the ATPase
domain disrupt the formation of inhibitory complexes and activate these pro-apoptotic
proteins.
GRP78 has also been found on their cell surface of cancer cells (Mintz et al.,
2003; Misra et al., 2005). Its emerging role as a multifunctional receptor, mediating
various signaling pathways from cell surface to nucleus, opens new opportunities
regarding its function and the effects of its inhibition on cancer cell signaling (Gonzalez-
Gronow et al., 2009; Misra et al., 2009; Misra and Pizzo, 2010; Ni et al., 2011).
7.3.2 GRP78 in drug resistance
In addition to conferring a survival advantage on cancer cells, GRP78
overexpression also results in drug resistance and refractory tumors by activating UPR
and inhibiting drug-induced apoptotic-cell death. Several studies have shown that GRP78
overexpression protects cancer cells from the cytotoxic effects of chemotherapy and
97
downregulating or inhibiting its activity restores drug sensitivity (Li and Lee, 2006;
Pyrko et al., 2007; Virrey et al., 2008). Table 7.4 provides a brief summary of the effects
of GRP78 expression on the sensitivity of cancer cells to different therapeutic agents
from representative studies.
Table 7.4 GRP78 and chemoresistance in cancer
Cancer type Therapeutic
agent
Key Findings References
NSCLC Doxorubicin,
Taxol
GRR78 overexpression in doxorubinc and taxol
resistant non-small cell lung cancer cells
(Koomagi et
al., 1999)
Bladder cancer Etoposide,
Doxorubicin,
Camptothecin
GRP78 overexpression confers resistance to cell
death induced by topoisomerase I and II inhibitors
(Reddy et al.,
2003)
Breast cancer Etoposide siGRP78 sensitizes breast cancer cells to
etoposide-induced cell death
(Dong et al.,
2005)
Estrogen
starvation
GRP78 overexpression protects breast cancer cells
from estrogen starvation induced cell death
(Fu et al.,
2007)
Doxorubicin Retrospective study shows GRP78 overexpression
predictive of doxorubicin resistance in breast
cancer patient tumors
(Lee et al.,
2006)
Squamous
carcinoma
Doxorubicin siGRP78 sensitizes doxorubicin-resistant dormant
squamous carcinoma cells
(Ranganathan
et al., 2006)
Glioma Temozolomide Overexpression of GRP78 leads to temozolomide
resistance while silencing GRP78 sensitizes
(Pyrko et al.,
2007)
Endometrial
cancer
Cisplatin GRP78 induced in response to cisplatin treatment.
GRP78 kncokdown enhances sensitivity to
cisplatin
(Gray et al.,
2013)
Paclitaxel and
cisplatin
Estrogen induces GRP78 in endometrial cancer
and correlates with resitance to paclitaxel and
cisplatin.
(Luvsandagva
et al., 2012)
Hypopharynge
al carcinoma
Cisplatin GRP78 mediates severe hypoxia-induced cisplatin
resistance
(Pi et al.,
2014)
Tumor brain
endothelial
cells
CPT-11,
Temozolomide
, Etoposide
GRP78 overexpression in drug-resistant TuBEC
cells. GRP78 silencing overcomes drug resistance
(Virrey et al.,
2008)
98
7.3.3 GRP78 in breast cancer
Several lines of evidence highlight an important role of GRP78 in breast cancer
and drug resistance.
i) Elevated levels of basal and stress-induced GRP78 in breast cancer cell lines. A 1.5- to
3-fold increase in basal as well as stress-induced expression levels of GRP78 is observed
in several breast cancer cell lines (Gazit et al., 1999). In our study, we too analyzed
GRP78 expression in a panel of breast cancer cell lines. Our results confirm the previous
findings reported by Gazit et al. Steady state GRP78 expression was higher in some cell
lines such as MCF7 and MDA-MB-231, while comparatively lower in others such as
SKBR3 and MDA-MB-468. MCF7 was not very responsive to ER stress induced by
tunicamycin (Fig. 7.3). Similarly, we also observed that MDA-MB-468, SKBR3 and Hs-
578-T cells are not able to induce GRP78 and CHOP expression in response to ER stress
induced by tunicamycin.
Figure 7.3 GRP78 expression in a panel of breast cancer cell lines.
T47D
MCF7
MDA-MB-435
MDA-MB-231
MDA-MB-468
BT-549
SKBR3
Hs-578-T T47D
MCF7
MDA-MB-435
MDA-MB-231
MDA-MB-468
BT-549
SKBR3
Hs-578-T
GRP78
CHOP
Actin
Actin
GRP78
Stress induced (TU 1.5 µg/mL)
Basal
99
Top panel: Basal GRP78 expression in breast cancer cells grown in normal growth
conditions. Bottom panel: Breast cancer cells were grown in the presence of tunicamycin,
an ER stress inducer, for 24 h. Cell lysates were analyzed by western blotting for GRP78
and CHOP expression.
ii) GRP78 knockdown with siRNA reduces viability of breast cancer cells. We performed
siRNA studies in MCF7, T47D and MDA-MB-231 cell lines using GRP78-specific
siRNAs. Cell viability in all three cell lines was reduced in response to GRP78
knockdown for 72 and 96 h (Fig. 7.4). This confirms that GRP78 is important for breast
cancer cell survival.
Figure 7.4 GRP78 knockdown reduces breast cancer cell viability
MCF7, T47D and MDA-MB-231 cells (7500 cells/well) were seeded in a 96-well plate.
GRP78 siRNA (siGRP78-A, siGRP78-B, siGRP78-C) or negative control siRNA at 5 nM
was added to the cells and incubated at indicated times. Cell viability was measured at
48, 72 and 96 h post transfection using MTT assay.
iii) GRP78 induction in breast cancer cell lines confers resistance to chemotherapeutic
agents. Endogenous GRP78 expression levels are elevated in drug-resistant breast cancer
100
cell lines. Suppression of GRP78 levels using siRNA re-sensitizes these resistant cancer
cells (Dong et al., 2005).
iv) GRP78 confers growth advantage to tumor cells selectively. Progression of
endogenous mammary tumor in heterozygous mice with half the amount of GRP78 is
impeded with significant reduction in tumor microvessel density and increased apoptosis
(Dong et al., 2008).
v) Elevated levels of GRP78 expression in malignant breast tumors. GRP78 mRNA and
protein levels are elevated in malignant primary breast tumors when compared with
normal breast tissue. A significant overexpression is also been observed in higher grade,
estrogen receptor negative (aggressive) and pre-metastatic tumors (Fernandez et al.,
2000).
vi) GRP78 expression level is predictive of drug response in early breast cancer patients.
Retrospective analyses of tumor biopsies show that breast cancer patients, whose tumors
showed high levels of GRP78 expression before the initiation of doxorubicin
chemotherapy, have a shorter time to recurrence and developed resistance to doxorubicin
(Lee et al., 2006).
vii) GRP78 exerts anti-apoptotic role during estrogen-starvation induced apoptosis.
GRP78 forms inhibitory complexes with BIK, preventing its activation and further
complexes with BCL2, preventing its sequestration by BIK, thus preventing BIK-
mediated apoptosis in response to estrogen starvation of breast cancer cells (Fu et al.,
2007; Zhou et al., 2011).
101
viii) BRCA1 negatively regulated GRP78-mediated breast cancer cell survival. BRCA1
overexpression suppressed GRP78 expression, whereas BRCA1 mutation or silencing
increased GRP78 expression, cancer survival and resistance to apoptosis (Yeung et al.,
2008)
7.4 GRP78 inhibition
One of the common concerns over targeting chaperone proteins has been their
ubiquitous role in protein folding in both normal as well cancer cells. Successful clinical
development of HSP90 inhibitors has validated the feasibility of ‘drugging’ chaperone
proteins (Guzman et al., 2013; Workman et al., 2007). Similarly, in case of GRP78,
cancer cells have an elevated GRP78 expression compared to normal cells to survive the
stress of rapid proliferation and sustain an overloaded protein folding machinery.
Although homozygous disruption of GRP78 in mice results in embryonic lethality due to
proliferative defect and apoptosis of inner cell mass (Luo et al., 2006), heterozygous adult
mice with half the amount of GRP78 showed no abnormal organ development or
antibody production (Dong et al., 2008). Moreover, reduced GRP78 levels selectively
inhibited cell proliferation and induced apoptosis of the cancer cells, without any
abnormal effects on the normal cells. Therefore, given the unfavorable conditions in the
tumor microenvironment, cancer cells will be more dependent on GRP78-mediated
survival signals than normal cells. Selective cytotoxicity can be achieved by exploiting
this “chaperone addiction” of cancer cells.
102
Two main approaches to targeting GRP78 have been explored in literature. The
first approach is to decrease stress induction of GRP78 by targeting at the transcriptional
level. The second approach is to inhibit the chaperone functions of GRP78 by targeting
its ATPase or chaperone activity. In the following chapters, we have focused on this
second approach to develop novel GRP78 inhibitors.
7.4.1 GRP78 inhibitors in literature
Though several lines of evidence have confirmed the importance of GRP78 in
cancer and chemoresistance, few selective GRP78 inhibitors are currently available.
Table 7.5 summarizes the compounds reported in literature till date that target GRP78 at
transcriptional level or protein level. Among these, versipelostatin selectively
downregulates GRP78 gene expression induced by glucose starvation and inhibits tumor
growth (Matsuo et al., 2009; Park et al., 2004). However, being a macrocyclic compound,
its isolation and synthesis is complicated and its physicochemical properties are not
favorable. The green tea ingredient, epigallocatechin gallate (EGCG) has been shown to
inhibit GRP78 ATPase activity. It has been shown to induce apoptosis and sensitize
breast cancer cells to etoposide and paclitaxel (Ermakova et al., 2006; Luo et al., 2010).
However, it exhibits pleiotropic effects and inhibits a variety of other cellular target. This
lack of selectivity and potency is further compounded by the presence of reactive and
labile functional groups, resulting in its poor bioavailability. Several other ATP mimetics
have been reported as GRP78 inhibitors, although their cellular activity has not been well
characterized. Targeting the ATPase activity of GRP78 has been challenging. GRP78 has
103
a low Kd for ATP binding (0. 7 µM). On the other hand, cellular ATP levels are in the
millimolar range. Moreover, the ATP binding pocket is buried deep in the protein and is
polar in nature, making pocket accessibility by small-molecule compounds difficult.
Therefore, identifying ATP-competitive GRP78 inhibitors are challenging (Massey,
2010). Recently, HSP70 inhibitors acting via allosteric mechanism have been reported.
These compounds target sites on HSP70 that are outside the ATP binding pocket, that
regulate interdomain communication or that are important for co-chaperone binding
(Kang et al., 2014; Li et al., 2013; Miyata et al., 2013; Rodina et al., 2013; Rousaki et al.,
2011; Taldone et al., 2014). Therefore, new avenues for inhibiting HSP70 as well as
GRP78 are emerging.
104
Table 7.5 Compounds targeting GRP78
Drug Structure Mechanism of action Activity data Clinical
development
Reference
Versipelostatin
Inhibits ER-stress induction of
GRP78
Preclinical (Matsuo et
al., 2009)
EGCG
Binds GRP78 ATPase domain
and inhibits ATPase activity
Preclinical
(Breast)
(Luo et al.,
2010)
Honokiol
Binds GRP78 ATPase domain (Martin et
al., 2013)
Aspirin
Binds to polypeptide binding
site, inhibits ATPase activity
ATPase activity
– 32%
Inhibition at 500
µM
(Deng et al.,
2001)
O
O
OH
OH
OH
OH
HO
OH
OH
OH
HO
HO
O OH
O
O
105
Salicylic acid
Binds to polypeptide binding
site, inhibits ATPase activity
Kd = 54.6 µM
35% at 500 µM
(Deng et al.,
2001)
EGF-SubA
Cleaves GRP78 in the inter-
domain hinge region (at L416)
Preclinical
(Breast,
Prostate)
(Backer et
al., 2009)
VER-155008
Binds GRP78 and HSP70
ATPase domain, inhibits ATP
binding and ATPase activity
K
D
= 0.08 µM
(GRP78)
K
D
= 0.3 µM
(HSP70)
(Macias et
al., 2011)
JMC-7
Binds GRP78 and HSP70
ATPase domain, inhibits ATP
binding competitively
K
D
= 12.4 µM
(GRP78)
K
D
= 1.9 µM
(HSP70)
(Macias et
al., 2011)
JMC-10
Binds GRP78 and HSP70
ATPase domain, inhibits ATP
binding competitively
K
D
= 2.4 µM
(GRP78)
K
D
= 4.3 µM
(HSP70)
(Macias et
al., 2011)
O OH
OH
N
N
N
N
NH
Cl
Cl
O
NH2
HO
HO
O
N
N
N
N
N
NH
O
NH2
HO
HO
O
O
NH
2
N
N
N
N
NH N
O
NH2
HO
HO
OH
106
JMC-14
Binds GRP78 and HSP70
ATPase domain, inhibits ATP
binding competitively
K
D
= 0.06 µM
(GRP78)
K
D
= 0.05 µM
(HSP70)
(Macias et
al., 2011)
N
N
N
N
NH
O
NH2
HO
HO
O
N
N
107
CHAPTER 8: IDENTIFICATION OF NOVEL THIENO[2,3-
d]PYRIMIDIN-4(1H)-ONE-BASED GRP78 INHIBITORS BY
PHARMACOPHORE-BASED VIRTUAL SCREENING
8.1 Introduction
GRP78 is a member of the highly conserved Hsp70 family of chaperone proteins.
It is abundantly expressed in the endoplasmic reticulum (ER), where it facilitates proper
folding of nascent polypeptides. As an ER chaperone that regulates protein folding,
GRP78 is an integral component of the unfolded protein response and promotes cell
survival and ER homeostasis during ER stress (Wang et al., 2009). GRP78 has been
implicated in cancer progression and drug resistance and is a promising therapeutic target
for anti-cancer drug development (Lee, 2007, 2014; Wang et al., 2009).
While the N-terminal ATP binding domain (NBD) appears to be a promising site
for drug action, designing inhibitors targeting NBD has proved challenging (Massey,
2010). Few inhibitors targeting GRP78 NBD have been reported till date. Natural
product-derived epigallocatechin 3-gallate (EGCG) has been reported to bind to the NBD
of GRP78 and inhibit its ATPase activity (Ermakova et al., 2006). Adenosine-based
analogues such as VER-155008 bind at the ATP binding pocket in Hsp70 and GRP78
and inhibit their ATPase and chaperone activities (Macias et al., 2011; Massey et al.,
2010; Williamson et al., 2009). More recently, compounds targeting allosteric sites
108
adjacent to the ATP binding pocket in other Hsp70 isoforms have been reported (Li et al.,
2013; Rodina et al., 2013). Given the importance of GRP78 in cancer progression and
drug resistance and the scarcity of GRP78 inhibitors, we sought to identify novel non-
adenosine scaffolds that inhibit GRP78 ATPase activity using a pharmacophore-based
virtual screening approach. Here, we report the identification of a novel class of
thienopyrimidinones that inhibit GRP78 ATPase activity, aggravate ER stress and inhibit
cancer cell survival.
8.2 Results
8.2.1 Pharmacophore modeling
In order to identify novel scaffolds of GRP78 inhibitors, we generated a 3D-
pharmacophore model based on the structures of known GRP78 inhibitors. A
pharmacophore is a two- or three-dimensional spatial arrangement of chemical features
that are necessary for a ligand to bind into a pocket in a target biomolecule and to exhibit
pharmacological activity either by displacing endogenous substrate or by inducing
conformational changes in the active site. Ligand-based molecular modeling using
pharmacophore is a useful approach to explore new chemical space as well to identify
novel inhibitor scaffold that are completely different from the compounds, which are
used to generate the pharmacophore model. Several utilities of 3D pharmacophore-based
screenings have been successfully demonstrated in drug-discovery research to identify
modulators for different targets (Poyraz et al., 2013; Ritschel et al., 2014; Sabatini et al.,
2013; Verheij et al., 2012).
109
In this study, we used a qualitative HipHop method, as there are only a few
reported GRP78 inhibitors with narrow activity range. For the generation of a common
feature pharmacophore, it is important to select compounds that exhibit similar activity
profiles and similar binding modes. In this study, we generated a common feature
pharmacophore based on reported GRP78 inhibitors (Macias et al., 2011), JMC-10, JMC-
13 (VER-155008) and JMC-14 using Catalyst 4.11 from Accelrys, Inc. Highest
weighting was assigned to these three compounds by considering principal as 2 and
maximum omitting features as 0, which ensures that all the features in those compounds
will be considered in generating hypotheses and all features must mapped with the
generated hypotheses. The pharmacophore, which is complementary to the protein-ligand
(GRP78-JMC-10) interactions in the PDB: 3LDP (Macias et al., 2011), is comprised of
five features, three hydrogen bond acceptors, one each of hydrogen bond donor and
hydrophobic feature. The pharmacophore, along with the spatial distances among
different features, is depicted in Fig. 8.1a. The pharmacophore is completely mapped
onto all three GRP78 inhibitors with a maximum fitness value close to 5 (Fig. 8.1b-d).
With the unavailability of many GRP78 inhibitors, it was not possible to validate
the pharmacophore statistically. However, to see how good the pharmacophore is, we
tried to fit the pharmacophore on other reported GRP78 inhibitors. The pharmacophore
partially mapped onto the moderately active inhibitor JMC-7 (fit value of 3.97), while it
is mapped well onto EGCG (fit value of 4.55) (Fig. 8.1e-f).
110
Figure 8.1 Five-feature Pharmacophore model and mapping of GRP78 inhibitors.
a) The five-feature pharmacophore generated using JMC-10, VER-155008 (JMC-13) and
JMC-14. Color code: Green - hydrogen bond acceptors (HBA1-3), Purple – hydrogen
111
bond donor (HBD), cyan – hydrophobic features (HY). Distances among different
features are in Å. b-d) Pharmacophore mapping onto the three training set compounds,
JMC-10 (b), JMC-13 (c) and JMC-14 (d). e-f) Pharmacophore mapping onto other
GRP78 inhibitors, JMC-7 (e) and EGCG (f). Top panel: 3D mapping of the compounds
and Bottom pane: 2D representation of the pharmacophore.
8.2.2 Hit identification
Next, our in-house database of 50,000 drug-like diverse compounds
(Serrao et al.,
2013b), comprising representing a chemical space of 8 million commercially available
small molecules, was virtually screened using the pharmacophore. Pharmacophore fit
value was computed using best fit method. 50 compounds from the top 200 hits from the
virtual screening were selected for further screening in an in vitro GRP78 ATPase assay.
From this initial screening, we identified two such compounds with different scaffolds. A
general schema of our drug design strategy is outlined in Fig. 8.2a. In this paper, we have
explored one of these two hits with a thieno[2,3-d]pyrimidin-4(1H)-one scaffold. Figure
8.2b shows the pharmacophore mapping onto one of these two active hits, 122. Four out
of the five pharmacophore features were mapped onto 122 (5-(2-chlorophenyl)-2-(((2-
morpholino-1-(p-tolyl)ethyl)amino)methyl)thieno[2,3-d]pyrimidin-4(1H)-one) with a fit
value of 3.25. Two out of the three hydrogen bond acceptors mapped the
thienopyrimidinone carbonyl and morpholino oxygen, while the third hydrogen bond
acceptor missed the mapping. The hydrogen bond donor feature mapped onto the NH of
the thienopyrimidinone moiety, while the hydrophobic feature mapped onto one of the
phenyl groups of the compound. 122 showed a moderate inhibition of GRP78 steady state
ATPase activity with an average IC
50
value of 43 µM in the presence of ATP at K
m
.
GRP78 inhibition leads to accumulation of misfolded proteins in the ER and activation of
112
the unfolded protein response. Using CHOP as a marker for GRP78 inhibition and UPR
activation, we next assessed the effect of 122 on ER stress. In T47D cells, treatment with
122 caused mild ER stress with slight increase in CHOP and no increase in GRP78. On
the other hand, in the presence of tunicamycin, an ER stress inducer, CHOP induction
was significantly increased indicating an exacerbation of ER stress. Interestingly, GRP78
induction under ER stress was inhibited with 122 treatment (Fig 8.2b).
Figure 8.2 Hit identification and optimization.
a) Schematic shows overall drug discovery strategy. b-c) Pharmacophore mapping of
scaffold A. Left panel is the 3D mapping of the initial hit 122 (b) and the optimized lead
240 (c), while right panel is the 2D representation of the pharmacophore mapping of the
inhibitors.
113
8.2.3 Hit optimization through substructure-similarity search
To explore structure activity relationship as well as to optimize the compound
122, we performed a substructure and similarity search using commercially available
databases (www.enamine.net, www.asinex.com). In the substructure search, we tried to
keep basic scaffold intact, while in similarity search, we tried to find compounds most
similar in terms of chemical features and functional groups. We identified 110 such
analogs from the substructure and similarity search and tested them further using the in
vitro ATPase activity assay. The pharmacophore mapped onto best lead compound, 240
(2-(1-((2-morpholino-1-(p-tolyl)ethyl)amino)ethyl)-5-(thiophen-2-yl)thieno[2,3-
d]pyrimidin-4(1H)-one) is shown in Fig. 8.2c, with four of the five features are mapped
with fit value of 3.33.
8.2.4 Structure activity relationship
Clustering of the analogues from the substructure and similarity search further
helped to elucidate a limited structure activity relationship. The structures and GRP78
ATPase inhibitory activities of selected analogues are shown in Tables 8.1-8.3. Starting
with the initial hit 122, modifications at the 6-position in the thienopyrimidinone core
with small groups such as methyl were well tolerated (224, 225). Introduction of more
bulkier groups such as a phenyl ring proved unfavorable (228, 238). Next, upon replacing
the 2-chlorophenyl ring at 5-position with a 4-fluorophenyl or 4-methylphenyl yielded
more potent compounds (232, 229, 224, 225). On the other hand, replacement of the
phenyl ring with a 5-methyl-thiophenyl ring abolished activity (235). Interestingly,
114
introduction of a hydrophobic methyl group in the phenyl ring (Position R
4
) improved
activity in general. This is in agreement with the pharmacophore where the phenyl ring
was mapped onto the hydrophobic feature. Although, compounds 122 and 226 do not fit
this generalization. Finally, addition of a methyl group at R
3
position, hindering the
rotation of the substituents at 2-position in the thienopyrimidinone core (240), resulted in
potent inhibition of GRP78 steady state ATPase activity with an IC
50
value of 2 µM.
Based on the pharmacophore mapping, the phenyl ring maps the hydrophobic
feature while the morpholino oxygen is important as a hydrogen bond acceptor.
Modifications in the placement of the morpholino ring are shown in Table 8.2. Shifting
the morpholino moiety at meta (221, 261) and para-positions (267) in the phenyl ring was
tolerated. Replacement of the morpholino ring with a pyrrolidino ring resulted in loss of
potency (263) while replacement with a piperidino (268) abolished activity. Morpholino
ring was most preferred for activity. Further, substitutions on the benzyl methylamino
linker backbone cause steric hindrance and reduce rotational flexibility, resulting in a loss
of potency of GRP78 inhibition (Table 8.3). A comparison of the structure and activity
underscores the limited SAR with small changes resulting in drastic differences in
inhibitory activity.
115
Table 8.1. Inhibitory activity of compounds
a
% Inhibition values are average of duplicate wells, tested at 100 µM in the presence of 1 µM ATP.
b
% Inhibition values are average of duplicate wells, determined by single concentration screening at 100 µM in the presence of 2, 5, 50 µM ATP.
c
IC
50
values are mean ±
SD calculated from at least 3 independent dose response experiments.
d
Induction of CHOP in T47D cells treated with tunicamycin (1.5 µg/mL) and compounds (10 µM) assessed by western blotting.
Compd R
1
R
2
R
3
R
4
% I IC
50
(µM)
CHOP
induction
d
ADP-glo
a
ADP hunter
b
ADP hunter
c
ATP 1 µM 2 µM 5 µM 50 µM 2 µM
122 H 2-Cl-Ph H Me 76.6 83.2 43.6 8.2 43 ± 22
Y
226 H 2-Cl-Ph H H 86.9 76.4 43.4 0.5 7 ± 8 Y
232 H 4-F-Ph H H 81.5 80.0 18.4 0.9 5 ± 3 Y
229 H 4-F-Ph H Me 80.2 79.2 63.0 2.5 3 ± 1 Y
224 Me 4-Me-Ph H H 85.3 71.9 20.3 -0.6 4 ± 2 Y
225 Me 4-Me-Ph H Me 84.6 67.5 27.9 -2.2 - Y
235 H 5-Me-Thiophen-2-yl H H 26.3 10.2 2.9 -
N
236 H 5-Me-Thiophen-2-yl H Me 67.9 76.5 17.6 0.1 - Y
238 Ph H H H 23.2 11.6 2.3 -
228 Ph H H Me 47.8 58.2 17.5 5.2 21
240 H Thiophen-2-yl Me Me 74.6 77.3 33.4 7.7 2 ± 1
Y
VER-155008 82.7 98.2 74.3 9 ± 2 Y
HN
N
S
O
NH
N
O
R
1
R
2
R
3
R
4
116
Table 8.2. Inhibitory activity of compounds
Compd R
1
R
2
R
3
% I IC
50
(µM)
ADP
Hunter
a
ADP
Hunter
ATP 2 µM 2 µM
221 H Ph 2-morpholino 60.1 4 ± 6
261 H 4-F-Ph 2-morpholino 85.3 11
263 H 4-F-Ph 2-pyrrolidino 58.7 44
262 Ph H 2-morpholino 63.0 10
268 Ph H 2-piperidino 32.7 -
267 Ph H 4-morpholino 88 22
a
% Inhibition values are average of duplicate wells, tested at 100 µM in the presence of 2 µM ATP.
Table 8.3. Inhibitory activity of compounds
Compd R
1
R
2
R
3
R
4
R
5
R
6
% I IC
50
(µM)
ADP
Hunter
a
ADP Hunter
ATP 2 µM 2 µM
279 Me Me Me H Me 2-Me 71.8 10.5
274 Me 4-Me-Ph H H Me 3-F, 4-F 49.9
257 Me 2-Me-Butyl H Me H 3-Cl, 4-Cl 0.6
269 COOMe Me H Me H 3-Cl 37.8
271 COOMe Me H Me H 4-F 38.4
265 COOMe Me H Me Me H 50.1
258 COOEt Me Me Me H 2-F, 6-Cl 31.9
a
% Inhibition values are average of duplicate wells, tested at 100 µM in the presence of 2 µM ATP.
HN
N
S
O
N
R
1
R
2
R
3
HN
N
S
O
N R
5
R
1
R
2
R
4
R
6
R
3
117
8.2.5 240 binding to GRP78
To determine effect of 240 inhibited ATP binding to GRP78, we used a
fluorescence polarization assay using FAM-ATP. While the ATP-competitive inhibitor
VER-155008 inhibited ATP binding, 240 had no effect on ATP binding. It has been
previously reported that mutation of residues important for ATP hydrolysis does not
affect ATP binding (Gaut and Hendershot, 1993). Recently, allosteric HSP70 inhibitors
that bind outside the ATP binding pocket and inhibit ATP-induced conformational
change have been reported (Rousaki et al., 2011). Additional studies to elucidate
compound binding to GRP78 are to understand how it affects the ATPase activity are
ongoing.
Since 240 did not affect ATP binding to GRP78, we tested the specificity of
binding to GRP78 using a drug affinity responsive target stability (DARTS) assay.
Ligand/inhibitor binding to its target protein induces a conformational change in it and
protects against proteolytic degradation. This protection against proteolysis has been
widely used to demonstrate ligand binding (Eskew et al., 2011; Lomenick et al., 2009;
Lomenick et al., 2011; Xu et al., 2013). Recombinant GRP78 (full length) incubated with
240 showed a dose-dependent protection against protease digestion (Fig. 8.3). Similar
protection was also seen with VER-155008.
GRP78
1000 100 10 1 0.1 0.01 100
Pronase - + + + + + + + +
240
VER-155008
118
Figure 8.3 240 binds to GRP78.
DARTS assay was used to demonstrate the binding of 240 to GRP78. Recombinant
GRP78 (1 µg) was incubated with indicated concentrations of 240 for 1 h on ice,
followed by digestion with pronase (0.1 µg) for 10 min. The proteolysis was quenched by
the addition of protease inhibitor cocktail. The reaction was heated to 90 °C for 5 min in
the presence of SDS loading buffer and analyzed by western blotting for GRP78.
8.2.6 240 inhibits GRP78-mediated chaperone refolding
We performed a luciferase refolding experiment in which recombinant luciferase
is chemically denatured using Guanidium HCl and renatured by incubation with full
length GRP78. 240 inhibited GRP78-mediated luciferase refolding at 50 and 100 µM
Similar results were observed with VER-155008 (Fig. 8.3).
Figure 8.3 Inhibition of GRP78 chaperone function by 240.
Chemically denatured luciferase was incubated with full length GRP78 in the presence of
indicated concentrations of 240 or VER-155008 and ATP (100 µM). Luminescence was
measured using the luciferase substrate.
8.2.7 240 induces ER stress
We next investigated if inhibition of GRP78 ATPase activity by these compounds
leads to ER stress induction and activation of the unfolded protein response. GRP78 acts
240 VER-155008
0
20
40
60
80
100
% Inhibition of luciferase renaturation
10 µM
50 µM
100 µM
119
as the molecular switch that transduces ER stress signals to activate the unfolded protein
response (UPR). Mild ER stress results in the activation of a pro-survival UPR signaling
to restore homeostasis. However, when the stress is severe, UPR transducers activate
CHOP, which then activates apoptotic signaling resulting in cell death (Lai et al., 2007).
Inhibition of GRP78 ATPase activity results in accumulation of misfolded proteins,
which induces ER stress (Hendershot et al., 1995).
In T47D cells, treatment with 240 resulted in a dose-dependent increase in GRP78
and CHOP expression, indicating ER stress induction (Fig. 8.4a). Moreover, in the
presence of ongoing ER stress induced by tunicamycin, 240 treatment resulted in
significant increase in CHOP (Fig. 8.4 b,c). Similar results were also obtained with VER-
155008, while EGCG was not effective in inducing ER stress even at 50 µM. We also
observed that T47D cell line were more sensitive than MCF7 cells to the ER stress
induced by 240 treatment or tunicamycin as seen by GRP78 and CHOP induction (Fig.
7.3, Fig. 8.4). 240 did not show inhibition of PDI, another ER chaperone that is involved
in protein folding and ER stress.
This aggravation to severe ER stress was not restricted to the ER stress induced
by tunicamycin (N-glycosylation inhibitor). In the presence of other ER stress inducers
such as thapsigargin (SERCA pump inhibitor) and 2-deoxyglucose (hexokinase inhibitor,
hypoglycemia), 240 treatment caused a similar increase in CHOP induction (Fig. 8.4d).
Therefore, with the inhibition of GRP78 by 240, UPR fails to restore ER homeostasis.
Instead, severe and sustained ER stress is induced with CHOP. Interestingly, we observed
an inhibition of ER stress-induced GRP78 expression in presence of tunicamycin in both
120
T47D and MCF7 cell lines, but not with the other two ER stress inducers. The underlying
mechanism behind this selective inhibition is however not clear. We also observed that
compounds that were inactive in GRP78 ATPase assay screening did not show any
CHOP induction (data not shown). Therefore, CHOP induction serves as a good indicator
to confirm GRP78 inhibition.
Figure 8.4 240 treatment induces ER stress in breast cancer cells.
(a) Western blot showing induction of GRP78 and CHOP in T47D cell line treated with
indicated concentrations of 240 for 24 h. T47D (b) and MCF (c) cells were treated with
indicated concentrations of 240, VER-155008 (10 µM) or EGCG (50 µM) alone or in
combination with tunicamycin (1.5 µg/mL) for 24 h. Whole cell lysates were analyzed by
western blotting for GRP78 and CHOP expression. (d) Western blot showing GRP78 and
CHOP expression in MCF7 cell line upon treatment with 240 (10 µM) alone or in
combination with different ER stress inducers: tunicamycin (1.5 µg/mL), thapsigargin
(300 nM) or 2-deoxyglucose (20 mM). β-Tubulin was used as a control for loading.
Representative blots from multiple consistent experiments are shown.
121
8.2.8 Treatment with 240 causes degradation of HSP70 and HSP90 client proteins
Since GRP78 and HSP70 share 84% sequence similarity, these compounds may
also target HSP70. HSP70 acts in concert with HSP90 and together, they function to
maintain and stabilize several tumor-promoting oncoproteins and signaling molecules
such as Her-2, EGFR, c-Raf, Akt etc. HSP70, together with heat shock organizing protein
(HOP), plays an important role in delivering these client proteins to the HSP90 chaperone
machinery (Kampinga and Craig, 2010; Koren et al., 2010). Consequently, disrupting
HSP70 and HSC70 causes degradation of HSP90 client proteins (Powers et al., 2008),
similar to HSP90 inhibition. However, in contrast with HSP90 inhibition, HSP70
inhibition does not elicit a heat shock response mediated HSP70 induction (Massey et al.,
2010; Powers et al., 2008). Therefore, client protein degradation was used as a marker of
HSP70/HSP90 chaperone machinery inhibition.
In the Her2-overexperssing SKBR3 cell line, treatment with 240 resulted in a
dose-dependent degradation of the HSP70/HSP90 client proteins HER2, Akt and c-Raf,
with no change in the cellular abundance of HSP90 (Fig. 8.5a). As previously reported,
sensitivity of client proteins to degradation differs (Tillotson et al., 2010). Similarly,
another active analogue 236 induced also HER2 degradation, but not the inactive
analogue 235 (Fig. 8.5b) Treatment with HSP90 inhibitor, 17-DMAG, also caused client
protein degradation at a low micromolar concentration, while VER-155008 was effective
only at higher concentrations (Massey et al., 2010). Together with the observed effects on
GRP78 ATPase activity and no effect on HSP90 ATPase activity, these results suggest
122
that 240 acts on GRP78 as well as HSP70 chaperone pathways. Further studies will be
necessary to validate if this effect is through HSP70-binding or due to disruption of
HSP70-cochaperone interactions.
Figure 8.5 240 treatment induces degradation of HSP70/HSP90 client proteins.
SKBR3 cells were treated with 240 at indicated concentrations, HSP90 inhibitor 17-
DMAG (250 nM), VER-155008 (40 µM) (a) or 235 (ATPase assay - inactive) and 236
(ATPase assay - active) (20 µM) (b) for 24 h. Cells were lysed for western blotting and
immunoblotted for HER2, Akt, c-Raf, HSP90 or GAPDH. Representative blot from
multiple consistent experiments is shown.
8.2.9 240 induces cancer cell death
GRP78 has been shown to promote cancer cell proliferation and survival (Dong et
al., 2008; Jamora et al., 1996; Li and Lee, 2006). Through siRNA studies in breast cancer
cell lines, we have also shown that GRP78 knockdown induces cell death (Fig. 7.4).
Therefore, we next assessed the effect of 240 on cancer cell survival in a panel of breast
cancer cell lines. Treatment with 240 inhibited cancer cell viability in a dose-dependent
DMSO
17-DMAG
2.5 5 10 20 40 0.25
VER-155008
240
HER2
Akt
GAPDH
(µM)
SKBR3
Hsp90
C-Raf
GAPDH
a) b)
HER2
Akt
GAPDH
DMSO
235
236
(20 µM)
SKBR3
123
manner in a panel of breast cancer cell lines (Fig. 8.6a, b). More importantly, 240
treatment did not cause significant loss of cell viability in immortalized human foreskin
fibroblast HFF1, except at the highest concentration of 25 µM (Fig. 8.6a). T47D, MDA-
MB-231 and MDA-MB-468 showed a higher sensitivity to 240 treatment. MCF7 colony
survival was also inhibited following 240 treatment (Fig. 8.6c). VER-155008, on the
other hand, showed only moderate cytotoxicity in the high micromolar range. It is also
interesting to note that 240-induced cancer cell death occurs at a concentration where
HSP70/HSP90 client protein degradation is observed.
Figure 8.6 240 preferentially induces cancer cell death.
a)
T47D MCF7 HFF1
0
5
10
15
20
25
GI
50
(µM)
Cell line GI
50
(µ µM)
T47D 3.2 ± 1.7
MCF7 5.5 ± 1.4
MDA-MB-231 3.2 ± 1.3
MDA-MB-468 2.0 ± 0.9
Hs-578-T > 25
BT549 > 25
SKBR3 8.0
DMSO 20 µM 10 µM 5 µM
1 µM 0.5 µM 0.25 µM 2.5 µM
0.05 µM 0.01 µM DMSO 0.1 µM
240 VER-155008
DMSO 40 µM 20 µM 10 µM
2.5 µM 1 µM 0.5 µM 5 µM
0.01 µM DMSO DMSO 0.1 µM
MCF7 MCF7
b) c)
124
(a) Untransformed human foreskin fibroblasts and breast cancer cells (T47D and MCF7)
were treated with indicated concentrations of 240 for 72 h. Cell viability was measured
by MTT assay. Right panel shows selective cytotoxicity of 240 in cancer cells. (b) GI
50
of 240 in a panel of breast cancer cell lines. (c) MCF7 colony survival following 240
treatment. Colonies stained with crystal violet.
8.2.10 240 induces a nonapoptotic cell death.
We next evaluated the characteristics of the cell death induced by 240. 240
induces ER stress with CHOP activation and CHOP mediates ER-stress-induced
apoptosis (Szegezdi et al., 2006). Therefore, we next examined the role of apoptosis in
the cell death. Surprisingly, we did not detect any increase in caspase activation
following 240 treatment of T47D cells at concentrations that inhibited cell viability. Also
no increase in cytotoxicity was detected except at the highest concentration (25 µM),
indicating an absence of membrane permeabilization (Fig. 8.7a). In contrast,
camptothecin treatment resulted in a decrease in cell viability and an increase in caspase
activation, also with no cytotoxicity, indicating early-phase apoptosis. Additional
experiments to detect cleaved caspase or PARP by western blotting also failed to detect
any apoptosis (not shown). Similarly, combination with pan-caspase inhibitor, Z-VAD-
fmk, did not inhibit 240-induced cell death (Fig. 8.7b). We also did not detect any ROS
induction, ruling out oxidative cell death
Since cell death induced by 240 appeared to be caspase-independent, we explored
the role of other cell death mechanisms. We observed a progressive appearance and
accumulation of cytoplasmic vacuoles in T47D cells treated with 240 (Fig. 8.7c).
Cytoplasmic vacuoles commonly occur during processes such as autophagy, necroptosis,
125
paraptosis or lysosomal permeabilization (Broker et al., 2005). Therefore, we first
examined these cells for markers of autophagy. Proteolytic converison of the microtubule
associated protein-1 light-chain 3 (LC3) from its free form (LC3-I, 18 kDa) to a faster
migrating form (LC3-II, 16 kDa), for subsequent lipidation and incorporation into
autophagosomes, is considered a marker of autophagy induction (Klionsky et al., 2008).
As shown in Fig. 8.7d we observed a dose-dependent increase in LC3 conversion at 24 h
in T47D cells treated with 240. To further examine if autophagy contributed to cell death,
we assessed 240 cytotoxcity in the presence of chloroquine. Chloroquine, a
lysosomotropic agent, prevents the fusion of autophagosomes with lysosomes during the
late stages of autophagy. Blocking autophagy with chloroquine partially rescued T47D
cells from 240-induced cell death, suggesting a role for autophagy and/or lysosomal
degradation in its cytotoxicity (Fig. 8.7e). This protection was more evident at the higher
concentrations where extensive vacuolation was observed. Next, we examined the role
of necroptosis as the cell death mechanism. We observed that necroptosis inhibitor,
necrostain-1, significantly reversed 240-induced cytotoxicity (Fig. 8.7f). This suggests a
predominant role for necroptosis in the cell death mechanism of 240. Cell death-induced
by VER-155008 was not affected by combination with any of these cell death pathway
inhibitors.
126
Figure 8.7 240 induces a nonapoptotic, autophagic and necroptotic cell death.
(a) Caspase activation, cytotoxicity and viability of T47D cells treated with indicated
concentrations of 240 or 1 µM camptothecin for 24 h was assessed using ApoTox-glo
triplex assay. (b) Viability of T47D cells following treatment with 240 for 48 h in the
presence or absence of 20 µM Z-VAD-fmk. Viability measured by MTT assay. (c) Light
microscopy image showing extensive cytoplasmic vacuolization following treatment with
240 at indicated concentrations for 24 h. (d) Western blotting showing the conversion of
LC3B in T47D cells treated with 240 for 48 h at indicated concentrations. (e-f) Viability
of T47D cells following treatment with 240 for 48 h in the presence or absence of 15 µM
chloroquine (e) or 40 µM necrostatin-1 (f), measured by MTT assay. Statistical analysis
by one-way ANOVA using Prism 6.0. ** p < 0.01, *** p < 0.001, **** p < 0.0001
DMSO 1.5626 3.125 6.25 12.5 25 Cpt
0
4000
8000
12000
Apoptosis (Luminescence) or
Cytotoxicity (Fluorescence)
Apoptosis (Caspase 3/7) Cytotoxicity
0
10000
20000
30000
Viability (Fluorescence)
Viability
240 (µM)
LC-3B-I -
LC-3B-II -
GAPDH
DMSO
2.5
5
10
20
240
T47D, 24H
DMSO 240 – 12.5 µM 240 – 25 µM
c) d)
e)
b)
e)
a)
127
8.2.11 240 shows inhibition of tumor growth in MDA-MB-468 breast xenograft
model.
Figure 8.8 240 shows tumor growth inhibition up to day 28.
Antitumor activity of 240 administered i.p., on MDA-MB-468 breast cancer xenograft
implanted subcutaneously on athymic nude mice. Tumor volume of vehicle treated and
240 treated mice shown. * p < 0.05
In vivo efficacy of 240 on tumor progression was evaluated in MDA-MB-468
breast cancer xenograft model. MDA-MB-468 cells (1 x 10
6
cells in 50% v/v matrigel)
were implanted by subcutaneous injection into the dorsal flank of athymic nude mice
(Simenson laboratory). When the tumor size reached 40 mm
3
, mice were randomized into
control (n = 5) and treatment groups (n = 5). 240 was administered at 10 mg/Kg 5 times a
week. Compared to the vehicle-treated group, tumor growth in 240-treated mice was
impeded up to day 28 (p < 0.05 at day 28) (Fig. 8.8). However, the results were not
statistically significant beyond day 28. Although the preliminary results with 240 were
0 10 20 30 40
0
100
200
300
400
Time (days)
Tumor volume (mm
3
)
Vehicle control
240
*
128
promising, the tumor growth in control group was progressing slowly and the experiment
was terminated after 4 weeks of treatment.
8.3 Materials and Methods
Generation of Pharmacophore hypothesis: Common feature pharmacophore
hypotheses were generated using Catalyst 4.11 from Accelrys, Inc. installed on a Dell
Precision 690 work station running operating system Red Hat Enterprise Linux 5.
Compounds structures were 3D-minimized using CHARMM-like force-field within
Catalyst and a maximum of 250 conformations for each compounds were generated using
‘Best conformer generation’ method within a 20 Kcal/mol energy range cutoff. The
Poling algorithm implemented within Catalyst was used to generate conformations for all
the compounds
(Smellie et al., 1995). The Catalyst HipHop module was used to generate
the common feature hypotheses
(O.O.T.-M Clement, 2000). HipHop evaluates a
collection of conformational models of the training set molecules and a set of chemical
features from a defined dictionary of features, such as Hydrogen bond acceptor (A),
Hydrogen bond donor (D), Hydrophobic (H), Hydrophobic aromatic (Y), Hydrophobic
aliphatic (Z), Ring Aromatic (R), Positive ionizable (P), Negative ionizable (N), etc., and
identifies configurations or three dimensional spatial arrangements of chemical features
that are common to the molecules in the training set by a pruned exhaustive search. The
top ranking pharmacophores are expected to identify the hypothetical 3D orientation of
the active compounds and the common binding features interacting with the target.
129
Database Search. A common feature pharmacophore was used as a search query to
retrieve molecules with novel chemical structure and desired chemical features from our
in-house multi-conformer Catalyst-formatted database consisting of ~50,000 drug-like
diverse compounds. The Best Flexible Search Databases method in Catalyst was used to
search the database and fit values were calculated to rank the hits.
Substructure and similarity search: Substructure search of the active hits was done
using ChemBioFinder 12.0 software (Perkin Elmer) by keeping basic framework and
heteroatoms of the scaffold constant. Similarity search was carried out using Chemaxon
software (www.chemaxon.com) by considering Extended Connectivity Fingerprints
(ECFP) as descriptors and Tanimoto coefficients as similarity indices.
Compounds and Reagents. Compounds for initial screening and structure
optimization studies were procurred from Enamine (www.enamine.net) and Asinex
(www.asinex.com). All compounds were stored as 10 mM or 50 mM stock in
dimethylsulfoxide (DMSO) at -80 °C. The lead compound 240 (2-(1-((2-morpholino-1-
(p-tolyl)ethyl)amino)ethyl)-5-(thiophen-2-yl)thieno[2,3-d]pyrimidin-4(3H)-one) was
purchased from Enamine (95% purity). Pronase (Roche), VER-155008 (Tocris), EGCG
(LKT Laboratories), tunicamycin, thapsigargin and 2-deoxyglucose (Sigma) and 17-
DMAG (LC labs) were also used. Chloroquine, Necrostatin-1 and Z-Vad-fmk were
purchased from Tocris, respectively.
130
Antibodies, plasmids and recombinant proteins. The following antibodies were
purchased from Cell Signaling Technology: CHOP (L63F7), HER2/ErbB2 (44E7), Akt,
LC-3B, GAPDH, HSP90, c-Raf, CHIP, HSP40, Calreticulin and PDI. Anti-Hsc70 (B-6),
anti-GRP78 (H-129), anti-p62/SQMT1 and anti-β-Tubulin were purchased from Santa
Cruz Biotechnology. Anti-HSP70 was purchased from Enzo life sciences. Plasmid
encoding full length human GRP78 (pETDuet-His-GRP78) was a kind gift from Dr. Lai,
National Tsing Hua University, Taiwan. Full length human HSP70 plasmid (pBluescript-
His-Hsp70) was generously provided by Dr. Wieland, University of Heidelberg,
Germany. Plasmids for GRP78 ATPase domain (pNIC-Bsa4-His-HSPA5
D26-D410
) and
HSP70 ATPase domain (pNIC-Bsa4-His-HSPA1A
M1-N387
) were provided by Dr. Schüler,
Structural Genomics Consortium, Sweden. Human HSP90 recombinant protein (full
length) and human GRP78 recombinant protein (full length) were purchased from Enzo
life sciences and Abcam, respectively.
Cell lines. Breast cancer cell lines T47D, MCF7, MDA-MB-468, Hs-578T, BT-549
and SK-BR-3 were purchased from the American Type Cell Culture (Manassas, VA).
MDA-MB-231 was provided by Dr. Alan Epstein, University of Southern California.
Human foreskin fibroblast cell line HFF-1 was generously provided by Dr. Grandori,
Fred Hutchinson Cancer Research Center; Seattle, WA. Cell lines were cultured in RPMI
1640 or Dulbecco’s minimal essential media supplemented with 10% fetal bovine serum
at 37 °C in a humidified atmosphere of 5% CO
2
. Cell lines were routinely checked for
131
mycoplasma contamination using PlasmoTest (Life Technologies). All experiments were
carried out using cells growing in exponential phase.
Protein expression and purification. The His-tagged recombinant proteins were
expressed in E.Coli. BL21 (DE3) pRARE2 (Rosetta 2 (DE3); Novagen) and purified as
described previously (Wisniewska et al., 2010). Briefly, GRP78 plasmids were
transformed into Rosetta 2 cells, which were then grown in TB broth supplemented with
8g/L glycerol, 50 µg/mL kanamycin or 100 µg/mL Ampicillin at 37 °C to an OD
600
of
approximately 1.5-2. The culture was allowed to cool to 18 °C over a period of 1 h and
induced with 0.5 mM isopropyl thio-β-D-galactoside for 18h. Bacterial cells were
harvested by centrifugation at 4000 × g for 20 min at 4 °C and the cell pellet was
resuspended in lysis buffer (100 mM HEPES, 500 mM NaCl, 10% glycerol, 10 mM
imidazole, 0.5 mM TCEP, pH 8.0 supplemented with EDTA-free protease inhibitor and
Benzonase (Sigma, 2000 U); 1.5 mL lysis buffer/g cell pellet). Cells were lysed by
sonication and the cleared supernatant was purified using 1 mL Ni-NTA resin washed
sequentially with buffer (20 mM HEPES, 500 mM NaCl, 10% glycerol, 0.5 mM TCEP,
pH 7.5) containing increasing concentrations of imidazole (10 mM, 25 mM amd 500
mM). The recombinant enzymes were purified using PD-10 desalting column (GE
Healthcare) with filtration buffer (20 mM HEPES, 300 mM NaCl, 10% glycerol, 0.5 mM
TCEP, pH 7.5). Protein purity was judged by SDS-PAGE.
132
GRP78 ATPase assay. Preliminary screening of the focused-library identified using
the pharmacophore was performed as follows in an Opti plate (Perkin Elmer). Briefly, 0.1
µg (128 nM) recombinant human full-length GRP78 (Abcam, ab78432) in reaction buffer
(100 mM Tris HCl pH 7.4, 20 mM KCl, 6 mM MgCl
2
, 0.017% Triton X-100) was pre-
incubated with 10 µM inhibitors in DMSO for 30 min, followed by the addition of ATP
(1 µM) and further incubation for 2.5 h at 37 °C. ADP generated through ATP hydrolysis
was measured using an ADP glo assay reagent (Promega). Luminescence was read on an
Envision plate reader (Perkin Elmer). Further screening of analogues, performed at 100
µM, and dose response determinations were performed in a black low-volume non-
binding assay plate (Corning No. 3677). Expressed recombinant full-length GRP78 (128
nM or 250 nM where indicated) in reaction buffer (100 mM Tris HCl pH 7.4, 20 mM
KCl, 6 mM MgCl
2
, 0.017% Triton X-100) was incubated with test compounds for 30 min
followed by addition of 2-10 µM ATP. The reaction mixture (10 µL) was incubated at 37
°C for an additional 2.5 h. Generation of ADP through GRP78-mediated ATP hydrolysis
was measured using ADP Hunter Plus assay (Discoverx). Fluorescence (E
x
: 530 nm, E
m
:
590 nm) was read on Synergy II plate reader (BioTek). % Inhibition of ATPase activity
was calculated using DMSO controls (0% inhibition) and ATP controls (100%
inhibition). Effect of compounds on the coupled enzyme-based detection system was also
assessed to control for false positives. Background signal with recombinant protein alone
was comparable to that of wells containing buffer, indicating that there was no significant
contamination of ATP or ADP in the recombinant protein preparation.
133
ATP binding assay. Binding assay was performed based on previously reported
protocol (Williamson et al., 2009). Recombinant full length GRP78 (5 µM) was
incubated with different concentrations of test compounds in binding assay buffer
(100mM Tris HCl, 150 mM KCl, 5 mM CaCl
2
, pH 7.4) for 1 h, followed by addition of
N
6
-(6-amino)hexyl-ATP-5-FAM (Jenna Biosciences) and incubation for an additional 2 h
at RT. K
D
for FAM-ATP for full length GRP78 was 2.5 µM. Fluorescence polarization
was measured on a Synergy multi-mode microplate reader (BioTek).
Chaperone activity inhibition. Chaperone assay was done according to previously
reported protocol (Pubchem assay 547). Briefly, 20 mM firefly luciferase (Sigma) was
incubated in denaturation buffer (6 M guanidinium HCl, 30 mM Tris HCl, and 5 mM
DTT, pH 7.5) for 30 min at RT and 1h at 4 °C. Denatured luciferase was then diluted into
cold refolding buffer (10 mM MOPS, 50 mM KCl, 5 mM MgCl2, 5 mM DTT and 100
mM ATP, pH7.5) to a final concentration of 20 nM and incubated for an additional 1h at
4 °C. At the same time, GRP78 (full length or ATPase domain alone protein, 500 nM)
was incubated in refolding buffer (10 mM MOPS, 50 mM KCl, 5 mM MgCl2, 5 mM
DTT, pH 7.5) for 3 h. Refolding was initiated by incubating denatured luciferase with
GRP78 at 1:1 ratio in a 384-well opti plate (Perkin Elmer) for 3h in dark at RT.
Renaturation of luciferase was measured by addition of luciferase substrate solution
(Promega) and reading luminescence on an Envision multilabel plate reader (Perkin
Elmer).
134
Protease protection (DARTS) assay. Recombinant full length GRP78 (1 µg) was
incubated with DMSO, 240 or VER-155008 in assay buffer (100 mM Tris HCl pH 7.4,
20 mM KCl, 6 mM MgCl
2
, 0.017% Triton X-100) on ice for 1 h. The samples were
digested with Pronase (0.1 µg) for 10 min at 37 °C. Digestion was quenched by adding
5X sample buffer and boiling for 5 min. The digested products were resolved by SDS-
PAGE on a 10% polyacrylamide gel and analyzed with GRP78 (H129) antibody.
Western blotting. T47D, MCF7 and SK-BR-3 cells (0.5 × 10
6
cells/well) were coated
in a 6-well plate and treated with DMSO, test compounds, VER-155008, EGCG or 17-
DMAG, alone or in combination with tunicamycin (1.5 µg/mL) for 24 h. Cells were then
washed with ice cold PBS and lysed in a triple detergent lysis buffer containing 50 mM
Tris (pH 7.5), 0.5 M NaCl, 1% NP-40, 1% sodium deoxycholate, 0.1% SDS, 2 mM
EDTA supplemented with protease and phosphatase inhibitor cocktail. Lysate was
sonicated and centrifuged at 12,000 rpm for 10 min at 4 °C to remove cell debris. Total
protein concentration in the supernatant was measured using BCA protein assay kit
(Pierce Biotechnology). 40 µg of protein lysate was boiled with laemmli sample buffer
for 5 min, resolved on a 10% polyacrylamide gel or 4-20% gradient gel (Bio Rad) and
electroblotted on a PVDF membrane. Membranes were blocked in 5% non-fat milk in
Tris-buffered saline with 0.1% Tween-20 (TBST) for 1 h at RT and incubated overnight
with primary antibodies. Subsequently, the membranes were washed with TBST and
incubated with appropriate HRP-linked or DyLight 800 conjugated secondary antibodies
for 1 h at RT followed by further washing with TBST. The immunoblots were visualized
135
using ECL western blotting substrate (Pierce Biotechnology) on a Chemidoc XRS imager
(Bio Rad) or using Odyssey infrared imaging system (Li-Cor Biosciences).
Cell viability assay. Cell proliferation was assessed by a 3-(4,5-dimethylthiazol-2-yl)-
2,5-diphenyltetrazolium bromide (MTT) assay. Cancer cells were seeded in 96-well
microtitre plates and after overnight attachment, treated with test compounds. After 72 h,
MTT solution (3 mg/mL; 20 µL) was added to each well and cells were incubated for 3 h
at 37 °C. After incubation, media from each well was removed and the dark blue
formazan crystals formed by live cells were dissolved in DMSO (150 µL/well). The
absorbance intensity was measured at 570 nm on a microplate reader (Molecular Devices,
Sunnyvale, CA, USA). 50% growth inhibitory concentration (GI
50
) values were
determined for each drug from a plot of log (drug concentration) versus percentage of cell
growth inhibition using Prism 6.0 (Graphpad). At least 3 independent dose response
experiments with each concentration tested in triplicate were performed for each cell line.
For viability studies in combination with pathway inhibitors (Chloroquine, Necrostatin-1,
z-Vad-fmk), cells were seeded as described above and pre-incubated with inhibitors for 2
h before addition of test compounds.
siRNA transfection. siRNAs for GRP78 (siGRP78-2 and siGRP78-3) from Life
Technologies were used in this study. Briefly, sub-confluent MCF7, T47D and MDA-
MB-231 cells were transfected with siGRP78 (siGRP78-2 and siGRP78-3) or siControl
(Silencer negative control siRNA) at 5 nM using RNAiMax lipofectamine transfection
136
reagent (Life Technologies) according to manufacturer’s protocol. Protein expression was
determined by western blotting at 48 h and cell viability was measured by MTT assay at
48, 72 and 96 h post transfection, respectively.
Clonogenic survival assay. MCF7 cells were seeded at a density of 100 cells/well in a
12-well microtiter plate. After overnight attachment, cells were treated with different
concentrations of 240 and VER-155008. Colonies were allowed to grow for 2-3 weeks
and stained with 0.5% crystal violet.
8.4 DISCUSSION
GRP78 is a member of the HSP70 family of chaperone proteins. It plays key roles
in proper folding of nascent polypeptides, regulating the unfold protein response and cell
survival during ER stress. GRP78 has been implicated in cancer progression and
chemoresistance, and has emerged as an attractive drug target in the recent years (Lee,
2014). We were interested in developing small-molecule inhibitors of GRP78 as very few
small-molecules have been published to date. Towards this goal, we developed a five-
feature pharmacophore model based on the few known GRP78 inhibitors for
identification of novel inhibitor scaffolds through virtual screeening. Through subsequent
in vitro screening of selected subset of the virtual hits in a GRP78 ATPase assay, we
identified a novel thieno[2,3-d]pyrimidin-4(1H)-one-based inhibitor scaffold. Most
potent compound of this class, 240, inhibited GRP78 steady state ATPase activity in the
presence of k
m
ATP with an IC
50
value of 2 µM. 240 inhibited cancer cell viability in a
137
panel of breast cancer cell lines and is 4-10 fold more-selective for cancer cells over
nontransformed fibroblasts. 240 also inhibited GRP78 chaperone function and induced
degradation of HSP70/HSP90 client proteins. As a result of its effect on GRP78, 240
causes ER stress with CHOP induction, which is exacerbated in the presence of an
ongoing ER stress.
While 240 induced CHOP expression and triggered ER stress, no caspase
activation or sensitivity to caspase inhibitor was observed. Instead, 240 exhibited a novel
non-apoptotic mechanism of cell death involving extensive vacuolization and LC3
conversion. Furthermore, cell death was inhibited by chloroquine, autophagy inhibitor,
and necrostatin-1, necroptosis inhibitor. These features indicate an involvement of
caspase-independent cell death mechanisms such as lysosomal membrane
permeabilization induced cell death, paraptosis etc. Previously, antisense depletion of
HSP70 has been shown to elicit an apoptosis-like cell death without caspase activation,
accompanied by lysosomal membrane permeabilization and cathepsin release
(Nylandsted et al., 2000). Similarly, the HSP70 inhibitor, 2-phenylethynesulfonamide
(PES), has been shown previously to induce a non-apoptotic form of cell death
characterized by extensive vacuolation, LC3 conversion and impaired lysosomal
function. More recently, PES cytotoxicity in a lymphoma cell line has been shown to
involve lysosomal membrane permeabilization accompanied with cathepsin release
(Granato et al., 2014). HSP70 plays a role in regulating lysosomal membrane integrity
(Kirkegaard et al., 2010; Nylandsted et al., 2004). Extensive ER vacuolization in the
absence of caspase activation are also considered hallmarks of paraptosis mediated cell
138
death, which has been associated with UPR activation (Tardito et al., 2009). Indeed,
several recent studies with compounds inducing ER stress have described a caspase-
independent cell death with paraptosis like features (Gandin et al., 2014; Kosakowska-
Cholody et al., 2014; Wang et al., 2013).
Overall, 240 represents a novel non-adenosine GRP78 inhibitor that inhibits
GRP78 steady state ATPase activity, induces ER stress, inhibits the HSP70/HS90
chaperone machinery and causes cancer cell death through a non-apoptotic mechanism.
Such inhibitors function as useful investigational probes to examine the effects of
pharmacological inhibition of GRP78 and HSP70 in cancer cells.
139
CHAPTER 9: IDENTIFICATION OF NOVEL N
2
,N
4
-
DIPHENYL-1,3,5-TRIAZINE-2,4,6-TRIAMINE-BASED
GRP78 INHIBITORS BY PHARMACOPHORE-BASED
VIRTUAL SCREENING
9.1 Introduction
The HSP70 family of chaperone proteins plays an important role in regulating
protein quality control and maintaining cellular homeostasis. While HSP70 functions as a
key chaperone in tandem with HSP90 in the cytosol stabilizing client proteins, the ER
resident 78 kDa Glucose regulated protein, GRP78, plays a key role in regulating the
proper folding of nascent polypeptides. GRP78 binds to misfolded proteins, prevents
their aggregation and facilitates their refolding. In addition to its function as an ER
chaperone, GRP78 is an integral component of the unfolded protein response (UPR)
pathway, acting as a molecular switch that transduces ER stress signals to activate
downstream signals. In non-stressed cells, GRP78 remains bound to three ER
transmembrane proteins: PKR-related ER kinase (PERK), inositol requiring enzyme
(IRE-1) and activated transcription factor-6 (ATF6). These act as major transducers of
ER stress and remain inactive when bound to GRP78. During ER stress, GRP78
dissociates from these transmembrane proteins and thus activates them (Lee, 2001). In
turn, these proteins activate various downstream signaling pathways involved in cell
survival. When the stress is severe and the UPR pro-survival signaling cannot restore ER
140
homeostasis, the UPR transducers activate CHOP, caspase-7 and JNK, leading to
apoptotic cell death (Lai et al., 2007; Szegezdi et al., 2006). In addition, GRP78’s
emerging role as a multifunctional receptor, mediating various signaling pathways from
cell surface to nucleus, opens new possibilities regarding its functions and impact on cell
signaling (Ni et al., 2011).
In recent years, GRP78 has gained a lot of focus as an anticancer target. GRP78 is
found to be induced in a wide range of cancers such as breast, gastric, prostate and
hepatocellular carcinomas (Gazit et al., 1999; Lee, 2007). Several in vivo and
retrospective biopsy studies have implicated GRP78 in tumor proliferation and survival
(Dong et al., 2008; Jamora et al., 1996; Li and Lee, 2006). GRP78 overexpression also
results in drug resistance and refractory tumors (Fu et al., 2007; Lee et al., 2006; Pyrko et
al., 2007). While suppressing GRP78 levels using siRNA has been shown to slow cancer
cell progression and overcome drug resistance, the effect of inhibiting its ATPase activity
in cancer cells has not received much attention. GRP78 has a conserved ATPase domain
that regulates the affinity and kinetics of peptide binding. Due to high affinity to ADP
and a hydrophilic binding pocket, designing GRP78 inhibitors have been challenging.
Few GRP78 and HSP70 ATPase inhibitors have been reported so far (Ermakova et al.,
2006; Fewell et al., 2001; Leu et al., 2009; Macias et al., 2011; Massey et al., 2010;
Williamson et al., 2009). In this paper, we report the identification of a novel class of
non-adenosine GRP78 inhibitors that target its ATPase activity. We further describe the
141
antiproliferative effects and mechanism of cell death of these compounds against breast
cancer cell lines.
9.2 Results
9.2.1 Pharmacophore modeling
We have previously reported a common feature pharmacophore in our earlier
study. In short, a five feature pharmacophore (Fig. 8.1) was generated based on reported
GRP78 inhibitors, JMC-10, VER-155008 and JMC-14 (Macias et al., 2011), using the
HipHop module of Catalyst 4.11 from Accelrys Inc. We screened our in-house database
of 50,000 drug-like diverse compounds using the pharmacophore. Through further
screening of the top hits in an in vitro GRP78 ATPase assay, we identified two active hits
from two different scaffolds. The first scaffold was previously reported. The other hit,
249, with a triazine core, will be explored in this article. Figure 9.1a depicts the mapping
of the five-feature pharmacophore onto 249. All five features are mapped almost
completely the compound 249 with a fit value of 4.41. Two out of the three hydrogen
bond acceptors mapped two nitrogen atoms of the triazine ring, while the third hydrogen
bond acceptor was mapped onto the sulfur atom of the diazine-carbothioamide linker.
The hydrogen bond donor and hydrophobic features were mapped onto one of the NH
linked to the triazine moiety and cyclohexyl ring of the compound, respectively. 249
showed potent inhibition of GRP78 ATPase activity at K
m
ATP with an IC
50
value of 2
µM. It also inhibited closely related homologue, HSP70 with a lesser potency, while had
142
not effect on HSP90. Interestingly, 249 inhibited ATPase activity of the full length
GRP78 but not of the isolated ATPase domain fragment (Fig. 9.1b).
Figure 9.1 Pharmacophore mapping onto GRP78 inhibitors (scaffold B).
(a) Initial hit, 249. Above panel is the 3D mapping of the compounds, while below panel
is the 2D representation of the pharmacophore mapping. (b) Inhibitory profile of 249
against GRP78, HSP70 and HSP90. (c) Pharmacophore mapping of analogue, 781.
143
9.2.2 Substructure-similarity search
To explore structure activity relationship as well as to optimize compound 249,
we performed substructure and similarity search. In substructure search, we tried to keep
basic scaffold intact, while in similarity search, we tried to find most similar compounds
in terms of chemical features and functional groups using Tanimoto coefficient.
Substructure and similarity search are commonly used in virtual screening, as
there is a high probability of finding nearest active neighbors through these searches.
Tanimoto similarity is well established and the method of choice for computing
molecular similarity when there are very little SAR data is available
(Willett, 2006). We
identified 9 such analogs from the substructure and similarity search and tested them in
the in vitro activity assay. The initial pharmacophore was mapped onto one of the active
analogs, 781, with four out of five features mapping with a fit value of 3.92 (Fig. 9.1c).
Two out of three hydrogen bond acceptors are mapped onto two nitrogen atoms of the
triazine ring, as seen with 249, while a third hydrogen bond acceptor was absent.
Hydrogen bond donor was mapped onto one of the NH of hydrazine moiety while the 4-
methyl group attached to one of the phenyl group was mapped by the hydrophobic
feature.
9.2.3 Structure activity relationship
The hit compound 249 has a triazine di-amine hydrazine-based central scaffold.
We explored different modifications on the triamine groups using substructure and 2D
finger printing similarity. Replacing the cyclohexylethanethioamide in 249 with a free
144
hydrazine moiety was tolerable with a slight decrease in potency (430). On the other
hand, derivatizing the secondary amine into a tertiary amine (626 and 462) abolished
activity (Table 9.1).
Table 9.1 Biochemical activities and SAR of 249 and analogues
Replacing the hydrazine with a piperidine ring (707) or a chlorine (217) abolished
activity, indicating the importance of the NH groups in the hydrazine group (Table 9.2).
Confirming this, replacing the hydrazine with an amine and the phenyl amine with a
methoxybenzamide did not abolish activity, but decreased potency of ATPase inhibition
(428).
5
± 7
145
Table 9.2. Biochemical activities and SAR of 249 and analogues
Inclusion of an additional phenyl ring on the amine in 430 as in 781 improved
potency of ATPase inhibition. Further substitution on the phenyl ring with a methyl or
methoxy (782) also proved favorable for activity (Table 9.3). GRP78 ATPase activity
inhibition by 781 was also measured using 10 µM ATP (Fig. 9.2a). In the presence of
higher ATP concentration, potency of inhibition by 781 decreased, suggesting weakly
ATP competitive. On the other hand, ATP-mimetic VER-155008 showed potent
inhibition of GRP78 ATPase activity. Similar inhibitory activity of 249, analogues 781
and 782 was obtained using a phosphate detection assay (Fig. 9.2b). Overall, the triazine
triamine core is essential for activity. Aromatic substitutions on two amines together with
N
N
N
R
1
R
3
R
2
146
a free hydrogen on the third amine appear important for activity. Due to limitations in
compound availability, 781 was used as the lead compound in many of the experiments,
instead of the more potent 249.
Table 9.3 Biochemical activities and SAR of 430 and analogues
Figure 9.2 249 and related analogue, 781 inhibit GRP78 ATPase activity.
(a) Steady state ATPase activity of recombinant full length GRP78 (250 nM) was
measured in the presence of 10 µM ATP and indicated concentrations of 781 and VER-
155008. ADP accumulation upon ATP hydrolysis was measured using ADP Hunter Plus
assay. (b) Effect on GRP78 ATPase activity in the presence of compounds – 249, 781
and 782 (100 µM) was measured using CytoPhos reagent which detects inorganic
phosphate released through ATP hydrolysis.
a b
Compound R
1
R
2
% I at 100 µM IC
50
(µM)
430 H H 127.9 5
± 7
781 Ph Me 58.9 1.6
782 Ph OMe 70.4 2.2
N
N
N
N N
H
NH
H
2
N
R
2
R
1
147
9.2.4 Inhibition of chaperone activity
The C-terminal substrate-binding domain of GRP78 is involved in binding
unfolded peptides with exposed hydrophobic residues. Propelled by the nucleotide-
induced conformational changes occurring in the N-terminal ATPase domain, the
substrate binding domain binds, refolds and releases unfolded polypeptides, thus
preventing protein aggregation (Flynn et al., 1989; Gaut and Hendershot, 1993;
Hendershot et al., 1995). We modified a luciferase refolding assay in which recombinant
luciferase is chemically denatured using Guanidium HCl, unfolding its polypeptide chain,
and renatured by incubation with full length GRP78 in the presence of ATP (Pubchem
assay 547). Denatured luciferase has low luminescence signal, but when refolded by
GRP78, luminescence from luciferase is restored. Recombinant GRP78 with only the
ATPase domain, lacking the peptide-binding domain, is unable to restore denatured
luciferase activity (Fig. 9.3a). We found that incubation with 249 inhibited GRP78-
assisted luciferase refolding (Fig. 9.3b). This provides further validation that these novel
compounds as GRP78 inhibitors.
Figure 9.3 Inhibition of GRP78 chaperone activity by 249.
100 µM 50 µM 10 µM
0
20
40
60
80
100
% Inhibition of
luciferase renaturation
249
VER
0
20000
40000
60000
80000
100000
120000
140000
Denat Luc
+dmso
GRP78 alone Deant Luc +
GRP78 + DMSO
GRP78 ATPASE
alone
Denat Luc+
GRP78-ATPase
+ DMSO
Luminescence Signal (cps)
Specificity of Luciferase refolding
A" B "
148
(a) Optimization of GRP78-mediated luciferase renaturation assy. Full length GRP78, but
not ATPase domain protein, refolded chemically denatured luciferase. (b) 249 showed
inhibition of GRP78-mediated luciferase refolding.
9.2.5 249 and 781 induce ER stress
Since inhibition of GRP78 triggers ER stress and UPR activation, we analyzed the
expression of downstream UPR marker, CHOP by western blotting (Fig. 9.4). Increasing
doses of 249 did not induce GRP78 or CHOP under normal growth conditions (Fig.
9.4a). On the other hand, 781 caused an increase in GRP78 and CHOP, indicating ER
stress induction (Fig. 9.4b). On the other hand, in the presence of ER stress environment
induced by tunicamycin, both 249 and 781 induced significant increase in CHOP without
further increase in GRP78, suggesting enhanced ER stress (Fig. 9.4c-d). Similar profiles
were observed with the other analogues 430 and 782.
Figure 9.4 ER stress induction and UPR activation by 249 and 781.
DMSO
1
5
10
10
10
Tu
1
5
10
10
10
249 781 782 249 781 782
GRP78
Actin
CHOP
DMSO Tunicamycin (1.5 ug/mL)
T47D !
MCF7! DMSO
249
430
781
782
VER
DMSO
249
430
781
782
VER
CHOP
DMSO Tunicamycin (1.5 ug/mL)
GRP78
ACTIN
CHOP
- + Tuni
1B4 (uM)
- - - + + +
GRP78
Actin
1 5 10 1 5 10 DMSO
0 2.5 5 10 20
GRP78
CHOP
β-Tubulin
(µM) 781
249
b) a)
d) c)
149
Western blotting showing GRP78 and CHOP expression in T47D (a-c) and MCF7 (d)
cells treated with 249, 781, 430, 782 or VER-155008, in the presence or absence of
tunicamycin (1.5 µg/mL) for 24 h. Actin and β-Tubulin were used as loading controls.
9.2.6 781 causes degradation of HSP70 client proteins
Inhibition of HSP70, a closely related homologue of GRP78, disrupts the HSP90
chaperone system leading to destabilization and degradation of its client proteins by the
proteasome machinery. Figure 9.5a shows decrease in expression of HSP90 client
proteins, HER2 and Akt, following 781 treatment in SKBR3 cells. Surprisingly, there
was no concurrent induction of HSP70 (Fig. 9.5b). 781 did not show a direct inhibition of
HSP90 ATPase activity (data not shown). However, degradation of HSP90 client proteins
occurs at concentrations higher than those at which ER stress is induced or cancer cell
viability is compromised. This could suggest that 781 inhibits GRP78 with a greater
potency and induces ER stress at lower concentrations. Only at higher concentrations,
there is inhibition of HSP70 system causing client protein degradation.
Figure 9.5 781 depletes HSP90 client proteins without HSP70 induction.
DMSO
17-DMAG
2.5 5 10 20 40 0.25
VER-155008
781
(µM)
SKBR3
GAPDH
HSP70
HER2
Akt
LC3B
GAPDH
DMSO
17-DMAG
5 10 20 40 40 0.25
VER-155008
781
(µM)
SKBR3
a) b)
150
SKBR3 breast carcinoma cells were treated with indicated concentrations of 781, 17-
DMAG or VER-155008 for 24 h. Western blot analysis shows expression levels of
HER2, Akt and LC3B (a) and HSP70 (b). GAPDH was used as a loading control.
9.2.7 781 induces cancer cell death
The cytotoxicity of 781 was assessed in a panel of breast cancer cell lines. 781
was moderately cytotoxic in most cancer cell lines with the exception of T47D and
MDA-MB-468, which showed greater sensitivity to 781 treatment (Table 9.4). Against
nontransformed human foreskin fibroblasts, 781 showed similar cytotoxicity profile,
indicating a lack of selectivity for cancer cells (Fig. 9.6a). MCF7 cells treated with 781
also exhibited a diminished clonogenic ability (Fig. 9.6b).
Table 9.4 Cytotoxicity of 781 in a panel of human breast cancer cell lines
GI
50
a
(µ µM)
T47D MCF7 MDA-MB-231 MDA-MB-468 Hs-578-T BT549
781 4.0 ±
1.7 11.2 ±
1.5 12.3 ±
4.5 7.8 ±
0.6 >25 20.7 ±
1.4*
a
GI
50
is the drug concentration that causes a 50% reduction in cell proliferation. GI
50
values are reported
as Mean ± S.D. calculated from at least 3 independent experiments.
*Calculated from two independent experiments.
Figure 9.6 Cytotoxicity profile of 781 in cancer versus nontransformed cells.
a) b)
151
(a) Viability of HFF-1 and T47D cell line were determined following 72 h treatment with
781 by MTT assay. (b) MCF7 cells were treated with 781 or VER-155008 at indicated
concentrations. Cells were allowed to grow 7-10 days to form colonies before staining
with crystal violet.
9.2.8 781 inhibits tumor growth in a breast cancer xenograft model
Based on the in vitro activity of 781 in breast cancer cells, we evaluated its in vivo
efficacy in a xenograft model of breast cancer. Tumor growth was significantly inhibited
by 781 administration (p < 0.001 by t-test) and the efficacy was comparable to that of the
standard chemotherapeutic agent, doxorubicin (Fig. 9.8a). While 781 showed moderate
activity in vitro against cell lines, against solid tumor, 781 potently inhibited tumor
growth. GRP78 has been implicated in doxorubicin resistance (Jiang et al., 2009; Lee et
al., 2006). Therefore, we examined the effect of combining 781 with doxorubicin. In
MDA-MB-231 cells, 781 at 30 µM significantly enhanced doxorubicin induced cell death
(Fig. 9.8b). In line with the in vitro synergistic combination of 781 and doxorubicin,
administration of doxorubicin together with 781 impeded tumor growth even further (Fig.
9.8a). Immunohistochemical staining of tumor sections with Ki-67 showed a significant
decrease in 781-treated tumors, suggesting an inhibition of tumor cell proliferation (Fig.
9.8c). Histological analysis of tumor sections with hematoxylin/eosin staining also
showed extensive areas of necrosis. No significant loss in weight was observed in mice
treated with 781 alone or in combination with doxorubicin (Fig. 9.8d). Histological
analysis of major organs also indicated that treatment with 781 was well tolerated with no
obvious systemic toxicity (Fig. 9.9).
152
Figure 9.8 In vivo efficacy of 781 in MDA-MB-231 breast cancer xenograft.
MDA-MB-231 cells were injected s.c. into the dorsal flank of nude mice. Tumor-bearing
mice were treated with 781 or vehicle. (a) Tumor progression in mice treated with
vehicle (n=6) or 781 (n=4). Data shown are mean tumor volumes (Error bars = SEM).
Treatment schedule is described in the methods section (p < 0.001 by two-tailed t-test).
(b) MDA-MB-231 cells were treated with a combination of 781 and doxorubicin for 72
h. Cell viability was measured by MTT assay. (c) Immunohistochemical staining (Ki-67
and Hematoxylin/Eosin) of tumor sections vehicle, 781, doxorubicin and combination
treatment groups. Representative image of 4 independent sections is shown. Bottom
0 10 20 30
0
200
400
600
800
1000
1200 Control
781
781+Dox
Dox
***
***
**
***
***
***
**
Day
Tumor Volume(mm
3)
Control'(20X)'
Doxorubicin'(20X)'
781'(20X)'
781+Doxorubicin'(20X)'
Control
Dox
781
781+Dox
0
10
20
30
40
50
p = 0.012
p = 0.026
Ki67 Index (%)
a)
c)
Ki67 Ki67
Ki67 Ki67
H & E H & E
H & E H & E
d)
0 10 20 30
0
5
10
15
20
25
30
Control
781
781+Dox
Dox
Day
Weight (g)
b)
0 270 810
0
20
40
60
80
100
120
Doxorubicin concentration (nM)
% Cell Viability
0
30 µM
****
****
**** ****
781
153
panel shows quantification of Ki-67-positive nuclei using ImageJ software. (d) Body
weights of 781, doxorubicin and vehicle-treated mice over the duration of treatment.
Figure 9.9 781 treatment did not cause any significant systemic toxicity.
Histochemical analysis of organ sections from MDA-MB-231 xenograft mice treated
with vehicle, doxorubicin, 781 or combination.
781$
781+Dox$
Live r & Kidne y& He ar t& Lung& Sple e n& Pa n crea s&
Control-MDA-MB-231
Doxor ubicin*
Liver Kidney Heart Lung Spleen Pancreas
154
9.3 Materials and Methods
9.3.1 Generation of pharmacophore hypothesis and database search
Detailed method of building pharmacophore hypothesis has been reported in the
earlier study. In short, common feature pharmacophore hypotheses were generated using
HipHop module of Catalyst 4.11 from Accelrys, Inc. installed on a Dell Precision 690
work station running operating system Red Hat Enterprise Linux 5. HipHop generates a
common three dimensional arrangements of chemical features by evaluating
conformations of a set of training molecules from a defined dictionary of chemical
features, such as hydrogen bond acceptor, hydrogen bond donor, hydrophobic, ring
aromatic, etc. The common feature pharmacophore was used to screen in-house multi-
conformer Catalyst-formatted database consisting of ~50,000 drug-like diverse
compounds through Best Flexible Search method.
9.3.2 Substructure and similarity search
Substructure search of the active hits was carried out within ChemBioFinder 12.0
software from Perkin Elmer by keeping basic framework and heteroatoms of the scaffold
constant. Similarity search was carried out using Chemaxon software considering
Extended Connectivity Fingerprints (ECFP) as descriptors and Tanimoto coefficients as
similarity indices (www.chemaxon.com). When there are two molecules with a and b bits
sets, respectively and c is the common bits set between these two molecules, the
Tanimoto similarity coefficient between two molecules is calculated as below:
155
The Tanimoto coefficient ranges from 0 (no similarity) to 1 (exact similar) values.
9.3.3 Xenograft studies
MDA-MB-231 cells (2.0 x 10
6
) in a 100 µL suspension of 50% Matrigel/50%
PBS (v/v) were injected subcutaneously into the dorsal flank of 8-week old female
athymic nude mice (The Jackson Laboratory, Bar Harbor, Maine). All animal
experiments were done in accordance with protocols approved by the Institutional
Animal Care and Use Committee. Tumor size was monitored twice a week through
caliper measurement and tumor volumes were calculated using the formula: 0.5 x D x d
2
,
where D and d were the longest and shortest perpendicular diameters, respectively. Mice
were randomly grouped (n=6 in control group and n=4 in three treatment groups) when
average tumor size reached 50 mm
3
. Control mice (n=6) received vehicle (5% DMSO,
60% Propylene glycol and 35% Saline v/v, 100 µL) alone. Doxorubicin (4 mg/kg in 5%
DMSO and 95% Saline, 100 µL) was used as positive control and administrated by
intraperitoneal injection once a week. 781 (20 mg/kg in vehicle, 100 µL) was
administrated via intraperitoneal injection five times weekly alone or in combination with
doxorubicin. Tumor volumes and body weights were measured twice weekly to monitor
tumor burden and weight loss during treatment. Study was concluded when the tumor
size in control group reached 1000 mm
3
. At the end of the experiment, animals were
156
euthanized and tumor, kidney and liver were collected, fixed, and paraffin embedded for
histology. Tumor volumes were compared using unpaired t test.
9.4 Discussion
Cancer cells exploit cellular stress response mechanisms, such as the unfolded
protein response, to survive through cytotoxic treatments. A key mediator of the unfolded
protein response, the 78kDa glucose regulated protein (GRP78) chaperone, has frequently
been found to be overexpressed in breast cancer. In addition to providing a growth
advantage to cancer cells, GRP78 induction leads to drug resistance. Suppressing GRP78
levels using siRNA has been shown to slow cancer cell progression and overcome drug
resistance. Currently, there are only few known GRP78 inhibitors. Here, we report a
novel class of triazine-triamine-based small-molecule inhibitors of GRP78 ATPase
activity. The most active compound, 249, showed potent inhibitory activity with an IC
50
value of 2 µM and induced CHOP significantly under conditions of ER stress. Another
close analogue, 781, was pursued further to its effects on cancer cells. 781 showed
moderate cytotoxicity against a panel of breast cancer cell lines and increased sensitivity
to doxorubicin, further supporting the role for GRP78 in cancer survival and drug
resistance. In a breast cancer xenograft mode, 781 was able to potently inhibit tumor
growth. These compounds are promising starting points for further optimization into
more potent GRP78 inhibitors.
157
CHAPTER 10: IN VITRO AND IN VIVO ANTICANCER
EFFECTS OF A NOVEL GRP78 INHIBITOR WITH A
QUINOLINE SCAFFOLD
10.1 Results
10.1.1 Identification of quinolone-based compounds through high throughput
screening
Our laboratory previously reported the generation of an in-house library of 50,000
diverse compounds, representing a chemical space of 8 million commercially
compounds, built employing a machine learning technique. This library has been
successfully used to identify novel integrase inhibitors through various high-throughput
screening endeavors (Serrao et al., 2013b). From this library, we randomly screened 200
compounds using an in vitro GRP78 activity assay, described previously for
identification of inhibitors of steady state GRP78 ATPase activity. Initial screening at 10
µM using K
m
ATP (1 µM) yielded very few active hits, underlining the difficulty in
identifying non-adenosine-based GRP78 ATPase inhibitors. Therefore, we selected
compounds that showed at least 20% inhibition of GRP78 ATPase activity in this screen
for further evaluation. Repeat screening and dose response determination confirmed two
of the initial hits. Here, we report the identification and biological characterization of 3-
(benzodioxolyl)-3-(quinolin-8-ylthio)-1-(thiophen-2-yl)propan-1-one (148) and N-
158
(benzodioxolyl(5-chloro-8-hydroxyquinolinyl)methyl)butyramide (170), which showed
promising GRP78 ATPase inhibition.
Figure 10.1 Identification of GRP78 ATPase inhibitor.
(a) Screening of a focused library of small-molecule compounds selected through
structure-based docking. 170 compounds were screened at 10 µM in an ATPase assay
using recombinant human GRP78 (128 nM, Abcam ab78432) in the presence of 1 µM
ATP, hydrolysis was measured using ADP glo reagent. Each point represents the average
of duplicate wells for a single compound. Compounds showing >25% inhibition of
control GRP78 ATPase activity were selected as hits. Compounds 148 and 170 showed
good activity and selectivity in this screen.
Table 10.1 Inhibitory activities of active compounds 148 and 170
0 50 100 150 200
0
25
50
75
100
125
150
Compound number
ATPase activity - % Control
148$
170$
N
S
O
S
O
O
N
Cl
OH NH O
O
O
148
170
Compound
GRP78 ATPase
% Inhibition of FGFR kinase
activity (at 50 µM)
IC
50
(µM)
148 29.9 ±
25.6 21.5
170 8 ±
5 19
159
10.1.2 Structure activity relationship
Structure activity relationship of 148 was explored next. 148 has a quinolinyl,
benzodioxo and thiophene group on a mercaptopropan-1-one backbone. The importance
of each of these groups was investigated. Bioisosteric replacement of the thiophene ring
with a furan ring (210, 212) was tolerable. Opening up the benzodioxo ring to a
dimethoxybenzene ring decreased potency of inhibition (212). Replacing the benzodioxo
ring with a fluorobenzene (213) or difluoro methoxybenzene (210) reduced potency of
ATPase inhibition (Table 10.2).
Table 10.2 Biochemical activities and SAR of 148 and analogues
Compound R
1
R
2
% I at
100 µM
IC
50
(µM)
148 S
115 30
210 O
54.2
212 O
49.0
213 S
63.8
N
S R
2
O
R
1
O
O
O F
F
O
O
F
160
Table 10.3 Biochemical activities and SAR of 148 and analogues
Replacing the quinolinyl moiety with a methylfuran (208) or a N-
phenylacetamide (219) abolished activity (Table 10.3). Replacing the quinolone ring
with benzene or a toluene ring results in an overall reduction in potency. Similarly,
replacing the benzodioxo ring with a dimethoxybenze or a trimethoxybenze substitution
did not improve activity. Interestingly, a furan ring appeared to be more favorable than
the thiophene ring (Table 10.4). Overall, SAR of 148 revealed that on a propanone
backbone, a quinolin-8-thiol or a 5-chloro, 8-hydroxyquinoline ring were more favorable.
Replacing the benzodioxo with other phenyl derivatives may also be tolerable. Also, a
furan ring instead of the thiophene may prove more active.
Compound R
1
R
2
% I at
100 µM
IC
50
(µM)
148 S
115 30
208 O
8.1
219 O
17.2
R
2
S
O
O
O
R
1
N
O
H
N O
161
Table 10.4 Biochemical activities and SAR of 148 and analogues
10.1.2 148 and 170 induce ER stress
GRP78 knockdown results in UPR activation with induction of the pro-apoptotic
transcription factor CHOP (Li et al., 2008). Similarly, inhibition of GRP78 ATPase
activity results in the accumulation of misfolded proteins, which could trigger ER stress
Compound R
1
R
2
R
3
% I at 100
µM
209 O
H 40.4
211 S
H 16.7
217 S
CH
3
29.0
220 S
H 42.7
218 O
CH
3
55.2
S R
2
O
R
1
R
3
O
O
O
O
O
O
O
O
O
O
O
O
162
(Hendershot et al., 1995). Therefore, we next tested the effect of 148 and 170 on ER
stress induction. As shown in Fig. 10.2a, in T47D cells, under normal growth conditions,
both 148 and 170 caused mild ER stress at higher concentrations. In combination with
another ER stress inducer such as tunicamycin, both compounds resulted in significant
CHOP induction, indicating severe ER stress. These results were comparable to VER-
155008 (Fig. 10b). Similar results were also obtained with MCF7 cells (Fig. 10c).
Figure 10.2 148 and 170 induce ER stress in breast cancer cells.
Western blotting shows induction of GRP78 and CHOP following 24 h treatment with
148 and 170 at indicated concentrations, under normal growth conditions and in the
presence of tunicmaycin (1 µg/mL). β-Tubulin was used as loading control.
CHOP
β-Tubulin
DMSO
1 5 10 10 50
148
VER-155008
EGCG
DMSO
1 5 10 10 50
148
VER155008
EGCG
GRP78
T47D
(µM)
Tunicamycin - - - - - - + + + + + +
CHOP
β-Tubulin
DMSO
1 5 10 10 50
148
VER-155008
EGCG
DMSO
1 5 10 10 50
148
VER155008
EGCG
GRP78
MCF7
(µM)
Tunicamycin
- - - - - - + + + + + +
CHOP
β-Tubulin
10 5 10 50
170 EGCG
GRP78
T47D
(µM)
Tunicamycin + - + - + - + - + - + -
148
5
DMSO
b c
a
163
10.1.3 148 disrupts HSP70/HSP90 chaperone machinery and induces client protein
degradation.
HSP90 stabilizes the active conformations of several oncoproteins while HSP70
binds and delivers these client proteins to HSP90. Inhibition of HSP90 and HSP70 elicit a
well-characterized molecular signature, which includes proteasomal degradation of client
proteins and induction of HSP70 (Aherne et al., 2003; Powers et al., 2008). To assess the
effect of 148 on the HSP70/HSP90 chaperone machinery, we determined client protein
expression as well as heat shock induction of HSP70. 148 induces degradation of client
proteins HER2, EGFR and Akt at a concentration of 2.5 µM or higher (Fig. 10.3a). VER-
155008 causes a similar effect only at higher concentrations.
148 also caused an increase in HSP70 levels at these concentrations with no effect
on HSP90 expression (Fig. 10.3b). In addition, Hsc70 levels were induced but at much
higher concentrations (Fig. 10.3c). HSP70 induction is commonly seen with HSP90
inhibitors. However, we ruled out inhibition of HSP90 (ATPase activity) by 148 (26.2 %
inhibition at 100 µM). On the other hand, silencing only Hsc70 using siRNA could also
cause a concurrent induction in HSP70, although the reverse does not (Powers et al.,
2008). VER155008, an HSP70/Hsc70 inhibitor, also causes an increase in HSP70 levels
at concentrations inducing client protein degradation, but with no effect on Hsc70 (Fig.
10.3b,c, (Massey et al., 2010). This suggests that 148 could either have an inhibitory
effect on Hsc70, considering the high degree of homology with GRP78, or 148 could
disrupt the HSP90 chaperone machinery through another mechanism.
164
Figure 10.3 Effect of 148 treatment on abundance of HSP70/HSP90 client proteins.
Western blotting analysis of HSP70/HSP90 client proteins – HER2, Akt and EGFR (a) as
well as HSP70, Hsc70 and HSP90 (b, c) in SKBR3 cells following 24 h exposure to 148,
VER155008 or 17-DMAG at indicated concentrations. GAPDH was used as a loading
control.
10.1.4 148 inhibits cell survival and induces cancer cell death in a panel of breast
cancer cell lines.
148 showed potent inhibition of cancer cell proliferation in a panel of breast
cancer cell lines, with sub-micromolar IC
50
values (Table 10.5). Similar results were
obtained for inhibition of colony formation in MCF7 cells (Fig. 10.4).
DMSO
17-DMAG
2.5 5 10 20 40 0.25
VER-155008
148
Hsc70
β-Tubulin
(µM)
SKBR3
DMSO
17-DMAG
5 2.5 1 0.5 40 0.25
VER-155008
148
(µM)
GAPDH
Hsp70
Hsp90
DMSO
17-DMAG
5 2.5 1 0.5 40 0.25
VER-155008
148
HER2
Akt
GAPDH
(µM)
GAPDH
EGFR
LC3B-I -
LC3B-II -
a) b)
c)
165
Table 10.5 Cytotoxicity of 148 in a panel of human breast cancer cell lines
GI
50
a
(µ µM)
T47D MCF7 MDA-MB-231 MDA-MB-468 Hs-578-T BT549
148 0.3 ±
0.1 0.3 ±
0.1 0.4 ±
0.1 0.9 ±
0.6* 0.4 ±
0.2 1.1 ±
0.7
a
GI
50
is the drug concentration that causes a 50% reduction in cell proliferation. GI
50
values are
reported as Mean ± S.D. calculated from at least 3 independent experiments.
*Calculated from two independent experiments.
Figure 10.4 Reduction of colony forming ability of MCF7 cells following 148
exposure.
MCF7 cells were treated with 148 or VER-155008 at indicated concentrations. Cells
were allowed to grow 7-10 days to form colonies before staining with crystal violet.
It is interesting to note that 148-induced cancer cell death occurs at a
concentration much lower than that required for HSP70/HSP90 client protein
degradation. For instance, in SKRB3 cell line, IC
50
for cytotoxicity is 0.3 µM, while
significant HER2 degradation occurs only at 2.5 and 5 µM. This indicates that cell death
is not primarily driven by the degradation of the oncogenic client proteins and occurs
possibly due to another mechanism.
DMSO%
40%µM% 20%µM% 10%µM%
2. 5%µM% 1%µM% 0. 5%µM% 5%µM%
0. 0%µM% DMSO% DMSO% 0. 1%µM%
VER$
DMSO 20 µM 10 µM 5 µM
1 µM 0.5 µM 0.25 µM 2.5 µM
0.05 µM 0.01 µM DMSO 0.1 µM
148$
$$
166
10.1.5 Mechanism of cell death
Next, we sought to investigate the mechanism underlying 148-induced
cytotoxicity. 148 treatment induces ER stress and activates UPR as evidenced by an
increase in GRP78 and CHOP expression (Fig. 10.2). ER stress has been shown to induce
autophagy as a survival pathway to clear aggregation of polyubiquitinated proteins and
protect against cell death (Ding et al., 2007; Ogata et al., 2006; Yorimitsu and Klionsky,
2007). Furthermore, GRP78 has been shown to play an important role in ER-stress
induced autophagy. Upon GRP78 knockdown in cells under ER stress, LC3B conversion
occurs while autophagosome formation is inhibited (Li et al., 2008). Therefore, we
assessed the expression of LC3B, which is a marker of autophagy. Upon initiation of
autophagy, LC3B is converted from its free form (LC3B-I) to a lipidated form (LC3B-II)
and is incorporated into autophagosomes (Murrow and Debnath, 2013). Treatment with
148 resulted in increase in LC3B conversion indicating autophagy (Fig. 105a). Next, we
examined if cell death caused by 148 is dependent on autophagy using chloroquine, a
lysosomotropic agent and an inhibitor of autophagy. Interestingly, at the lower
concentration of 0.3 µM, inhibiting autophagy seemed to have a protective effect limiting
148-induced cell death. On the other hand, at the high concentration of 0.6 µM,
combination with chloroquine exacerbated 148-induced cell death, indicating a protective
role of the ER stress-induced autophagy (Fig. 105b). We also examined if necroptosis,
another cell death mechanism, was involved in 148 cytotoxicity. Cell viability following
148 treatment in the presence of necrostain-1, a necroptosis inhibitor, was assessed.
Similar to the results with chloroquine, combination with necrostain-1 protected against
167
148-induced cytotoxicity at 0.3 µM whereas enhanced cell death at 0.6 µM (Fig. 105c).
These bimodal concentration-dependent effect on cytotoxicity in combination with
autophagic and necroptosis inhibitors are interesting and warrant further investigation.
Figure 10.5 Mechanism of cell death induced by 148.
(a) T47D cells were treated with 148 at indicated concentrations for 24 h. LC3B-I/II
levels were analyzed by western blotting. (b,c) Viability of T47D cells treated with 148
(0.31 and 0.62 µM) for 48 h in the presence or absence of (b) chloroquine (CQ; 15 µM)
or (c) necrostatin-1 (Nec1, 40 µM) measured by MTT assay.
10.1.6 148 impedes cancer growth in a breast cancer xenograft model
To investigate the in vivo antitumor efficacy of GRP78 inhibitors, xenograft
studies were performed in athymic nude mice. The maximum tolerated dose (MTD) for
148 was determined by dose-escalation experiment in these mice and no significant
toxicity was observed up to 60mg/kg body weight. Subcutaneous human breast cancer
xenograft was established on the dorsal flank of the immune-deficient mice, and treated
with 148 or vehicle for 29 days until tumor size in the control group reached 1000 mm
3
.
At 20 mg/kg, 148 treatment significantly suppressed growth of tumors from day 8, as
compared with the vehicle controls (Fig. 10.6a). H&E staining of tumor sections showed
0 0.3125 0.625
0
20
40
60
80
100
120
148 (µM)
% Cell Viability
- Nec1
+ Nec1
***
***
0 0.3125 0.625
0
20
40
60
80
100
120
148 (µM)
% Cell Viability
- CQ
+ CQ
***
*
b) c) a)
0 0.5 1 2.5 5
LC3B - I
- II
GAPDH
T47D, 24 h
(µM) 148
168
extensive regions of necrosis in 148-treated mice (Fig. 10.6b). The FDA-approved
chemotherapeutic agent doxorubicin was used as positive control in this experiment, and
its combination with GRP78 inhibitors was also investigated. 148 showed comparable
efficacy with doxorubicin administered at 4 mg/kg once a week. Interestingly,
pretreatment with 148 followed by weekly once doxorubicin treatment was able to inhibit
tumor growth even further. While the average tumor size in control group reached 956.1
± 102.3 mm
3
on day 29, the average tumor size was 237.3 ± 47.7 mm
3
for doxorubicin
treatment and 74.2 ± 18.9 mm
3
for combination with 148.
Control'(20X)' 148'(20X)'
148'+'Doxorubicin'(20X)' Doxorubicin'(20X)'
0 10 20 30
0
200
400
600
800
1000
1200
Control
148
148+Dox
Dox
**
***
**
**
**
*
**
Day
Tumor Volume(mm
3)
Ki67 Ki67
Ki67 Ki67
H & E H & E
H & E H & E
a)
c)
b)
0 10 20 30
0
5
10
15
20
25
30
Control
148
148+Dox
Dox
Day
Weight (g)
Control
Dox
148
148+Dox
0
10
20
30
40
50
p = 0.017
p = 0.004
Ki67 Index (%)
d)
169
Figure 10.6 GRP78 inhibitor 148 inhibits tumor growth in vivo.
MDA-MB-231 cells were injected s.c. into the dorsal flank of nude mice. Tumor-bearing
mice were treated with 148 or vehicle. (a) Tumor progression in mice treated with
vehicle (n=6) or 148 (n=4). Data shown are mean tumor volumes (Error bars = SEM).
Treatment schedule is described in the methods section. Significant reduction in tumor
volumes on day 29 of 148 treatment (p < 0.01 by two-tailed test), which was comparable
to doxoribucin. (b) Ki-67 staining of proliferating cells showed a significant decrease in
148-treated tumors which also show extensive areas of necrosis, as seen by histological
analysis of tumor sections with hematoxylin/eosin staining. Representative image of 4
independent sections is shown. (c) Quantification of Ki-67-positive nuclei using ImageJ
software shows significant decrease in tumor cell proliferation in 148-treated groups (d)
Comparison of body weights of 148, doxorubicin and vehicle-treated mice over the
duration of treatment show no significant weight loss.
Mice were also monitored daily and weighed twice a week to detect potential
drug related toxicity. In the combination treatment group with 148 and Doxorubicin, one
mouse died on day 21 with weight of 21.1g, but the other three mice on the same
treatment did not show any toxicity symptom throughout the study. In other treatment
groups, no systemic symptoms of toxicity such as weakness, weight loss or lethargy were
detected. The average body weight of the vehicle-treated or 148-treated mice did not vary
significantly throughout the duration of the study. H&E staining also showed no gross
morphological differences (Fig. 10.6d, Fig. 10.7). These results indicate that 148 as a
single agent reduced tumor growth without any obvious adverse effects to normal tissues.
170
Figure 10.7 148 treatment did not cause any significant systemic toxicity.
Histochemical analysis of organ sections from MDA-MB-231 xenograft mice treated
with vehicle, doxorubicin, 148 or combination.
Control-MDA-MB-231
Doxor ubicin*
Liver Kidney Heart Lung Spleen Pancreas
148$
148+Dox$
Live r & Kidne y& He ar t& Lung& Sple e n& Pa n crea s&
171
10.2 Materials and Methods
10.2.1 ATPase assay
ATP turnover and ADP generation was measured using ADP glo assay kit
(Promega). Reaction mixtures were prepared in 384-well white Opti plate (Perkin Elmer)
and contained 0.1 µg His-tagged recombinant protein (full length or ATPase domain) and
increasing concentrations of respective compounds in standard ATPase assay buffer.
Reactions were pre-incubated with compounds for 30 min at 37°C, followed by addition
of 1 µM ATP and further 2h incubation. Luminescence was read on an Envision plate
reader (Perkin Elmer).
10.2.2 Counter-screening against FGFR kinase
FGFR kinase activity was measured using a HTScan kinase assay kit (Cell Signaling
Technology). Briefly, recombinant GST-FGFR-1 fusion protein (50 ng) in kinase
reaction buffer was preincubated with test compounds (at a final concentration of 50 µM)
for 15 min, followed by incubation with a biotinylated substrate peptide (1.5 µM) in the
presence of ATP (20 µM) for 60 min. The level of phosphorylation was measured using
anti-phosphotyrosine and HRP-linked secondary antibodies (Cell Signaling Technology).
Absorbance at 450 nm was measured on a microplate reader (Molecular devices,
Sunnyvale, CA). % Inhibition of enzyme activity by test compound compared to control
was calculated.
172
10.2.3 Xenograft studies
MDA-MB-231 cells (2.0 x 10
6
) in a 100 µL suspension of 50% Matrigel/50%
PBS (v/v) were injected subcutaneously into the dorsal flank of 8-week old female
athymic nude mice (The Jackson Laboratory, Bar Harbor, Maine). All animal
experiments were done in accordance with protocols approved by the Institutional
Animal Care and Use Committee. Tumor size was monitored twice a week through
caliper measurement and tumor volumes were calculated using the formula: 0.5 x D x d
2
,
where D and d were the longest and shortest perpendicular diameters, respectively. Mice
were randomly grouped (n=6 in control group and n=4 in three treatment groups) when
average tumor size reached 50 mm
3
. Control mice (n=6) received vehicle (5% DMSO,
60% Propylene glycol and 35% Saline v/v, 100 µL) alone. Doxorubicin (4 mg/kg in 5%
DMSO and 95% Saline, 100 µL) was used as positive control and administrated by
intraperitoneal injection once a week. 148 (20 mg/kg in vehicle, 100 µL) was
administrated via intraperitoneal injection five times weekly alone or in combination with
doxorubicin. Tumor volumes and body weights were measured twice weekly to monitor
tumor burden and weight loss during treatment. Study was concluded when the tumor
size in control group reached 1000 mm
3
. At the end of the experiment, animals were
euthanized and tumor, kidney and liver were collected, fixed, and paraffin embedded for
histology. Tumor volumes were compared using unpaired t test.
173
10.3 Discussion
GRP78 has emerged as a novel anticancer drug target, given its role in cancer cell
survival, tumor progression and drug resistance. However, design of GRP78 inhibitors
has been slow. Unlike HSP90, GRP78 and its closely related homologue HSP70 exhibit
potent ATP binding, with Kd for ATP in the low micromolar range. Such tight binding
combined with high cellular ATP concentrations makes identification of ATP
competitive inhibitors challenging (Massey, 2010). Given the paucity of GRP78
inhibitors, we sought to identify novel non-adenosine based GRP78 ATPase inhibitors
using a high throughput ATPase screening of our enriched in-house library of diverse
drug-like small-molecules. We have identified a novel class of small-molecules,
containing a quinolone scaffold, with GRP78 inhibitory activities. Two compounds –
148, with a quinolin-8-thiol core, and 170, with a 5-chloro-8-hydroxyquinoline core,
showed promising GRP78 inhibitory activities. We found that in addition to inhibition of
GRP78 ATPase activity, 148 induces depletion of HSP70/HSP90 client proteins,
indicating a disruption of the chaperone machinery. Our results further show that 148 and
170 cause ER stress overload in the presence of tunicamycin and show potent inhibition
of cancer cell survival. 148 also showed promising in vivo efficacy in breast cancer
xenograft model with no significant systemic toxicity.
The quinolin-8-thiol and 8-hydroxyquinoline scaffolds are widely found in
several enzyme inhibitors (Adlard et al., 2008; Chockalingam et al., 2011; Johnson et al.,
2008; Sato et al., 2006), antibiotics (Wiles et al., 2010), antimalarials (Achan et al., 2011)
and anticancer agents (Solomon and Lee, 2011). Our lab previously reported quinolone
174
carboxylic acids as HIV-1 integrase inhibitors (Dayam et al., 2008) and more recently, 8-
hydroxyquinoline derivatives as inhibitors of the interaction between viral integrase and
the cellular protein – Lens epithelium-derived growth factor/p75 (Serrao et al., 2013a).
Some of these compounds also showed potent cytotoxicity in Lncap cell line.
LEDGF/p75 is a stress response oncoprotein, which promotes cancer cell survival
(Matsui et al., 2001; Sharma et al., 2000). It is interesting to note here that LEDGF/p75
has been shown to regulate HSP70-2-mediated lysosomal stability (Daugaard et al., 2007;
Gyrd-Hansen et al., 2004; Nylandsted et al., 2004). Inhibition of LEDGF/p75 or HSP70
through siRNA destabilizes lysosome leading to a caspase-independent cell death. We
have shown here that quinoline-based compound 148 inhibits GRP78 as well as the
chaperone functions of HSP70/HSP90 pathway and induces a non-apoptotic cell death
that involves activation of autophagy and necroptosis. Previous studies with compounds
that inhibit HSP70 or induce ER stress have also reported non-apoptotic cell death
mechanisms involving lysososmal destabilization and cytoplasmic vacuolization
(Granato et al., 2014; Kosakowska-Cholody et al., 2014; Wang et al., 2013). Therefore, it
will be interesting to examine the role of lysosomes in 148-induced cell death. In
summary, we have identified novel quinoline-based inhibitors of GRP78 and HSP70
chaperone system. These show potent anticancer activity and in vivo efficacy in breast
xenograft model and elicit a novel mechanism of cell death, encouraging further
structural optimization and preclinical development.
175
CHAPTER 11: CONCLUSIONS AND FUTURE
DIRECTIONS
As outlined in the introduction (chapter 1), cancer cells exploit cellular stress
response pathways to thrive under unfavorable growth conditions and to evade cell death.
We have identified several novel small-molecule compounds targeting two cellular stress
response pathways – oxidative stress response (target: GSTO1) and unfolded protein
response (target: GRP78). By inhibiting these stress response pathways with GSTO1 and
GRP78 inhibitors, we have shown that cancer cell survival is hampered.
GSTO1 is a novel GST isoform with a cysteine residue in the active site. It is
overexpressed in several cancers as well as in drug-resistant cell lines. Through extensive
structure-activity relationships, screening, biochemical assays, and X-ray crystallography,
we have identified several potent and selective GSTO1 inhibitors. Through
crystallography and mechanistic studies, we have shown that the lead compound C1-27 is
a reversible covalent inhibitor. Reversible covalent inhibitors offer potent inhibition,
prolonged duration of action and target selectivity, with minimal irreversible off-target
effects and have become an attractive drug design approach in the recent years (Lee and
Grossmann, 2012). However the compounds we identified in this study as well as
previously reported GSTO1 inhibitors contain electrophilic and labile moieties that suffer
from reactivity with cellular glutathione and instability in solution. Therefore, tempered
176
electrophiles are needed to increase stability and improve selectivity. We have
determined the crystal structure of GSTO1 in complex with our potent inhibitors. Using
the structural information on key amino acid residues important for GSTO1 inhibition,
newer generation of GSTO1 inhibitors with a non-chloroacetamide scaffold can now be
designed.
We have also demonstrated in our study that knocking down GSTO1 by siRNA or
inhibiting its catalytic activity by small-molecule inhibitors blocked cell proliferation and
caused cytotoxicity in a panel of cancer cell lines. The lead compound, C1-27, reduced
cell viability, caused S-phase arrest, and induced apoptotic cell death. Furthermore, C1-
27 targets GSTO1 in tumor mass and shows in vivo efficacy in a xenograft model of
colon cancer, validating GSTO1 as an important target in oncology. We further observed
a moderately increased sensitivity of KRAS mutant cancer cell lines to GSTO1 inhibition
than cells with wild-type KRAS in a small panel of cancer cell lines. The RAS-RAF-
MEK-ERK pathway is commonly deregulated in several cancers and is associated with
poor treatment outcomes. Therefore, selectively targeting KRAS mutant cancer cells
using GSTO1 inhibitors could be a potential approach for anticancer therapy.
Furthermore, cell death induced by GSTO1 inhibitor appeared to be Ras-signaling
dependent. Therefore, the role of GSTO1 and chloroacetamide-containing compounds in
the RAS-selective lethality merits further investigation.
177
Lastly, our results have shown that GSTO1 inhibitors block IL-1β secretion.
Furthermore, GSTO1 and IL-1β are co-overexpressed in many cancers. Therefore,
targeting IL-1β−dependent cancers could be an attractive niche for GSTO1 inhibitors.
For example, colorectal cancer is the third leading cause of cancer-related deaths in the
US. Despite significant advances in radiation and chemotherapy, only 20% of patients
who receive pre-operative chemoradiation show complete response. A major cause for
treatment failure is the inherent and acquired resistance to radiation and chemotherapy.
GSTO1 is overexpressed in colorectal cancer and confers chemo- and radio-resistance.
Similarly, head and neck cancer is another cancer with significant overexpression of
GSTO1. It is also the sixth leading cause of cancer-related deaths with a poor prognosis
and patients frequently develop recurrence and distant metastases. These cancers have a
highly invasive phenotype and secrete high levels of several cytokines including IL-1β.
GSTO1 inhibitors could have a two-fold effect of inhibiting cancer progression and IL-
1β-mediated oncogenic response. Therefore, it will be useful to investigate the effect of
GSTO1 inhibitors on immune cells in the tumor microenvironment in terms of cancer cell
invasion, migration and expression of pro-inflammatory signaling. Figure 11.1
summarizes some of the key findings and a proposed mechanism for GSTO1 inhibitors in
IL-1β dependent cancers.
178
Figure 11.1 First-in-class GSTO1 inhibitors target key hallmarks of cancer.
We also identified novel inhibitors targeting GRP78, a protein involved in the
unfolded protein response. In the recent years, GRP78 as well as other HSP70 chaperones
have emerged as attractive therapeutic targets. However, there is a paucity of GRP78
inhibitors. We undertook a drug discovery effort to design novel small-molecule
inhibitors of GRP78 and have identified three novel structural inhibitor classes. However,
identification of GRP78 ATPase inhibitors has been quite challenging as previously
noted (Massey, 2010). While we have demonstrated the effect of these small molecules
on GRP78 and HSP70 through various direct and indirect methods, discerning their
binding mechanism has been challenging. We speculate that these compounds perhaps
Tumor associated
macrophages
Infiltrating monocytes
Tumor cells
IL-1β β " "
IL-1β β ( (tumor cell derived) " "
Cancer cell
proliferation " "
Chemoresistance " "
GSTO1
Pro-IL-1 β β" "
IL-1 β β" "
GSTO1 inhibitor " "
Cellular stress
homeostasis " "
Chemotactic
signals " "
Tumor invasion
and metastasis " "
( (immune cell derived) " "
179
act via an allosteric mechanism. In the recent years, there has been an increasing focus on
allostery and interdomain communication among HSP70 family chaperones (Zuiderweg
et al., 2013). Several allosteric inhibitors have been reported, indicating new druggable
binding sites in the protein (Kang et al., 2014; Li et al., 2013; Miyata et al., 2013; Rodina
et al., 2013; Rousaki et al., 2011; Taldone et al., 2014).
Through our studies, we also uncovered an atypical mechanism of cell death that
appears to be caspase-independent, involving autophagy, necroptosis and cytoplasmic
vacuolization. This form of cell death is becoming increasingly common among
compounds targeting GRP78 and HSP70 or causing ER stress (Chiang et al., 1989;
Gandin et al., 2014; Granato et al., 2014; Kosakowska-Cholody et al., 2014; Leu et al.,
2009; Wang et al., 2013). It will be of great value to explore the underlying mechanism.
Cancer cells are often resistant to apoptosis and do not respond to chemotherapeutic
drugs, where cell death is primarily elicit through apoptosis. In such cases, exploiting
non-apoptotic cell death pathways using such compounds could prove useful.
180
BIBLIOGRAPHY
Achan, J., Talisuna, A.O., Erhart, A., Yeka, A., Tibenderana, J.K., Baliraine, F.N.,
Rosenthal, P.J., and D'Alessandro, U. (2011). Quinine, an old anti-malarial drug in a
modern world: role in the treatment of malaria. Malar. J. 10.
Adam, G.C., Sorensen, E.J., and Cravatt, B.F. (2002). Proteomic profiling of
mechanistically distinct enzyme classes using a common chemotype. Nat. Biotechnol. 20,
805-809.
Adlard, P.A., Cherny, R.A., Finkelstein, D.I., Gautier, E., Robb, E., Cortes, M., Volitakis,
I., Liu, X., Smith, J.P., Perez, K., et al. (2008). Rapid restoration of cognition in
Alzheimer's transgenic mice with 8-hydroxy quinoline analogs is associated with
decreased interstitial Abeta. Neuron 59, 43-55.
Aherne, W., Maloney, A., Prodromou, C., Rowlands, M.G., Hardcastle, A., Boxall, K.,
Clarke, P., Walton, M.I., Pearl, L., and Workman, P. (2003). Assays for HSP90 and
inhibitors. Methods Mol. Med. 85, 149-161.
Ali, M.M., Bagratuni, T., Davenport, E.L., Nowak, P.R., Silva-Santisteban, M.C.,
Hardcastle, A., McAndrews, C., Rowlands, M.G., Morgan, G.J., Aherne, W., et al.
(2011). Structure of the Ire1 autophosphorylation complex and implications for the
unfolded protein response. EMBO J. 30, 894-905.
Allen, M., Zou, F., Chai, H.S., Younkin, C.S., Miles, R., Nair, A.A., Crook, J.E.,
Pankratz, V.S., Carrasquillo, M.M., Rowley, C.N., et al. (2012). Glutathione S-
transferase omega genes in Alzheimer and Parkinson disease risk, age-at-diagnosis and
brain gene expression: an association study with mechanistic implications. Mol.
Neurodegener. 7.
Atkins, C., Liu, Q., Minthorn, E., Zhang, S.Y., Figueroa, D.J., Moss, K., Stanley, T.B.,
Sanders, B., Goetz, A., Gaul, N., et al. (2013). Characterization of a novel PERK kinase
inhibitor with antitumor and antiangiogenic activity. Cancer Res. 73, 1993-2002.
Axten, J.M., Romeril, S.P., Shu, A., Ralph, J., Medina, J.R., Feng, Y., Li, W.H., Grant,
S.W., Heerding, D.A., Minthorn, E., et al. (2013). Discovery of GSK2656157: An
optimized PERK inhibitor selected for preclinical development. ACS Med. Chem. Lett.
4, 964-968.
181
Bachovchin, D.A., Brown, S.J., Rosen, H., and Cravatt, B.F. (2009). Identification of
selective inhibitors of uncharacterized enzymes by high-throughput screening with
fluorescent activity-based probes. Nat. Biotechnol. 27, 387-394.
Backer, J.M., Krivoshein, A.V., Hamby, C.V., Pizzonia, J., Gilbert, K.S., Ray, Y.S.,
Brand, H., Paton, A.W., Paton, J.C., and Backer, M.V. (2009). Chaperone-targeting
cytotoxin and endoplasmic reticulum stress-inducing drug synergize to kill cancer cells.
Neoplasia 11, 1165-1173.
Blond-Elguindi, S., Fourie, A.M., Sambrook, J.F., and Gething, M.J. (1993). Peptide-
dependent stimulation of the ATPase activity of the molecular chaperone BiP is the result
of conversion of oligomers to active monomers. J. Biol. Chem. 268, 12730-12735.
Board, P.G. (2011). The omega-class glutathione transferases: structure, function, and
genetics. Drug Metab. Rev. 43, 226-235.
Board, P.G., and Anders, M.W. (2007). Glutathione transferase omega 1 catalyzes the
reduction of S-(phenacyl)glutathiones to acetophenones. Chem. Res. Toxicol. 20, 149-
154.
Board, P.G., Coggan, M., Cappello, J., Zhou, H., Oakley, A.J., and Anders, M.W. (2008).
S-(4-Nitrophenacyl)glutathione is a specific substrate for glutathione transferase omega
1-1. Anal. Biochem. 374, 25-30.
Board, P.G., Coggan, M., Chelvanayagam, G., Easteal, S., Jermiin, L.S., Schulte, G.K.,
Danley, D.E., Hoth, L.R., Griffor, M.C., Kamath, A.V., et al. (2000). Identification,
characterization, and crystal structure of the Omega class glutathione transferases. J. Biol.
Chem. 275, 24798-24806.
Board, P.G., and Menon, D. (2013). Glutathione transferases, regulators of cellular
metabolism and physiology. Biochim. Biophys. Acta 1830, 3267-3288.
Boyce, M., Bryant, K.F., Jousse, C., Long, K., Harding, H.P., Scheuner, D., Kaufman,
R.J., Ma, D., Coen, D.M., Ron, D., et al. (2005). A selective inhibitor of eIF2alpha
dephosphorylation protects cells from ER stress. Science 307, 935-939.
Brock, J., Board, P.G., and Oakley, A.J. (2013). Structural insights into omega-class
glutathione transferases: a snapshot of enzyme reduction and identification of a non-
catalytic ligandin site. PLoS One 8, e60324.
182
Broker, L.E., Kruyt, F.A., and Giaccone, G. (2005). Cell death independent of caspases: a
review. Clin. Cancer Res. 11, 3155-3162.
Burmeister, C., Luersen, K., Heinick, A., Hussein, A., Domagalski, M., Walter, R.D., and
Liebau, E. (2008). Oxidative stress in Caenorhabditis elegans: protective effects of the
Omega class glutathione transferase (GSTO-1). FASEB J. 22, 343-354.
Capurso, C., Panza, F., Seripa, D., Frisardi, V., Imbimbo, B.P., Verdile, G., Vendemiale,
G., Pilotto, A., and Solfrizzi, V. (2010). Polymorphisms in glutathione S-transferase
omega-1 gene and increased risk of sporadic Alzheimer disease. Rejuvenation Res 13,
645-652.
Chen, W.T., Zhu, G., Pfaffenbach, K., Kanel, G., Stiles, B., and Lee, A.S. (2013). GRP78
as a regulator of liver steatosis and cancer progression mediated by loss of the tumor
suppressor PTEN. Oncogene.
Chiang, H.L., Terlecky, S.R., Plant, C.P., and Dice, J.F. (1989). A role for a 70-kilodalton
heat shock protein in lysosomal degradation of intracellular proteins. Science 246, 382-
385.
Chockalingam, P.S., Sun, W., Rivera-Bermudez, M.A., Zeng, W., Dufield, D.R.,
Larsson, S., Lohmander, L.S., Flannery, C.R., Glasson, S.S., Georgiadis, K.E., et al.
(2011). Elevated aggrecanase activity in a rat model of joint injury is attenuated by an
aggrecanase specific inhibitor. Osteoarthritis Cartilage 19, 315-323.
Coll, R.C., and O'Neill, L.A. (2011). The cytokine release inhibitory drug CRID3 targets
ASC oligomerisation in the NLRP3 and AIM2 inflammasomes. PLoS One 6, e29539.
Cross, B.C., Bond, P.J., Sadowski, P.G., Jha, B.K., Zak, J., Goodman, J.M., Silverman,
R.H., Neubert, T.A., Baxendale, I.R., Ron, D., et al. (2012). The molecular basis for
selective inhibition of unconventional mRNA splicing by an IRE1-binding small
molecule. Proc. Natl. Acad. Sci. U. S. A. 109, E869-878.
Daugaard, M., Kirkegaard-Sorensen, T., Ostenfeld, M.S., Aaboe, M., Hoyer-Hansen, M.,
Orntoft, T.F., Rohde, M., and Jaattela, M. (2007). Lens epithelium-derived growth factor
is an Hsp70-2 regulated guardian of lysosomal stability in human cancer. Cancer Res. 67,
2559-2567.
Dayam, R., Al-Mawsawi, L.Q., Zawahir, Z., Witvrouw, M., Debyser, Z., and Neamati, N.
(2008). Quinolone 3-carboxylic acid pharmacophore: design of second generation HIV-1
integrase inhibitors. J. Med. Chem. 51, 1136-1144.
183
Deans, D.A., Wigmore, S.J., Gilmour, H., Paterson-Brown, S., Ross, J.A., and Fearon,
K.C. (2006). Elevated tumour interleukin-1beta is associated with systemic inflammation:
A marker of reduced survival in gastro-oesophageal cancer. Br. J. Cancer 95, 1568-1575.
Dejeans, N., Manie, S., Hetz, C., Bard, F., Hupp, T., Agostinis, P., Samali, A., and
Chevet, E. (2014). Addicted to secrete - novel concepts and targets in cancer therapy.
Trends Mol. Med. 20, 242-250.
Deng, W.G., Ruan, K.H., Du, M., Saunders, M.A., and Wu, K.K. (2001). Aspirin and
salicylate bind to immunoglobulin heavy chain binding protein (BiP) and inhibit its
ATPase activity in human fibroblasts. FASEB J. 15, 2463-2470.
Ding, W.X., Ni, H.M., Gao, W., Hou, Y.F., Melan, M.A., Chen, X., Stolz, D.B., Shao,
Z.M., and Yin, X.M. (2007). Differential effects of endoplasmic reticulum stress-induced
autophagy on cell survival. J. Biol. Chem. 282, 4702-4710.
Dixon, S.J., Lemberg, K.M., Lamprecht, M.R., Skouta, R., Zaitsev, E.M., Gleason, C.E.,
Patel, D.N., Bauer, A.J., Cantley, A.M., Yang, W.S., et al. (2012). Ferroptosis: an iron-
dependent form of nonapoptotic cell death. Cell 149, 1060-1072.
Djukic, T.I., Savic-Radojevic, A.R., Pekmezovic, T.D., Matic, M.G., Pljesa-Ercegovac,
M.S., Coric, V.M., Radic, T.M., Suvakov, S.R., Krivic, B.N., Dragicevic, D.P., et al.
(2013). Glutathione S-transferase T1, O1 and O2 polymorphisms are associated with
survival in muscle invasive bladder cancer patients. PLoS One 8, e74724.
Dong, D., Ko, B., Baumeister, P., Swenson, S., Costa, F., Markland, F., Stiles, C.,
Patterson, J.B., Bates, S.E., and Lee, A.S. (2005). Vascular targeting and
antiangiogenesis agents induce drug resistance effector GRP78 within the tumor
microenvironment. Cancer Res. 65, 5785-5791.
Dong, D., Ni, M., Li, J., Xiong, S., Ye, W., Virrey, J.J., Mao, C., Ye, R., Wang, M., Pen,
L., et al. (2008). Critical role of the stress chaperone GRP78/BiP in tumor proliferation,
survival, and tumor angiogenesis in transgene-induced mammary tumor development.
Cancer Res. 68, 498-505.
Dong, D., Stapleton, C., Luo, B., Xiong, S., Ye, W., Zhang, Y., Jhaveri, N., Zhu, G., Ye,
R., Liu, Z., et al. (2011). A critical role for GRP78/BiP in the tumor microenvironment
for neovascularization during tumor growth and metastasis. Cancer Res. 71, 2848-2857.
184
Dulhunty, A., Gage, P., Curtis, S., Chelvanayagam, G., and Board, P. (2001). The
glutathione transferase structural family includes a nuclear chloride channel and a
ryanodine receptor calcium release channel modulator. J. Biol. Chem. 276, 3319-3323.
Elaraj, D.M., Weinreich, D.M., Varghese, S., Puhlmann, M., Hewitt, S.M., Carroll, N.M.,
Feldman, E.D., Turner, E.M., and Alexander, H.R. (2006). The role of interleukin 1 in
growth and metastasis of human cancer xenografts. Clin. Cancer Res. 12, 1088-1096.
Engel, R.H., and Evens, A.M. (2006). Oxidative stress and apoptosis: a new treatment
paradigm in cancer. Front. Biosci. 11, 300-312.
Ermakova, S.P., Kang, B.S., Choi, B.Y., Choi, H.S., Schuster, T.F., Ma, W.Y., Bode,
A.M., and Dong, Z. (2006). (-)-Epigallocatechin gallate overcomes resistance to
etoposide-induced cell death by targeting the molecular chaperone glucose-regulated
protein 78. Cancer Res. 66, 9260-9269.
Eskew, J.D., Sadikot, T., Morales, P., Duren, A., Dunwiddie, I., Swink, M., Zhang, X.,
Hembruff, S., Donnelly, A., Rajewski, R.A., et al. (2011). Development and
characterization of a novel C-terminal inhibitor of Hsp90 in androgen dependent and
independent prostate cancer cells. BMC Cancer 11.
Feldman, D., and Koong, A. (2007). Irestatin, a potent inhibitor of IRE1 {alpha} and the
unfolded protein response, is a hypoxia-selective cytotoxin and impairs tumor growth. J.
Clin. Oncol. 25, 3514.
Fernandes, A.P., and Holmgren, A. (2004). Glutaredoxins: glutathione-dependent redox
enzymes with functions far beyond a simple thioredoxin backup system. Antioxid Redox
Signal 6, 63-74.
Fernandez, P.M., Tabbara, S.O., Jacobs, L.K., Manning, F.C., Tsangaris, T.N., Schwartz,
A.M., Kennedy, K.A., and Patierno, S.R. (2000). Overexpression of the glucose-
regulated stress gene GRP78 in malignant but not benign human breast lesions. Breast
Cancer Res. Treat. 59, 15-26.
Fewell, S.W., Day, B.W., and Brodsky, J.L. (2001). Identification of an inhibitor of
hsc70-mediated protein translocation and ATP hydrolysis. J. Biol. Chem. 276, 910-914.
Fiaschi, T., and Chiarugi, P. (2012). Oxidative stress, tumor microenvironment, and
metabolic reprogramming: a diabolic liaison. Int. J. Cell Biol. 2012.
185
Flynn, G.C., Chappell, T.G., and Rothman, J.E. (1989). Peptide binding and release by
proteins implicated as catalysts of protein assembly. Science 245, 385-390.
Franchi, L., Eigenbrod, T., Munoz-Planillo, R., and Nunez, G. (2009). The
inflammasome: a caspase-1-activation platform that regulates immune responses and
disease pathogenesis. Nat. Immunol. 10, 241-247.
Fu, Y., Li, J., and Lee, A.S. (2007). GRP78/BiP inhibits endoplasmic reticulum BIK and
protects human breast cancer cells against estrogen starvation-induced apoptosis. Cancer
Res. 67, 3734-3740.
Fu, Y., Wey, S., Wang, M., Ye, R., Liao, C.P., Roy-Burman, P., and Lee, A.S. (2008).
Pten null prostate tumorigenesis and AKT activation are blocked by targeted knockout of
ER chaperone GRP78/BiP in prostate epithelium. Proc. Natl. Acad. Sci. U. S. A. 105,
19444-19449.
Gandin, V., Tisato, F., Dolmella, A., Pellei, M., Santini, C., Giorgetti, M., Marzano, C.,
and Porchia, M. (2014). In vitro and in vivo anticancer activity of copper(I) complexes
with homoscorpionate tridentate tris(pyrazolyl)borate and auxiliary monodentate
phosphine ligands. J. Med. Chem., 4745-4760.
Gaut, J.R., and Hendershot, L.M. (1993). Mutations within the nucleotide binding site of
immunoglobulin-binding protein inhibit ATPase activity and interfere with release of
immunoglobulin heavy chain. J. Biol. Chem. 268, 7248-7255.
Gazit, G., Lu, J., and Lee, A.S. (1999). De-regulation of GRP stress protein expression in
human breast cancer cell lines. Breast Cancer Res. Treat. 54, 135-146.
Giri, U., Terry, N.H., Kala, S.V., Lieberman, M.W., and Story, M.D. (2005). Elimination
of the differential chemoresistance between the murine B-cell lymphoma LY-ar and LY-
as cell lines after arsenic (As2O3) exposure via the overexpression of gsto1 (p28). Cancer
Chemother. Pharmacol. 55, 511-521.
Gonzalez-Gronow, M., Selim, M.A., Papalas, J., and Pizzo, S.V. (2009). GRP78: a
multifunctional receptor on the cell surface. Antioxid Redox Signal 11, 2299-2306.
Granato, M., Lacconi, V., Peddis, M., Lotti, L.V., Di Renzo, L., Gonnella, R., Santarelli,
R., Trivedi, P., Frati, L., D'Orazi, G., et al. (2014). HSP70 inhibition by 2-
phenylethynesulfonamide induces lysosomal cathepsin D release and immunogenic cell
death in primary effusion lymphoma. Cell Death Dis. 5, e1254.
186
Gray, M.J., Mhawech-Fauceglia, P., Yoo, E., Yang, W., Wu, E., Lee, A.S., and Lin, Y.G.
(2013). AKT inhibition mitigates GRP78 (glucose-regulated protein) expression and
contribution to chemoresistance in endometrial cancers. Int. J. Cancer 133, 21-30.
Guzman, M.L., Lowery, M.A., Taldone, T., Koren Iii, J., Gomes, E.D., and Chiosis, G.
(2013). Molecular chaperones and how addiction matters in cancer therapy. In Cell Death
Signaling in Cancer Biology and Treatment (Springer), pp. 181-203
Gyrd-Hansen, M., Nylandsted, J., and Jaattela, M. (2004). Heat shock protein 70
promotes cancer cell viability by safeguarding lysosomal integrity. Cell Cycle 3, 1484-
1485.
Hanahan, D., and Weinberg, R.A. (2000). The hallmarks of cancer. Cell 100, 57-70.
Hanahan, D., and Weinberg, R.A. (2011). Hallmarks of cancer: the next generation. Cell
144, 646-674.
Hayes, J.D., Flanagan, J.U., and Jowsey, I.R. (2005). Glutathione transferases. Annu.
Rev. Pharmacol. Toxicol. 45, 51-88.
Hayes, J.D., and McLellan, L.I. (1999). Glutathione and glutathione-dependent enzymes
represent a co-ordinately regulated defence against oxidative stress. Free Radic. Res. 31,
273-300.
Hendershot, L.M., Wei, J.Y., Gaut, J.R., Lawson, B., Freiden, P.J., and Murti, K.G.
(1995). In vivo expression of mammalian BiP ATPase mutants causes disruption of the
endoplasmic reticulum. Mol. Biol. Cell 6, 283-296.
Herrero, E., and de la Torre-Ruiz, M.A. (2007). Monothiol glutaredoxins: a common
domain for multiple functions. Cell. Mol. Life Sci. 64, 1518-1530.
Hetz, C., Chevet, E., and Harding, H.P. (2013). Targeting the unfolded protein response
in disease. Nat Rev Drug Discov 12, 703-719.
Hoffstrom, B.G., Kaplan, A., Letso, R., Schmid, R.S., Turmel, G.J., Lo, D.C., and
Stockwell, B.R. (2010). Inhibitors of protein disulfide isomerase suppress apoptosis
induced by misfolded proteins. Nat. Chem. Biol. 6, 900-906.
Honda, T., Inagawa, H., and Yamamoto, I. (2011). Differential expression of mRNA in
human monocytes following interaction with human colon cancer cells. Anticancer Res.
31, 2493-2497.
187
Ishikawa, T., Casini, A.F., and Nishikimi, M. (1998). Molecular cloning and functional
expression of rat liver glutathione-dependent dehydroascorbate reductase. J. Biol. Chem.
273, 28708-28712.
Jamora, C., Dennert, G., and Lee, A.S. (1996). Inhibition of tumor progression by
suppression of stress protein GRP78/BiP induction in fibrosarcoma B/C10ME. Proc.
Natl. Acad. Sci. U. S. A. 93, 7690-7694.
Jiang, C.C., Mao, Z.G., Avery-Kiejda, K.A., Wade, M., Hersey, P., and Zhang, X.D.
(2009). Glucose-regulated protein 78 antagonizes cisplatin and adriamycin in human
melanoma cells. Carcinogenesis 30, 197-204.
Johnson, R.L., Huang, W., Jadhav, A., Austin, C.P., Inglese, J., and Martinez, E.D.
(2008). A quantitative high-throughput screen identifies potential epigenetic modulators
of gene expression. Anal. Biochem. 375, 237-248.
Kaler, P., Augenlicht, L., and Klampfer, L. (2009a). Macrophage-derived IL-1beta
stimulates Wnt signaling and growth of colon cancer cells: a crosstalk interrupted by
vitamin D3. Oncogene 28, 3892-3902.
Kaler, P., Galea, V., Augenlicht, L., and Klampfer, L. (2010). Tumor associated
macrophages protect colon cancer cells from TRAIL-induced apoptosis through IL-
1beta-dependent stabilization of Snail in tumor cells. PLoS One 5, e11700.
Kaler, P., Godasi, B.N., Augenlicht, L., and Klampfer, L. (2009b). The NF-kappaB/AKT-
dependent induction of Wnt signaling in colon cancer cells by macrophages and IL-1beta.
Cancer Microenviron. 2, 69-80.
Kampinga, H.H., and Craig, E.A. (2010). The HSP70 chaperone machinery: J proteins as
drivers of functional specificity. Nat. Rev. Mol. Cell Biol. 11, 579-592.
Kang, Y., Taldone, T., Patel, H.J., Patel, P.D., Rodina, A., Gozman, A., Maharaj, R.,
Clement, C.C., Patel, M.R., Brodsky, J.L., et al. (2014). Heat shock protein 70 inhibitors.
1. 2,5'-thiodipyrimidine and 5-(phenylthio)pyrimidine acrylamides as irreversible binders
to an allosteric site on heat shock protein 70. J. Med. Chem. 57, 1188-1207.
Kim, K., Kim, S.H., Kim, J., Kim, H., and Yim, J. (2012). Glutathione S-transferase
Omega 1 activity is sufficient to suppress neurodegeneration in a drosophila model of
Parkinson's disease. J. Biol. Chem. 287, 6628-6641.
188
Kirkegaard, T., Roth, A.G., Petersen, N.H., Mahalka, A.K., Olsen, O.D., Moilanen, I.,
Zylicz, A., Knudsen, J., Sandhoff, K., Arenz, C., et al. (2010). Hsp70 stabilizes
lysosomes and reverts Niemann-Pick disease-associated lysosomal pathology. Nature
463, 549-553.
Klionsky, D.J., Abeliovich, H., Agostinis, P., Agrawal, D.K., Aliev, G., Askew, D.S.,
Baba, M., Baehrecke, E.H., Bahr, B.A., Ballabio, A., et al. (2008). Guidelines for the use
and interpretation of assays for monitoring autophagy in higher eukaryotes. Autophagy 4,
151-175.
Kodym, R., Calkins, P., and Story, M. (1999). The cloning and characterization of a new
stress response protein. A mammalian member of a family of theta class glutathione s-
transferase-like proteins. J. Biol. Chem. 274, 5131-5137.
Kodym, R., Calkins, P.R., and Story, M.D. (2001). Anthracycline-induced erythroid
differentiation of K562 cells is inhibited by p28, a novel mammalian glutathione-binding
stress protein. Leuk. Res. 25, 151-156.
Kolsch, H., Linnebank, M., Lutjohann, D., Jessen, F., Wullner, U., Harbrecht, U., Thelen,
K.M., Kreis, M., Hentschel, F., Schulz, A., et al. (2004). Polymorphisms in glutathione
S-transferase omega-1 and AD, vascular dementia, and stroke. Neurology 63, 2255-2260.
Koomagi, R., Mattern, J., and Volm, M. (1999). Glucose-related protein (GRP78) and its
relationship to the drug-resistance proteins P170, GST-pi, LRP56 and angiogenesis in
non-small cell lung carcinomas. Anticancer Res. 19, 4333-4336.
Koren, J., 3rd, Jinwal, U.K., Jin, Y., O'Leary, J., Jones, J.R., Johnson, A.G., Blair, L.J.,
Abisambra, J.F., Chang, L., Miyata, Y., et al. (2010). Facilitating Akt clearance via
manipulation of Hsp70 activity and levels. J. Biol. Chem. 285, 2498-2505.
Kosakowska-Cholody, T., Lin, J., Srideshikan, S.M., Scheffer, L., Tarasova, N.I., and
Acharya, J.K. (2014). HKH40A downregulates GRP78/BiP expression in cancer cells.
Cell Death Dis. 5, e1240.
Kozutsumi, Y., Segal, M., Normington, K., Gething, M.J., and Sambrook, J. (1988). The
presence of malfolded proteins in the endoplasmic reticulum signals the induction of
glucose-regulated proteins. Nature 332, 462-464.
Laborde, E. (2010). Glutathione transferases as mediators of signaling pathways involved
in cell proliferation and cell death. Cell Death Differ. 17, 1373-1380.
189
Lai, E., Teodoro, T., and Volchuk, A. (2007). Endoplasmic reticulum stress: signaling the
unfolded protein response. Physiology (Bethesda) 22, 193-201.
Laliberte, R.E., Perregaux, D.G., Hoth, L.R., Rosner, P.J., Jordan, C.K., Peese, K.M.,
Eggler, J.F., Dombroski, M.A., Geoghegan, K.F., and Gabel, C.A. (2003). Glutathione s-
transferase omega 1-1 is a target of cytokine release inhibitory drugs and may be
responsible for their effect on interleukin-1beta posttranslational processing. J. Biol.
Chem. 278, 16567-16578.
Latz, E., Xiao, T.S., and Stutz, A. (2013). Activation and regulation of the
inflammasomes. Nat. Rev. Immunol. 13, 397-411.
Lee, A.S. (1981). The accumulation of three specific proteins related to glucose-regulated
proteins in a temperature-sensitive hamster mutant cell line K12. J. Cell. Physiol. 106,
119-125.
Lee, A.S. (2001). The glucose-regulated proteins: stress induction and clinical
applications. Trends Biochem. Sci. 26, 504-510.
Lee, A.S. (2007). GRP78 induction in cancer: therapeutic and prognostic implications.
Cancer Res. 67, 3496-3499.
Lee, A.S. (2014). Glucose-regulated proteins in cancer: molecular mechanisms and
therapeutic potential. Nat. Rev. Cancer 14, 263-276.
Lee, C.U., and Grossmann, T.N. (2012). Reversible covalent inhibition of a protein
target. Angew. Chem. Int. Ed. Engl. 51, 8699-8700.
Lee, E., Nichols, P., Spicer, D., Groshen, S., Yu, M.C., and Lee, A.S. (2006). GRP78 as a
novel predictor of responsiveness to chemotherapy in breast cancer. Cancer Res. 66,
7849-7853.
Leu, J.I., Pimkina, J., Frank, A., Murphy, M.E., and George, D.L. (2009). A small
molecule inhibitor of inducible heat shock protein 70. Mol. Cell 36, 15-27.
Li, J., and Lee, A.S. (2006). Stress induction of GRP78/BiP and its role in cancer. Curr.
Mol. Med. 6, 45-54.
Li, J., Ni, M., Lee, B., Barron, E., Hinton, D.R., and Lee, A.S. (2008). The unfolded
protein response regulator GRP78/BiP is required for endoplasmic reticulum integrity
and stress-induced autophagy in mammalian cells. Cell Death Differ. 15, 1460-1471.
190
Li, X., Srinivasan, S.R., Connarn, J., Ahmad, A., Young, Z.T., Kabza, A.M., Zuiderweg,
E.R., Sun, D., and Gestwicki, J.E. (2013). Analogs of the Allosteric Heat Shock Protein
70 (Hsp70) Inhibitor, MKT-077, as Anti-Cancer Agents. ACS Med. Chem. Lett. 4, 1042-
1047.
Li, X., Zhang, K., and Li, Z. (2011). Unfolded protein response in cancer: the physician's
perspective. J. Hematol. Oncol. 4, 8.
Li, Y.J., Oliveira, S.A., Xu, P., Martin, E.R., Stenger, J.E., Scherzer, C.R., Hauser, M.A.,
Scott, W.K., Small, G.W., Nance, M.A., et al. (2003). Glutathione S-transferase omega-1
modifies age-at-onset of Alzheimer disease and Parkinson disease. Hum. Mol. Genet. 12,
3259-3267.
Li, Y.J., Scott, W.K., Zhang, L., Lin, P.I., Oliveira, S.A., Skelly, T., Doraiswamy, M.P.,
Welsh-Bohmer, K.A., Martin, E.R., Haines, J.L., et al. (2006). Revealing the role of
glutathione S-transferase omega in age-at-onset of Alzheimer and Parkinson diseases.
Neurobiol. Aging 27, 1087-1093.
Lomenick, B., Hao, R., Jonai, N., Chin, R.M., Aghajan, M., Warburton, S., Wang, J.,
Wu, R.P., Gomez, F., Loo, J.A., et al. (2009). Target identification using drug affinity
responsive target stability (DARTS). Proc. Natl. Acad. Sci. U. S. A. 106, 21984-21989.
Lomenick, B., Olsen, R.W., and Huang, J. (2011). Identification of direct protein targets
of small molecules. ACS Chem. Biol. 6, 34-46.
Luo, J., Solimini, N.L., and Elledge, S.J. (2009). Principles of cancer therapy: oncogene
and non-oncogene addiction. Cell 136, 823-837.
Luo, S., Mao, C., Lee, B., and Lee, A.S. (2006). GRP78/BiP is required for cell
proliferation and protecting the inner cell mass from apoptosis during early mouse
embryonic development. Mol. Cell. Biol. 26, 5688-5697.
Luo, T., Wang, J., Yin, Y., Hua, H., Jing, J., Sun, X., Li, M., Zhang, Y., and Jiang, Y.
(2010). (-)-Epigallocatechin gallate sensitizes breast cancer cells to paclitaxel in a murine
model of breast carcinoma. Breast Cancer Res. 12, R8.
Luvsandagva, B., Nakamura, K., Kitahara, Y., Aoki, H., Murata, T., Ikeda, S., and
Minegishi, T. (2012). GRP78 induced by estrogen plays a role in the chemosensitivity of
endometrial cancer. Gynecol. Oncol. 126, 132-139.
191
Macias, A.T., Williamson, D.S., Allen, N., Borgognoni, J., Clay, A., Daniels, Z.,
Dokurno, P., Drysdale, M.J., Francis, G.L., Graham, C.J., et al. (2011). Adenosine-
derived inhibitors of 78 kDa glucose regulated protein (Grp78) ATPase: insights into
isoform selectivity. J. Med. Chem. 54, 4034-4041.
Mahajan, S., and Atkins, W.M. (2005). The chemistry and biology of inhibitors and pro-
drugs targeted to glutathione S-transferases. Cell. Mol. Life Sci. 62, 1221-1233.
Mariathasan, S., and Monack, D.M. (2007). Inflammasome adaptors and sensors:
intracellular regulators of infection and inflammation. Nat. Rev. Immunol. 7, 31-40.
Martin, S., Lamb, H.K., Brady, C., Lefkove, B., Bonner, M.Y., Thompson, P., Lovat,
P.E., Arbiser, J.L., Hawkins, A.R., and Redfern, C.P. (2013). Inducing apoptosis of
cancer cells using small-molecule plant compounds that bind to GRP78. Br. J. Cancer
109, 433-443.
Martindale, J.L., and Holbrook, N.J. (2002). Cellular response to oxidative stress:
signaling for suicide and survival. J. Cell. Physiol. 192, 1-15.
Massey, A.J. (2010). ATPases as drug targets: insights from heat shock proteins 70 and
90. J. Med. Chem. 53, 7280-7286.
Massey, A.J., Williamson, D.S., Browne, H., Murray, J.B., Dokurno, P., Shaw, T.,
Macias, A.T., Daniels, Z., Geoffroy, S., Dopson, M., et al. (2010). A novel, small
molecule inhibitor of Hsc70/Hsp70 potentiates Hsp90 inhibitor induced apoptosis in
HCT116 colon carcinoma cells. Cancer Chemother. Pharmacol. 66, 535-545.
Matsui, H., Lin, L.R., Singh, D.P., Shinohara, T., and Reddy, V.N. (2001). Lens
epithelium-derived growth factor: increased survival and decreased DNA breakage of
human RPE cells induced by oxidative stress. Invest. Ophthalmol. Vis. Sci. 42, 2935-
2941.
Matsuo, J., Tsukumo, Y., Sakurai, J., Tsukahara, S., Park, H.R., Shin-ya, K., Watanabe,
T., Tsuruo, T., and Tomida, A. (2009). Preventing the unfolded protein response via
aberrant activation of 4E-binding protein 1 by versipelostatin. Cancer Sci. 100, 327-333.
McIlwain, C.C., Townsend, D.M., and Tew, K.D. (2006). Glutathione S-transferase
polymorphisms: cancer incidence and therapy. Oncogene 25, 1639-1648.
Menon, D., and Board, P.G. (2013). A role for glutathione transferase Omega 1 (GSTO1-
1) in the glutathionylation cycle. J. Biol. Chem. 288, 25769-25779.
192
Mimura, N., Fulciniti, M., Gorgun, G., Tai, Y.T., Cirstea, D., Santo, L., Hu, Y., Fabre, C.,
Minami, J., Ohguchi, H., et al. (2012). Blockade of XBP1 splicing by inhibition of
IRE1alpha is a promising therapeutic option in multiple myeloma. Blood 119, 5772-
5781.
Mintz, P.J., Kim, J., Do, K.A., Wang, X., Zinner, R.G., Cristofanilli, M., Arap, M.A.,
Hong, W.K., Troncoso, P., Logothetis, C.J., et al. (2003). Fingerprinting the circulating
repertoire of antibodies from cancer patients. Nat. Biotechnol. 21, 57-63.
Misra, U.K., Deedwania, R., and Pizzo, S.V. (2005). Binding of activated alpha2-
macroglobulin to its cell surface receptor GRP78 in 1-LN prostate cancer cells regulates
PAK-2-dependent activation of LIMK. J. Biol. Chem. 280, 26278-26286.
Misra, U.K., Mowery, Y., Kaczowka, S., and Pizzo, S.V. (2009). Ligation of cancer cell
surface GRP78 with antibodies directed against its COOH-terminal domain up-regulates
p53 activity and promotes apoptosis. Mol. Cancer Ther. 8, 1350-1362.
Misra, U.K., and Pizzo, S.V. (2010). Ligation of cell surface GRP78 with antibody
directed against the COOH-terminal domain of GRP78 suppresses Ras/MAPK and PI 3-
kinase/AKT signaling while promoting caspase activation in human prostate cancer cells.
Cancer Biol. Ther. 9, 142-152.
Miyata, Y., Li, X., Lee, H.F., Jinwal, U.K., Srinivasan, S.R., Seguin, S.P., Young, Z.T.,
Brodsky, J.L., Dickey, C.A., Sun, D., et al. (2013). Synthesis and initial evaluation of
YM-08, a blood-brain barrier permeable derivative of the heat shock protein 70 (Hsp70)
inhibitor MKT-077, which reduces tau levels. ACS Chem. Neurosci. 4, 930-939.
Muhlenhoff, U., Molik, S., Godoy, J.R., Uzarska, M.A., Richter, N., Seubert, A., Zhang,
Y., Stubbe, J., Pierrel, F., Herrero, E., et al. (2010). Cytosolic monothiol glutaredoxins
function in intracellular iron sensing and trafficking via their bound iron-sulfur cluster.
Cell Metab 12, 373-385.
Murrow, L., and Debnath, J. (2013). Autophagy as a stress-response and quality-control
mechanism: implications for cell injury and human disease. Annu. Rev. Pathol. 8, 105-
137.
Ni, M., Zhang, Y., and Lee, A.S. (2011). Beyond the endoplasmic reticulum: atypical
GRP78 in cell viability, signalling and therapeutic targeting. Biochem. J. 434, 181-188.
193
Nishimura, M., Kuno, S., Kaji, R., Yasuno, K., and Kawakami, H. (2005). Glutathione-S-
transferase-1 and interleukin-1beta gene polymorphisms in Japanese patients with
Parkinson's disease. Mov. Disord. 20, 901-902.
Nishimura, M., Sakamoto, T., Kaji, R., and Kawakami, H. (2004). Influence of
polymorphisms in the genes for cytokines and glutathione S-transferase omega on
sporadic Alzheimer's disease. Neurosci. Lett. 368, 140-143.
Nylandsted, J., Gyrd-Hansen, M., Danielewicz, A., Fehrenbacher, N., Lademann, U.,
Hoyer-Hansen, M., Weber, E., Multhoff, G., Rohde, M., and Jaattela, M. (2004). Heat
shock protein 70 promotes cell survival by inhibiting lysosomal membrane
permeabilization. J. Exp. Med. 200, 425-435.
Nylandsted, J., Rohde, M., Brand, K., Bastholm, L., Elling, F., and Jaattela, M. (2000).
Selective depletion of heat shock protein 70 (Hsp70) activates a tumor-specific death
program that is independent of caspases and bypasses Bcl-2. Proc. Natl. Acad. Sci. U. S.
A. 97, 7871-7876.
O.O.T.-M Clement, A. (2000). HipHop: Pharmacophores based on multiple common-
feature alignments. In Pharmacophore Perception, Development, and Use in Drug
Design, O. Güner, ed. (La Jolla, CA: International University Line), pp. 69-83.
Ogata, M., Hino, S., Saito, A., Morikawa, K., Kondo, S., Kanemoto, S., Murakami, T.,
Taniguchi, M., Tanii, I., Yoshinaga, K., et al. (2006). Autophagy is activated for cell
survival after endoplasmic reticulum stress. Mol. Cell. Biol. 26, 9220-9231.
Okada, T., Haze, K., Nadanaka, S., Yoshida, H., Seidah, N.G., Hirano, Y., Sato, R.,
Negishi, M., and Mori, K. (2003). A serine protease inhibitor prevents endoplasmic
reticulum stress-induced cleavage but not transport of the membrane-bound transcription
factor ATF6. J. Biol. Chem. 278, 31024-31032.
Ozturk, A., Desai, P.P., Minster, R.L., Dekosky, S.T., and Kamboh, M.I. (2005). Three
SNPs in the GSTO1, GSTO2 and PRSS11 genes on chromosome 10 are not associated
with age-at-onset of Alzheimer's disease. Neurobiol. Aging 26, 1161-1165.
Pace, N.J., Pimental, D.R., and Weerapana, E. (2012). An inhibitor of glutathione S-
transferase omega 1 that selectively targets apoptotic cells. Angew. Chem. Int. Ed. Engl.
51, 8365-8368.
Pajaud, J., Kumar, S., Rauch, C., Morel, F., and Aninat, C. (2012). Regulation of signal
transduction by glutathione transferases. Int. J. Hepatol. 2012.
194
Pantoliano, M.W., Petrella, E.C., Kwasnoski, J.D., Lobanov, V.S., Myslik, J., Graf, E.,
Carver, T., Asel, E., Springer, B.A., Lane, P., et al. (2001). High-density miniaturized
thermal shift assays as a general strategy for drug discovery. J Biomol Screen 6, 429-440.
Papandreou, I., Denko, N.C., Olson, M., Van Melckebeke, H., Lust, S., Tam, A., Solow-
Cordero, D.E., Bouley, D.M., Offner, F., Niwa, M., et al. (2011). Identification of an
Ire1alpha endonuclease specific inhibitor with cytotoxic activity against human multiple
myeloma. Blood 117, 1311-1314.
Park, H.R., Tomida, A., Sato, S., Tsukumo, Y., Yun, J., Yamori, T., Hayakawa, Y.,
Tsuruo, T., and Shin-ya, K. (2004). Effect on tumor cells of blocking survival response to
glucose deprivation. J. Natl. Cancer Inst. 96, 1300-1310.
Pi, L., Li, X., Song, Q., Shen, Y., Lu, X., and Di, B. (2014). Knockdown of glucose-
regulated protein 78 abrogates chemoresistance of hypopharyngeal carcinoma cells to
cisplatin induced by unfolded protein in response to severe hypoxia. Oncol. Lett. 7, 685-
692.
Piacentini, S., Polimanti, R., Squitti, R., Mariani, S., Migliore, S., Vernieri, F., Rossini,
P.M., Manfellotto, D., and Fuciarelli, M. (2012). GSTO1*E155del polymorphism
associated with increased risk for late-onset Alzheimer's disease: association hypothesis
for an uncommon genetic variant. Neurosci. Lett. 506, 203-207.
Piaggi, S., Marchi, S., Ciancia, E., Debortoli, N., Lazzarotti, A., Saviozzi, M., Raggi, C.,
Fierabracci, V., Visvikis, A., Bisgaard, H.C., et al. (2009). Nuclear translocation of
glutathione transferase omega is a progression marker in Barrett's esophagus. Oncol. Rep.
21, 283-287.
Piaggi, S., Raggi, C., Corti, A., Pitzalis, E., Mascherpa, M.C., Saviozzi, M., Pompella,
A., and Casini, A.F. (2010). Glutathione transferase omega 1-1 (GSTO1-1) plays an anti-
apoptotic role in cell resistance to cisplatin toxicity. Carcinogenesis 31, 804-811.
Pollard, J.W. (2004). Tumour-educated macrophages promote tumour progression and
metastasis. Nat. Rev. Cancer 4, 71-78.
Pongstaporn, W., Pakakasama, S., Sanguansin, S., Hongeng, S., and Petmitr, S. (2009).
Polymorphism of glutathione S-transferase Omega gene: association with risk of
childhood acute lymphoblastic leukemia. J. Cancer Res. Clin. Oncol. 135, 673-678.
195
Powers, M.V., Clarke, P.A., and Workman, P. (2008). Dual targeting of HSC70 and
HSP72 inhibits HSP90 function and induces tumor-specific apoptosis. Cancer Cell 14,
250-262.
Poyraz, O., Jeankumar, V.U., Saxena, S., Schnell, R., Haraldsson, M., Yogeeswari, P.,
Sriram, D., and Schneider, G. (2013). Structure-guided design of novel thiazolidine
inhibitors of O-acetyl serine sulfhydrylase from Mycobacterium tuberculosis. J. Med.
Chem. 56, 6457-6466.
Pyrko, P., Schonthal, A.H., Hofman, F.M., Chen, T.C., and Lee, A.S. (2007). The
unfolded protein response regulator GRP78/BiP as a novel target for increasing
chemosensitivity in malignant gliomas. Cancer Res. 67, 9809-9816.
Qin, Y., Ekmekcioglu, S., Liu, P., Duncan, L.M., Lizee, G., Poindexter, N., and Grimm,
E.A. (2011). Constitutive aberrant endogenous interleukin-1 facilitates inflammation and
growth in human melanoma. Mol. Cancer Res. 9, 1537-1550.
Raj, L., Ide, T., Gurkar, A.U., Foley, M., Schenone, M., Li, X., Tolliday, N.J., Golub,
T.R., Carr, S.A., Shamji, A.F., et al. (2011). Selective killing of cancer cells by a small
molecule targeting the stress response to ROS. Nature 475, 231-234.
Ranganathan, A.C., Zhang, L., Adam, A.P., and Aguirre-Ghiso, J.A. (2006). Functional
coupling of p38-induced up-regulation of BiP and activation of RNA-dependent protein
kinase-like endoplasmic reticulum kinase to drug resistance of dormant carcinoma cells.
Cancer Res. 66, 1702-1711.
Reddy, R.K., Mao, C., Baumeister, P., Austin, R.C., Kaufman, R.J., and Lee, A.S.
(2003). Endoplasmic reticulum chaperone protein GRP78 protects cells from apoptosis
induced by topoisomerase inhibitors: role of ATP binding site in suppression of caspase-7
activation. J. Biol. Chem. 278, 20915-20924.
Ri, M., Tashiro, E., Oikawa, D., Shinjo, S., Tokuda, M., Yokouchi, Y., Narita, T.,
Masaki, A., Ito, A., Ding, J., et al. (2012). Identification of Toyocamycin, an agent
cytotoxic for multiple myeloma cells, as a potent inhibitor of ER stress-induced XBP1
mRNA splicing. Blood Cancer J 2, e79.
Ritschel, T., Hermans, S.M., Schreurs, M., van den Heuvel, J.J., Koenderink, J.B.,
Greupink, R., and Russel, F.G. (2014). In silico identification and in vitro validation of
potential cholestatic compounds through 3D ligand-based pharmacophore modeling of
BSEP inhibitors. Chem. Res. Toxicol., 873-881.
196
Rodina, A., Patel, P.D., Kang, Y., Patel, Y., Baaklini, I., Wong, M.J., Taldone, T., Yan,
P., Yang, C., Maharaj, R., et al. (2013). Identification of an allosteric pocket on human
hsp70 reveals a mode of inhibition of this therapeutically important protein. Chem. Biol.
20, 1469-1480.
Rousaki, A., Miyata, Y., Jinwal, U.K., Dickey, C.A., Gestwicki, J.E., and Zuiderweg,
E.R. (2011). Allosteric drugs: the interaction of antitumor compound MKT-077 with
human Hsp70 chaperones. J. Mol. Biol. 411, 614-632.
Russell, S.M., Lechner, M.G., Gong, L., Megiel, C., Liebertz, D.J., Masood, R., Correa,
A.J., Han, J., Puri, R.K., Sinha, U.K., et al. (2011). USC-HN2, a new model cell line for
recurrent oral cavity squamous cell carcinoma with immunosuppressive characteristics.
Oral Oncol. 47, 810-817.
Sabatini, S., Gosetto, F., Iraci, N., Barreca, M.L., Massari, S., Sancineto, L., Manfroni,
G., Tabarrini, O., Dimovska, M., Kaatz, G.W., et al. (2013). Re-evolution of the 2-
phenylquinolines: ligand-based design, synthesis, and biological evaluation of a potent
new class of Staphylococcus aureus NorA efflux pump inhibitors to combat antimicrobial
resistance. J. Med. Chem. 56, 4975-4989.
Sampayo-Reyes, A., and Zakharyan, R.A. (2006). Tocopherol esters inhibit human
glutathione S-transferase omega. Acta Biochim. Pol. 53, 547-552.
Sanguansin, S., Petmitr, S., P, O.C., and Pongstaporn, W. (2012). Association of
glutathione S-transferase omega gene polymorphisms with progression of head and neck
cancer. Mol. Biol. Rep. 39, 10915-10920.
Sato, M., Motomura, T., Aramaki, H., Matsuda, T., Yamashita, M., Ito, Y., Kawakami,
H., Matsuzaki, Y., Watanabe, W., Yamataka, K., et al. (2006). Novel HIV-1 integrase
inhibitors derived from quinolone antibiotics. J. Med. Chem. 49, 1506-1508.
Schmuck, E., Cappello, J., Coggan, M., Brew, J., Cavanaugh, J.A., Blackburn, A.C.,
Baker, R.T., Eyre, H.J., Sutherland, G.R., and Board, P.G. (2008). Deletion of Glu155
causes a deficiency of glutathione transferase Omega 1-1 but does not alter sensitivity to
arsenic trioxide and other cytotoxic drugs. Int. J. Biochem. Cell Biol. 40, 2553-2559.
Schmuck, E.M., Board, P.G., Whitbread, A.K., Tetlow, N., Cavanaugh, J.A., Blackburn,
A.C., and Masoumi, A. (2005). Characterization of the monomethylarsonate reductase
and dehydroascorbate reductase activities of Omega class glutathione transferase
variants: implications for arsenic metabolism and the age-at-onset of Alzheimer's and
Parkinson's diseases. Pharmacogenet. Genomics 15, 493-501.
197
Serafimova, I.M., Pufall, M.A., Krishnan, S., Duda, K., Cohen, M.S., Maglathlin, R.L.,
McFarland, J.M., Miller, R.M., Frodin, M., and Taunton, J. (2012). Reversible targeting
of noncatalytic cysteines with chemically tuned electrophiles. Nat. Chem. Biol. 8, 471-
476.
Serrao, E., Debnath, B., Otake, H., Kuang, Y., Christ, F., Debyser, Z., and Neamati, N.
(2013a). Fragment-based discovery of 8-hydroxyquinoline inhibitors of the HIV-1
integrase-lens epithelium-derived growth factor/p75 (IN-LEDGF/p75) interaction. J.
Med. Chem. 56, 2311-2322.
Serrao, E., Xu, Z.L., Debnath, B., Christ, F., Debyser, Z., Long, Y.Q., and Neamati, N.
(2013b). Discovery of a novel 5-carbonyl-1H-imidazole-4-carboxamide class of
inhibitors of the HIV-1 integrase-LEDGF/p75 interaction. Bioorg. Med. Chem. 21, 5963-
5972.
Sharma, P., Singh, D.P., Fatma, N., Chylack, L.T., Jr., and Shinohara, T. (2000).
Activation of LEDGF gene by thermal-and oxidative-stresses. Biochem. Biophys. Res.
Commun. 276, 1320-1324.
Shaw, A.T., Winslow, M.M., Magendantz, M., Ouyang, C., Dowdle, J., Subramanian, A.,
Lewis, T.A., Maglathin, R.L., Tolliday, N., and Jacks, T. (2011). Selective killing of K-
ras mutant cancer cells by small molecule inducers of oxidative stress. Proc. Natl. Acad.
Sci. U. S. A. 108, 8773-8778.
Shen, Y., and Hendershot, L.M. (2005). ERdj3, a stress-inducible endoplasmic reticulum
DnaJ homologue, serves as a cofactor for BiP's interactions with unfolded substrates.
Mol. Biol. Cell 16, 40-50.
Shen, Y., Meunier, L., and Hendershot, L.M. (2002). Identification and characterization
of a novel endoplasmic reticulum (ER) DnaJ homologue, which stimulates ATPase
activity of BiP in vitro and is induced by ER stress. J. Biol. Chem. 277, 15947-15956.
Smellie, A., Teig, S.L., and Towbin, P. (1995). Poling - promoting conformational
variation. J. Comput. Chem. 16, 171-187.
Solimini, N.L., Luo, J., and Elledge, S.J. (2007). Non-oncogene addiction and the stress
phenotype of cancer cells. Cell 130, 986-988.
Solomon, V.R., and Lee, H. (2011). Quinoline as a privileged scaffold in cancer drug
discovery. Curr. Med. Chem. 18, 1488-1508.
198
Son, J., Lee, J.J., Lee, J.S., Schuller, A., and Chang, Y.T. (2010). Isozyme-specific
fluorescent inhibitor of glutathione s-transferase omega 1. ACS Chem. Biol. 5, 449-453.
Story, M.D., and Meyn, R.E. (1999). Modulation of apoptosis and enhancement of
chemosensitivity by decreasing cellular thiols in a mouse B-cell lymphoma cell line that
overexpresses bcl-2. Cancer Chemother. Pharmacol. 44, 362-366.
Stricher, F., Macri, C., Ruff, M., and Muller, S. (2013). HSPA8/HSC70 chaperone
protein: Structure, function, and chemical targeting. Autophagy 9, 1937-1954.
Szegezdi, E., Logue, S.E., Gorman, A.M., and Samali, A. (2006). Mediators of
endoplasmic reticulum stress-induced apoptosis. EMBO Rep 7, 880-885.
Taldone, T., Kang, Y., Patel, H.J., Patel, M.R., Patel, P.D., Rodina, A., Patel, Y.,
Gozman, A., Maharaj, R., Clement, C.C., et al. (2014). Heat shock protein 70 inhibitors.
2. 2,5'-thiodipyrimidines, 5-(phenylthio)pyrimidines, 2-(pyridin-3-ylthio)pyrimidines,
and 3-(phenylthio)pyridines as reversible binders to an allosteric site on heat shock
protein 70. J. Med. Chem. 57, 1208-1224.
Tardito, S., Isella, C., Medico, E., Marchio, L., Bevilacqua, E., Hatzoglou, M., Bussolati,
O., and Franchi-Gazzola, R. (2009). The thioxotriazole copper(II) complex A0 induces
endoplasmic reticulum stress and paraptotic death in human cancer cells. J. Biol. Chem.
284, 24306-24319.
Tillotson, B., Slocum, K., Coco, J., Whitebread, N., Thomas, B., West, K.A.,
MacDougall, J., Ge, J., Ali, J.A., Palombella, V.J., et al. (2010). Hsp90 (heat shock
protein 90) inhibitor occupancy is a direct determinant of client protein degradation and
tumor growth arrest in vivo. J. Biol. Chem. 285, 39835-39843.
Townsend, D.M., and Tew, K.D. (2003). The role of glutathione-S-transferase in anti-
cancer drug resistance. Oncogene 22, 7369-7375.
Tsaytler, P., Harding, H.P., Ron, D., and Bertolotti, A. (2011). Selective inhibition of a
regulatory subunit of protein phosphatase 1 restores proteostasis. Science 332, 91-94.
Tsuboi, K., Bachovchin, D.A., Speers, A.E., Spicer, T.P., Fernandez-Vega, V., Hodder,
P., Rosen, H., and Cravatt, B.F. (2011). Potent and selective inhibitors of glutathione S-
transferase omega 1 that impair cancer drug resistance. J. Am. Chem. Soc. 133, 16605-
16616.
199
van de Giessen, E., Fogh, I., Gopinath, S., Smith, B., Hu, X., Powell, J., Andersen, P.,
Nicholson, G., Al Chalabi, A., and Shaw, C.E. (2008). Association study on glutathione
S-transferase omega 1 and 2 and familial ALS. Amyotroph. Lateral. Scler. 9, 81-84.
Verheij, M.H., Thompson, A.J., van Muijlwijk-Koezen, J.E., Lummis, S.C., Leurs, R.,
and de Esch, I.J. (2012). Design, synthesis, and structure-activity relationships of highly
potent 5-HT(3) receptor ligands. J. Med. Chem. 55, 8603-8614.
Vidal-Vanaclocha, F., Fantuzzi, G., Mendoza, L., Fuentes, A.M., Anasagasti, M.J.,
Martin, J., Carrascal, T., Walsh, P., Reznikov, L.L., Kim, S.H., et al. (2000). IL-18
regulates IL-1beta-dependent hepatic melanoma metastasis via vascular cell adhesion
molecule-1. Proc. Natl. Acad. Sci. U. S. A. 97, 734-739.
Virrey, J.J., Dong, D., Stiles, C., Patterson, J.B., Pen, L., Ni, M., Schonthal, A.H., Chen,
T.C., Hofman, F.M., and Lee, A.S. (2008). Stress chaperone GRP78/BiP confers
chemoresistance to tumor-associated endothelial cells. Mol. Cancer Res. 6, 1268-1275.
Voronov, E., Shouval, D.S., Krelin, Y., Cagnano, E., Benharroch, D., Iwakura, Y.,
Dinarello, C.A., and Apte, R.N. (2003). IL-1 is required for tumor invasiveness and
angiogenesis. Proc. Natl. Acad. Sci. U. S. A. 100, 2645-2650.
Walter, P., and Ron, D. (2011). The unfolded protein response: from stress pathway to
homeostatic regulation. Science 334, 1081-1086.
Wang, L., Perera, B.G., Hari, S.B., Bhhatarai, B., Backes, B.J., Seeliger, M.A., Schurer,
S.C., Oakes, S.A., Papa, F.R., and Maly, D.J. (2012). Divergent allosteric control of the
IRE1alpha endoribonuclease using kinase inhibitors. Nat. Chem. Biol. 8, 982-989.
Wang, M., Wey, S., Zhang, Y., Ye, R., and Lee, A.S. (2009). Role of the unfolded
protein response regulator GRP78/BiP in development, cancer, and neurological
disorders. Antioxid Redox Signal 11, 2307-2316.
Wang, M., Ye, R., Barron, E., Baumeister, P., Mao, C., Luo, S., Fu, Y., Luo, B., Dubeau,
L., Hinton, D.R., et al. (2010). Essential role of the unfolded protein response regulator
GRP78/BiP in protection from neuronal apoptosis. Cell Death Differ. 17, 488-498.
Wang, Y., Zhu, X., Yang, Z., and Zhao, X. (2013). Honokiol induces caspase-
independent paraptosis via reactive oxygen species production that is accompanied by
apoptosis in leukemia cells. Biochem. Biophys. Res. Commun. 430, 876-882.
200
Weinstein, I.B., and Joe, A. (2008). Oncogene addiction. Cancer Res. 68, 3077-3080;
discussion 3080.
Weiwer, M., Bittker, J.A., Lewis, T.A., Shimada, K., Yang, W.S., MacPherson, L.,
Dandapani, S., Palmer, M., Stockwell, B.R., Schreiber, S.L., et al. (2012). Development
of small-molecule probes that selectively kill cells induced to express mutant RAS.
Bioorg. Med. Chem. Lett. 22, 1822-1826.
Wey, S., Luo, B., and Lee, A.S. (2012a). Acute inducible ablation of GRP78 reveals its
role in hematopoietic stem cell survival, lymphogenesis and regulation of stress
signaling. PLoS One 7, e39047.
Wey, S., Luo, B., Tseng, C.C., Ni, M., Zhou, H., Fu, Y., Bhojwani, D., Carroll, W.L., and
Lee, A.S. (2012b). Inducible knockout of GRP78/BiP in the hematopoietic system
suppresses Pten-null leukemogenesis and AKT oncogenic signaling. Blood 119, 817-825.
Whitbread, A.K., Masoumi, A., Tetlow, N., Schmuck, E., Coggan, M., and Board, P.G.
(2005). Characterization of the omega class of glutathione transferases. Methods
Enzymol. 401, 78-99.
Whitbread, A.K., Tetlow, N., Eyre, H.J., Sutherland, G.R., and Board, P.G. (2003).
Characterization of the human Omega class glutathione transferase genes and associated
polymorphisms. Pharmacogenetics 13, 131-144.
Wiles, J.A., Bradbury, B.J., and Pucci, M.J. (2010). New quinolone antibiotics: a survey
of the literature from 2005 to 2010. Expert Opin. Ther. Pat. 20, 1295-1319.
Willett, P. (2006). Similarity-based virtual screening using 2D fingerprints. Drug Discov
Today 11, 1046-1053.
Williamson, D.S., Borgognoni, J., Clay, A., Daniels, Z., Dokurno, P., Drysdale, M.J.,
Foloppe, N., Francis, G.L., Graham, C.J., Howes, R., et al. (2009). Novel adenosine-
derived inhibitors of 70 kDa heat shock protein, discovered through structure-based
design. J. Med. Chem. 52, 1510-1513.
Wisniewska, M., Karlberg, T., Lehtio, L., Johansson, I., Kotenyova, T., Moche, M., and
Schuler, H. (2010). Crystal structures of the ATPase domains of four human Hsp70
isoforms: HSPA1L/Hsp70-hom, HSPA2/Hsp70-2, HSPA6/Hsp70B', and
HSPA5/BiP/GRP78. PLoS One 5, e8625.
201
Workman, P., Burrows, F., Neckers, L., and Rosen, N. (2007). Drugging the cancer
chaperone HSP90: combinatorial therapeutic exploitation of oncogene addiction and
tumor stress. Ann. N. Y. Acad. Sci. 1113, 202-216.
Xu, S., Grande, F., Garofalo, A., and Neamati, N. (2013). Discovery of a novel orally
active small-molecule gp130 inhibitor for the treatment of ovarian cancer. Mol. Cancer
Ther. 12, 937-949.
Yan, X.D., Pan, L.Y., Yuan, Y., Lang, J.H., and Mao, N. (2007). Identification of
platinum-resistance associated proteins through proteomic analysis of human ovarian
cancer cells and their platinum-resistant sublines. J. Proteome Res. 6, 772-780.
Yang, W.S., and Stockwell, B.R. (2008). Synthetic lethal screening identifies compounds
activating iron-dependent, nonapoptotic cell death in oncogenic-RAS-harboring cancer
cells. Chem. Biol. 15, 234-245.
Yano, S., Nokihara, H., Yamamoto, A., Goto, H., Ogawa, H., Kanematsu, T., Miki, T.,
Uehara, H., Saijo, Y., Nukiwa, T., et al. (2003). Multifunctional interleukin-1beta
promotes metastasis of human lung cancer cells in SCID mice via enhanced expression of
adhesion-, invasion- and angiogenesis-related molecules. Cancer Sci. 94, 244-252.
Yeung, B.H., Kwan, B.W., He, Q.Y., Lee, A.S., Liu, J., and Wong, A.S. (2008). Glucose-
regulated protein 78 as a novel effector of BRCA1 for inhibiting stress-induced
apoptosis. Oncogene 27, 6782-6789.
Yorimitsu, T., and Klionsky, D.J. (2007). Endoplasmic reticulum stress: a new pathway
to induce autophagy. Autophagy 3, 160-162.
Zhou, H., Zhang, Y., Fu, Y., Chan, L., and Lee, A.S. (2011). Novel mechanism of anti-
apoptotic function of 78-kDa glucose-regulated protein (GRP78): endocrine resistance
factor in breast cancer, through release of B-cell lymphoma 2 (BCL-2) from BCL-2-
interacting killer (BIK). J. Biol. Chem. 286, 25687-25696.
Zitvogel, L., Kepp, O., Galluzzi, L., and Kroemer, G. (2012). Inflammasomes in
carcinogenesis and anticancer immune responses. Nat. Immunol. 13, 343-351.
Zuiderweg, E.R., Bertelsen, E.B., Rousaki, A., Mayer, M.P., Gestwicki, J.E., and Ahmad,
A. (2013). Allostery in the Hsp70 chaperone proteins. Top. Curr. Chem. 328, 99-153.
202
Appendix
A1. Initial GSTO1 hit, 4a (401) inhibits cancer cell proliferation.
We studied the effects of the initial hit, 4a in colon cancer cell lines. 4a inhibited cell
proliferation (a) and colony formation (b) in HCT116 cell line. Further, 24 h treatment with 4a
blocked cell cycle progression with an S-phase accumulation at lower concentrations and a G2/M
arrest at high concentration (c). We also observed an atypical accumulation of p21 (d). p21
accumulation results in G0/G1 arrest. But in some cases, it also occurs during G2/M. Similar
a)
c)
b)
p21
β β-Tubulin
0 0.5 1 2.5 5 10 K401 (µ µM)
Cell$line:$HCT116, $Drug$treatment$–$24H$
0"
20"
40"
60"
80"
100"
%"Total"labeled"cells"
Cel l "cy cl e"a n a l y si s"
G2/M"
S"
G0/ G1"
Cell$line:$HCT116, $24H$treatme nt$
401"(µ µM) "" 0" 1" 5" 10"
KR-401
-1 0 1 2
-20
0
20
40
60
80
100
Log Concentration (µM)
% Inhibition of cell proliferation
HCT116
p53+/+
HT29
H630
4a (µ µM)
DMSO 10 µ µM 5 µ µM
2.5 µ µM 1 µ µM 0.5 µ µM
401
d)
203
G2/M arrest with p21 accumulation have been reported in some cell lines treated with
doxorubicin and proteasome inhibitors.
A2. Time course of 4a cytotoxicity.
Following a time course for induction of cytotoxicity, we also found that treatment
with GSTO1 inhibitors for 6 h or longer induced irreversible cytotoxicity.
0
20
40
60
80
100
120
0 2 4 6 8 10
% Cell Viability
Concentration (uM)
TIME COURSE - K401
1
2
4
6
12
24
72
204
A3. Effect of 4a on cell signaling pathways.
Treatment with 4a activated JNK-mediated stress signaling. 4a induced
phosphorylation of JNK in HCT116 cells within 10 min of treatment with significant and
sustained activation even after 6 h, point of no return for cytotoxicity (a,b). Interestingly,
the p46 isoform of JNK was selectively phosphorylated over the p54 isoform. 4a
treatment also induced ERK phosphorylation as an immediate response which was
sustained activation upon longer treatment (c). Sustained phosphorylation of ERK has
been reported in response to reactive oxygen species. A decrease in expression of
antioxidant protein SOD2 and increased phosphorylation of p53 at ser 392 (d,e). p53 has
been shown to repress SOD2 expression and induce cellular oxidative stress
5
.
5
Drane, P.; Bravard, A.; Bouvard, V.; May, E. Reciprocal down-regulation of p53 and SOD2 gene
expression-implication in p53 mediated apoptosis. Oncogene 2001, 20, 430-439
0 10
’!
1h 4h 6h 12h 24h
pJNK
T183/Y185
p-c-Jun
S73
GAPDH
DMSO 0.5 1 2.5 5 10
pJNK
T183/Y185
p54$
p46$
KR401 (µ µM)
GAPDH
p-c-Jun
S73
KR401 (10 µ µM)
30
’!
GAPDH
0 10
’
30
’
1h 4h 6h 12h 24h KR401 (10 µ µM)
pERK
GAPDH
0 1h 4h 6h 12h 24h
K401 (10 µ µM)
GAPDH
pP53
S392
10
’
30
’
0 1h 4h 6h 12h 24h
K401 (10 µ µM)
GAPDH
SOD2
10
’
30
’
Cell line: HCT116 Cell line: HCT116
a)
b)
c)
d)
e)
205
A4. 4a induces ER stress.
Treatment with 4a caused an increase in GRP78 and CHOP expression, indicating
an activation of ER stress signaling (a, b). While GSTO1 has been snown to inhibit ER
JNK1/2 c-Jun
P
P
G2/M arrest,
Apoptosis
MEK ERK1/2
P
Apoptosis
CK2 p53
P
Growth
suppression
GSTO inhibitor
OXIDATIVE/ REDOX
STRESS ?
---- Proposed mechanism
GRP78
β β-Tubulin
CHOP
DMSO 0.5 1 2.5 5 10 K401 (µ µM)
0 10’ 30’ 1h 4h 6h 12h 24h
GRP78
GAPDH
K401 (10 µ µM)
(A)
(B)
206
ryanodine receptor (RyR2), inhibition of GSTO1 activity or mutating C32 activates it
6
.
Interestingly, activation of the ryanodine receptor has been shown to increase in
cytoplasmic calcium and induce ER stress
7
.
A5. 4a enhances cisplatin induced cytotoxicity.
Combination with 4a enhanced the cytotoxic effects of cisplatin in HCT116 cells,
as assessed by colony survival.
6
Dulhunty, A., Gage, P., Curtis, S., Chelvanayagam, G., and Board, P. (2001). The glutathione transferase
structural family includes a nuclear chloride channel and a ryanodine receptor calcium release channel
modulator. The Journal of biological chemistry 276, 3319-3323
7
Luciani, D.S., Gwiazda, K.S., Yang, T.L., Kalynyak, T.B., Bychkivska, Y., Frey, M.H., Jeffrey, K.D.,
Sampaio, A.V., Underhill, T.M., and Johnson, J.D. (2009). Roles of IP3R and RyR Ca2+ channels in
endoplasmic reticulum stress and beta-cell death. Diabetes 58, 422-432.
* p<0.05
0
20
40
60
80
100
120
KR401 (µ µM) 0 0.1 0.5 1
Cisplatin
(1µ µM)
- - - -
0 0.1 0.5 1
+ + + +
*
*
% Colony survival
207
A6. Co-incubation of antioxidants (GSH and N-acetylcysteine) have no effect on GSTO1
inhibition by C1-27.
0
120’
60’
0
120
60’
0
120
60’
0
120
60’
10D1
(1µ µM)
DMS
5m
10m
DMS
5m
10m
DMS
5m
10m
DMS
5m
10m
0
120’
60’
0
120
60’
0
120
60’
0
120
60’
10D1
(1µ µM)
DMS
1m
10m
DMS
1m
10m
DMS
1m
10m
DMS
1m
10m
208
A7. GSTO1 inhibitor C2-22 inhibits IL-1 β β secretion from activated
monocytes.
THP-1 cells were stimulated with LPS and activated with ATP to secrete IL-1β.
Pre-treatment with GSTO1 inhibitor C2-22 (a) prior to stimulation with ATP caused an
inhibition of IL-1β release, as measured by ELISA (a). This was further confirmed by
western blotting (b). Inhibition of IL-1β release into the supernatant was accompanied by
an accumulation of pro-IL- 1β in the cell lysates (indicating inhibition of processing) and
accumulation of mature-IL-1β (indicating inhibition of release) (c).
-3 -2 -1 0 1 2
0
20
40
60
80
100
Log [µM]
%Inhibition of IL-1 β release
IC
50
= 3.9 µ µM
10C1
Mature-IL-1β β" "
50 10 5 0
Coomasie " "
C2-22 (µ µM)
ATP (5mM)
+ + + +
Supernatant " "
Pro-IL-1β β" "
50 10 5 0
β β-Tubulin" "
C2-22 (µ µM)
ATP (5mM) + + + +
Cell Lysate" "
Mature-IL-1β β" "
Cell$line:$THP+1, $primed$with$LPS$–$100$ng/mL, $3H, $Drug$
treatment$–$2H, $ATP$+$0.5H$$$
Cell$line:$THP+1, $primed$with$$LPS$–$100$ng/mL, $3H, $
Drug$treatment$–$2H, $ATP$+$0.5H$$$
a)
b)
c)
C2-22
Abstract (if available)
Abstract
Cancer cells exploit cellular stress response mechanisms, such as the unfolded protein response and oxidative stress response, to cope with unfavorable growth conditions and to survive through cytotoxic treatments. Cellular stress response pathways are therefore attractive targets for developing novel anticancer agents. We have identified novel small‐molecule compounds that target two players of the cellular stress response pathways—Glutathione S transferase omega 1 (GSTO1) and the 78 kDa glucose‐regulated protein (GRP78). ❧ GSTO1 is an atypical GST isoform, with an active‐site cysteine, that is involved in oxidative stress response. GSTO1 is overexpressed in several cancers and has been implicated in drug resistance. We show that silencing GSTO1 significantly impairs cancer cell survival and proliferation validating GSTO1 as a new target in oncology. Through extensive screening and hit optimization, we have identified a series of chloroacetamide‐containing small‐molecule compounds that are potent GSTO1 inhibitors with activity in the nanomolar range. Crystal structures and biochemical studies revealed a unique reversible covalent binding to the active site cysteine. Potent GSTO1 inhibitors suppressed the growth of cancer cells, enhanced the cytotoxic effects of cisplatin, and inhibited tumor growth without apparent systemic toxicity. GSTO1 inhibitors also blocked IL‐1β secretion from activated monocytes. Our findings demonstrate the therapeutic utility of selective GSTO1 inhibitors as anticancer and anti‐inflammatory agents. ❧ We also have identified small‐molecule inhibitors of the 78 kDa glucose regulated protein (GRP78) chaperone—a key mediator of the unfolded protein response. As a molecular chaperone belonging to the HSP70 family, GRP78 regulates protein quality control and relieves proteotoxic stress. In addition to providing a growth advantage to cancer cells, GRP78 induction can lead to drug resistance. Suppressing GRP78 levels using siRNA has been shown to slow cancer cell progression and overcome drug resistance. Another approach to inactivating GRP78 is by inhibiting its ATPase activity that has not received much attention. Here, we have identified novel classes of small‐molecule inhibitors of GRP78 ATPase activity through a ligand‐based drug design approach. Based on the initial hits, we further optimized several analogues with more potent activity for GRP78 inhibition. The lead compounds induced CHOP significantly under conditions of ER stress and activated autophagy. These compounds induced cancer cell death in a panel of breast cancer cell lines and increased sensitivity to chemotherapeutic agents, further supporting the role for GRP78 in cancer survival and drug resistance. The lead compounds also caused degradation of HSP70 client proteins. In vivo studies in a breast cancer xenograft mice model further demonstrated efficacy of these compounds in impeding tumor growth alone and in combination with standard chemotherapeutic agent, doxorubicin. Taken together, compounds identified in this dissertation provide novel approaches to target cellular stress pathways in cancer and will serve as initial leads for further optimization into more potent compounds.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Targeting cellular redox modulations for pancreatic cancer treatment
PDF
Discovery of novel small molecules for ovarian cancer treatment
PDF
Identification of small-molecules targeting CXCR2 function and signaling
PDF
Development of novel small molecules targeting mitochondrial and oxidative stress signaling pathways for pancreatic cancer therapy
PDF
Discovery of novel small-molecule compounds with antiviral and anticancer properties
PDF
The roles of endoplasmic reticulum chaperones in regulating liver homeostasis and tumorigenesis
PDF
Role of GRP78 and its novel cytosolic isoform GRP78va in regulating the unfolded protein response in cancer
PDF
Utilizing novel small molecules to modulate GRP78 expression and activity in cancer
PDF
Mechanistic studies of novel small molecule anti-cancer agents using next generation sequencing
PDF
Inhibition of tumor cell growth by mefloquine via multimechanistic effects involving increased cellular stress, inhibition of autophagy, and impairment of cellular energy metabolism
PDF
The role of glucose-regulated proteins in endometrial and pancreatic cancers
PDF
Discovery of novel small molecules targeting cancer cell metabolism
PDF
Design and discovery of small molecules inhibiting the interaction of cellular LEDGF/p75 and HIV-1 integrase
PDF
HIV-1 integrase and human APE1: two DNA-processing enzymes and two druggable targets
PDF
Targeting human base excision repair as a novel strategy in cancer therapeutics
PDF
Cell surface translocation mechanism of stress-inducible GRP78 in human cancer
PDF
The role of endoplasmic reticulum chaperones in adipogenesis, liver cancer and mammary gland development
PDF
The role of endoplasmic reticulum chaperone protein GRP78 in breast cancer
PDF
Novel modes of HIV-1 integrase inhibition
PDF
Molecular mechanisms of chemoresistance in breast cancer
Asset Metadata
Creator
Ramkumar, Kavya
(author)
Core Title
Design of novel anticancer agents targeting cellular stress response pathways
School
School of Pharmacy
Degree
Doctor of Philosophy
Degree Program
Pharmaceutical Sciences
Publication Date
08/07/2016
Defense Date
06/18/2014
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
anticancer,drug design,GRP78,GSTO1,inhibitor,OAI-PMH Harvest
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Neamati, Nouri (
committee chair
), Duncan, Roger (
committee member
), Okamoto, Curtis Toshio (
committee member
), Stiles, Bangyan L. (
committee member
)
Creator Email
kavya.usc@gmail.com
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-456380
Unique identifier
UC11287106
Identifier
etd-RamkumarKa-2762.pdf (filename),usctheses-c3-456380 (legacy record id)
Legacy Identifier
etd-RamkumarKa-2762.pdf
Dmrecord
456380
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Ramkumar, Kavya
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
anticancer
drug design
GRP78
GSTO1
inhibitor