University of Southern California Dissertations and Theses

Discovery of novel small molecules targeting cancer cell metabolism

(USC Thesis Other)

Discovery of novel small molecules targeting cancer cell metabolism

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content MELISSA MILLARD

DOCTORAL DISSERTATION

DISCOVERY OF NOVEL SMALL MOLECULES TARGETING CANCER CELL
METABOLISM
2
Dedication

This dissertation is dedicated to my family, to my kind and loving parents, Geoff
and Marie in recognition of their unconditional support throughout my life, to my
wonderful husband David, my source of strength and inspiration, in appreciation
of his commitment and dedication and the sacrifices made in helping me achieve
my dreams, to my father-in-law, Guy for his infectious enthusiasm and pride in
my scientific accomplishments and to my mother- in- law Beverly for her sweet
and touching letters so lovingly written and to my brother and sister-in-law, Geoff
and Agnes for their love and caring throughout the years.

Acknowledgements

I would like to acknowledge: my colleague Dr. Xue Fei Cao who was instrumental
in setting me on the path to earning my doctorate, all of my past mentors for their
patient and thorough instruction and my current mentor Dr. Nouri Neamati for
having the insight to recognize my potential and the willingness to guide, inspire
and educate me.

3
Table of Contents

List of Figures________________________________________________________ 5
List of Tables__________________________________________________ 10
Chapter 1 Background and Rationale ______________________________ 13
Chapter 2 Preclinical evaluation of novel triphenylphosphonium salts with
broad-spectrum activity _________________________________________ 38
Introduction __________________________________________________ 38
Results______________________________________________________ 44
Discussion ___________________________________________________ 56
Experimental Procedures________________________________________ 60
Chapter 3 TP compounds in the treatment of pancreatic ductal
adenocarcinoma _______________________________________________ 69
Introduction __________________________________________________ 69
Results______________________________________________________ 74
Discussion ___________________________________________________ 89
Experimental Procedures________________________________________ 94
Chapter 4 A selective mitochondrial-targeted chlorambucil with remarkable
cytotoxicity in breast and pancreatic cancers ______________________ 102
Introduction _________________________________________________ 102
Results_____________________________________________________ 106
Discussion __________________________________________________ 117
Experimental Procedures_______________________________________ 126
4
Chapter 5 A mitochondrial-targeted doxorubicin derivative, Mito-Dox__ 132
Introduction _________________________________________________ 132
Results_____________________________________________________ 134
Discussion __________________________________________________ 143
Experimental Procedures_______________________________________ 145
Chapter 6 A mitochondrial-targeted temozolomide, Mito-Tem ________ 150
Introduction _________________________________________________ 150
Results_____________________________________________________ 156
Discussion __________________________________________________ 157
Experimental Procedures_______________________________________ 160
Chapter 7 Novel sulfonamides targeting cancer cell metabolism______ 162
Introduction _________________________________________________ 162
Results_____________________________________________________ 164
Discussion __________________________________________________ 192
Experimental Procedures_______________________________________ 195
References___________________________________________________ 203

5
List of Figures

Figure 1.1 U.S. population, breast, colon, and pancreatic cancer age-related
incidence and 5-year survival, pancreatic cancer _______________________ 14

Figure 1.2 Chemical structures of chemotherapeutic drugs used in the treatment
of breast and pancreatic cancer.____________________________________ 17

Figure 1.3 Chemical structures of selective estrogen receptor modulators and
aromatase inhibitors _____________________________________________ 20

Figure 1.4 Metabolic reprogramming results in key differents between normal
and cancer cell mitochondria_______________________________________ 36

Figure 2.1 Triphenylphosphonium cation accrues in mitochondria based on
membrane potential. _____________________________________________ 39

Figure 2.2 Chemical structures of lead TP compounds __________________ 46

Figure 2.3 Colony formation assay HCT116 p53 +/+ treated with TP
compounds.____________________________________________________ 46

Figure 2.4 TP compounds arrest cell cycle progression _________________ 46

Figure 2.5 TP compounds are effective as single agents in delaying tumor
growth ________________________________________________________ 48

Figure 2.6 TP compounds rapidly accumulate in energized mitochondria ____ 53

Figure 2.7 TP compounds mitochondrial function ______________________ 53

6
Figure 3.1 Progression model of pancreatic cancer _____________________ 71

Figure 3.2 Mitochondrial function supports and is required for KRAS-dependent
cell proliferation. ________________________________________________ 73

Figure 3.3 Timeline of bioenergetic changes occurring during KRAS driven
ocogenic transformation __________________________________________ 73

Figure 3.4 TP compounds impinge on functions required for pancreastic cancer
progression ____________________________________________________ 75

Figure 3.5 TP197 and its close analogs suppress tube formation a in 3-D model
of endothelial cell differentiation ____________________________________ 77

Figure 3.6 Mitochondrial attributes of KRAS wild-type and mutant cell lines__ 81

Figure 3.7 TP 421 mitchondrial uptake is rapid and sustained_____________ 83

Figure 3.8 TP421 parent coumarin does not enter mitochondria.___________ 83

Figure 3.9 Chemical structures of TP187 analogs ______________________ 85

Figure 3.10 TP197 caused rapid phosphorylation of AMPK and ACC _______ 87

Figure 3.11 TP197 caused rapid turnover of LC3B and DNA damage response
_____________________________________________________________ 87

Figure 4.1 Chemical structures of chlorambucil and Mitochlor. ___________ 105

Figure 4.2 Confocal images of MIA PaCa-2 treated with Mito-Chlor _______ 109

7
Figure 4.3 Mitochondrial membrane potential of Rho wild-type and Rho θ cell
lines_________________________________________________________ 110

Figure 4.4 Cell cycle distrubtion of MIA PaCa-2_______________________ 112

Figure 4.5 Western blot for γH2A.X expression _______________________ 115

Figure 4.6 Western blot MIA PaCa-2 treated with chlorambucil and Mito-Chlor
____________________________________________________________ 116

Figure 5.1 Doxorubicin drug domains involved in topoisomerase II inhibition 134

Figure 5.2 Chemical structures of doxorubicin, Mito-Dox and Mito-Dox-
Hexanoamide _________________________________________________ 135

Figure 5.3 Micrograph demonstrating mitochondrial specificity of Mito-Dox
____________________________________________________________ 136

Figure 5.4 Colony formation assay_________________________________ 138

Figure 5.5 Mito-Dox induced cell cycle arrest in MDA MB 468 cells _______ 140

Figure 5.6 Mito-Dox hexanoamide causes cells cycle arrest but does not
increase hypodiploid content______________________________________ 141

Figure 5.7 Western blot of γH2A.X expression in MDA MB treated with
doxorubicin , Mito-Dox or Mito-Dox-Hexanoamide _____________________ 142

Figure 6.1 Schema illustrating temozolomide mechanism of action. _______ 152

Figure 6.2 Chemical Structures of temozolomide and Mito-Tem __________ 155
8

Figure 7.1 Target identification using DARTS assay and MS. ____________ 173

Figure 7.2 Target validation using DARTS-Western blotting _____________ 174

Figure 7.3 Upstream inhibition of the glycolytic pathway counters effects of
AV220 treatment on cell proliferation _______________________________ 175

Figure 7.4 Western blot, MDA MB 468 treated for indicated time points with 20
µM AV220 ____________________________________________________ 176

Figure 7.5 AV220 treatment under nutrient restricted conditions induced G0/G1
cell cycle arrest ________________________________________________ 178

Figure 7.6 AV220 treatment under nutrient restricted conditions caused
extensive intracellular vesicle formation _____________________________ 179

Figure 7.7 LC3B associates with AV220-induced vesicles_______________ 180

Figure 7.8 AV220 increases autophagic flux _________________________ 181

Figure 7.9 AV220-induced changes in morphology in a panel of cancer cell lines
____________________________________________________________ 182

Figure 7.10 AV220 caused a time- and dose-dependent increase in LC3B-II 182

Figure 7.11 HCT116 colony formation assay _________________________ 183

Figure 7.12 AV220 activates UPR _________________________________ 184

Figure 7.13 Erk 1/2, p38 and Src phosphorylation precede AV220 induced
autophagy ____________________________________________________ 185
9

Figure 7.14 PKM2 over-expression in pancreatic cancer tissue samples
____________________________________________________________ 189

Figure 7.15 PKM2 over-expression in breast cancer tissue samples_______ 189

Figure 7.16 PKM2 over–expression in gastric cancer tissue samples _________
____________________________________________________________ 190

Figure 7.17 Single agent AV220 inhibits HCT116 p53 +/+ colon cancer growth in
a mouse xenograft model ________________________________________ 191

10
List of Tables

Table 1.1 Breast cancer stage distribution and 5-year relative survival by stage
at diagnosis____________________________________________________ 16

Table 1.2 Chemotherapeutic regimens used in breast cancer treatment_____ 18

Table 1.3 Anti-endocrine therapies used in breast cancer________________ 19

Table 1.4 Pancreatic cancer stage distribution and 5-year relative survival by
stage at diagnosis _______________________________________________ 23

Table 1.5 Pancreatic cancer treatment regimens _______________________ 24

Table 1.6 Colorectal cancer stage distribution and 5-year relative survival by
stage at diagnosis for 2002-2008, all races, both sexes __________________ 25

Table 2.1 IC
50
values of TP187, TP197 in MTT assay___________________ 45

Table 3.1 Core signaling pathways altered in PDAC tumors ______________ 75

Table 3.2 IC
50
values of TP compounds in PDAC cell lines _______________ 76

Table 3.3 IC
50
values of TP187, TP197 in KRAS cell lines ________________ 78

Table 3.4 Anti-proliferative properties of Allyl TP and analogs _____________ 85

Table 4.1 Molecular attributes of selected breast cancer cell line panel. ____ 107

11
Table 4.2 IC
50
values (µM) of chlorambucil and Mtio-Chlor in a panel of breast
cancer cell lines________________________________________________ 107

Table 4.3 Molecular attributes of selected PDAC cell line panel___________ 108

Table 4.4 IC
50
values of chlorambucil and Mito-Chlor in a panel of PDAC cell
lines_________________________________________________________ 108

Table 4.5 IC
50
values of chlorambucil and Mito-Chlor in MDA MB 231 Rho wild-
type and paired Rho θ cell lines ___________________________________ 111

Table 4.6 Effect of ROS scavenging on IC
50
values of chlorambucil and Mito-
Chlor in MDA MB 468 cell line_____________________________________ 117

Table 5.1 IC
50
values of doxorubicin, Mito-Dox and Mito-Dox-hexanoamide _ 137

Table 6.1 IC
50
values of temozolomide, Mito-Tem in a panel of cancer cell lines
____________________________________________________________ 156

Table 7.1 IC
50
values of AV220 in a panel of cancer cell lines ____________ 164

Table 7.2 Scaffold 1 AV220 analogs________________________________ 166

Table 7.3 Scaffold 2 AV220 analogs________________________________ 168

Table 7.4 Scaffold 3 AV220 analogs________________________________ 171

Table 7.5 Scaffolds 5 and 6 AV220 analogs__________________________ 172

Table 7.6 Putative AV220 targets __________________________________ 175

12
Table 7.7 Analysis comparison for gene PKM2, upregulation in tumor vs normal
tissue________________________________________________________ 187

Table 7.8 Statistical ranking of PKM2 gene analyses___________________ 187

Table 7.9 Analysis comparison for gene PKM2, upregulation in tumor vs normal
tissue________________________________________________________ 188

Table 7.10 Statistical ranking of expanded PKM2 gene analyses _________ 188

Table 7.11 IC
50
values of AV220, erlotinib in a panel of cancer cell lines ____ 192
13
Chapter 1
Background and Rationale
Introduction

Drug discovery and development is essential to the advancement of cancer
treatment.
A multitude of factors necessitate the development of improved anti-cancer
agents. Traditional chemotherapeutic agents are a mainstay of clinical treatment
regimens but are generally limited in their efficacy by narrow therapeutic widows
and dose-limiting toxicities. Certain aggressive, drug-resistant cancers have no
standard of treatment due to a lack of effective drugs. Many late-stage,
metastatic cancers have no curative treatment options. The design and discovery
of novel anti-cancer agents, therefore, is an obvious step toward addressing the
unmet need for improved therapeutics.

A shift in population demographics will compound the already dire need for
effective cancer treatments.
Shifts in aged-based demographics toward an ‘aged population’ will augment the
need for effective cancer treatments. World population estimates predict steady
increases in the number of elderly such that by the year 2100 more than one-
third of the total U.S. population will consist of persons age 65 and over (Figure
1.1) (United Nations Department of Economics and Social Affairs, 2010). These
14

A C
B D
E

Figure 1.1 (A) Percentage of U.S. population, age 65 and over, by decade (B,C,E)
breast, colon, and pancreatic cancer age-related Incidence and (D) 5-year survival,
pancreatic cancer .

15
estimates are not unique to the United States; similar trends are projected for the
world’s population as well. Advanced age is a risk factor associated with the
majority of adult-onset cancers. Barring an unlikely decrease in cancer incidence,
it is reasonable to assume the projected shift in age demographics will bring with
it an unprecedented rise in the total number of cancer diagnoses. In coming
years the impending cancer ‘crisis’ will compound the ever present need to
expedite drug development. Thus great benefit could be achieved by focusing
drug-development efforts on treatments for cancers afflicting large segments of
the population or those lacking standard of care treatments. To this end the work
presented in this thesis describes the development of new cancer drugs as
potential treatments for breast, pancreatic and colon cancers.

Breast Cancer is the leading cause of cancer in women
Breast cancer can arise at any age, but the rate of incidence increases with
advancing age (Figure 1.1-C). According to SEER cancer statistics compiled
over the last three decades, breast cancer is already the most frequent cancer
occurring among females in the United States. In the years between 2005 - 2009,
breast cancer was diagnosed at a rate of 124 per 100,000 women (Howlader N,
2013). This figure accounts for more than a fourth of all cancers combined for
females during those years. The wide-spread use of screening methods such as
mammography and increased awareness regarding the need to perform self-
exams has been critical in the detection of breast cancer at earlier stages. (Table
1). The majority of breast cancers are now detected as localized disease that
16
Table 1.1 Breast Cancer Stage Distribution and 5-year Relative Survival by Stage at Diagnosis
Stage Percent distribution 5-year relative survival
local 60 98.4
regional 33 83.9
distant 5 23.8
unknown 2 50.7

with appropriate treatment can achieve 5-year survival rates as high as 98.4%.
Locally advanced disease is highly treatable as well, with nearly 84% survival
rate at 5-years. A small percentage of patients present with advanced disease
that carries little hope of cure (Howlader 2013).

Status and Limitations of Current Breast Cancer Treatment regimens
The current treatment options for available for breast cancer patients include
surgery, localized therapy, and systemic therapy. Localized therapy usually takes
the form of radiation therapy directed at the site of tumor occurrence. Systemic
therapies refer to chemotherapeutic agents, endocrine therapies and biologic
agents. Treatment plans are generally based on the extent of disease present at
the time of diagnosis. While surgery and therapy are generally reserved for
localized disease, systemic therapies are utilized in the treatment of all stages of
disease.
Chemotherapeutic agents are administered either as neoadjuvant or adjuvant
treatments. Commonly used agents include anthracyclines, taxanes, the nitrogen
mustard cyclophosphamide and the anti-metabolite, nucleoside analog 5-
fluoruracil (Figure 1.2). Treatment regimens consist of sequential or combined
administration of two or more of the agents, generally given over 4-8 three-week
17

Figure 1.2 Chemical structures of chemotherapeutic drugs used in the treatment of
breast and pancreatic cancer.

18
cycles (Table 1.2). These agents act by targeting processes critical to cell
division, each class having a distinct mechanism of action. In the neoadjuvant
setting, chemotherapy is administered with the intent to shrink the tumor prior to
surgical removal. Adjuvant use of chemotherapy is intended to prevent disease
recurrence (Taghian et al., 2013).
Table 1.2 Chemotherapeutic regimens used in breast cancer treatment.

Regimen Drugs
Dosage &
administration
Schedule
Drug-related
toxicities
AC
doxorubicin
cyclophosphamide
60 mg/m2 IV infusion
600 mg/m2 IV
infusion
4 x 21 day
cycles, drugs
administered on
day 1
myeolotoxicity
cardiotoxicity
AC/taxane
doxorubicin
cyclophosphamide
paclitaxel or
docetaxel
60 mg/m2 IV infusion
600 mg/m2 IV
infusion
175 mg/ m2 IV
infusion
100 mg/m2 IV
infusion
AC followed by
ptxl 4x 2-3 wk
cycles
dctxl 4 x 3 wk
cycles
myeolotoxicity,
cardiotoxicity,
neurologic toxicity
AC/
weekly T
doxorubicin
cyclophosphamide
paclitaxel
60 mg/m2 IV infusion
600 mg/m2 IV
infusion
80 mg/m2 IV infusion
AC schedule,
ptxl given on
days 1, 8, 15,
and 22
myeolotoxicity,
cardiotoxicity,
neurologic toxicity
FEC
5-fluorouracil
epirubicin
cyclophosphamide
500 mg/m2 IV bolus
100 mg/m2 IV
infusion
500 mg/m2 IV
infusion
6 to 8 x 21 day
cycles, given on
day 1
myeolotoxicity,
cardiotoxicity, GI
toxicity, palmar-plantar
erythrodysesthesia
FAC
5-fluorouracil
doxorubicin
cyclophosphamide
600 mg/m2 IV bolus
60 mg/m2 IV infusion
600 mg/m2 IV
infusion
6 to 8 x 21 day
cycles, given on
day 1
myeolotoxicity,
cardiotoxicity, GI
toxicity, palmar-plantar
erythrodysesthesia
TAC
docetaxel
doxorubicin
cyclophosphamide
75 mg/m2 IV infusion
50 mg/m2 IV infusion
500 mg/m2 IV
infusion
6 to 8 x 21 day
cycles, given on
day 1
myeolotoxicity,
cardiotoxicity, GI
toxicity, palmar-plantar
erythrodysesthesia
TC
docetaxel
cyclophosphamide
75 mg/m2 IV infusion
600 mg/m2 IV
infusion
4 x 21 day
cycles, drugs
administered on
day 1
myeolotoxicity,
cutaneous, mucosal,
neurologic toxicity
ACTH
doxorubicin
cyclophosphamide
paclitaxel
trastuzumab
60 mg/m2 IV infusion
600 mg/m2 IV
infusion
80 mg/m2 IV infusion
4 mg/kg IV loading, 2
mg/kg
AC schedule,
ptxl given on
days 1, 8, 15,
and 22
tstzmb weekly x
52 wks
myeolotoxicity,
cardiotoxicity,
neurologic toxicity,
pulmonary toxicity

19
Endocrine therapies consist of selective estrogen receptor modulators (SERMs)
and aromatase inhibitors. Ovarian suppression, either surgical or
pharmacological is another type of anti-endocrine therapy (Brenner et al.,
2013b). The characteristics of the various agents used in hormonal therapy are
listed in Table 1.3. The rationale for the use of anti-endocrine therapies is based
on the observation that many cancers of the female reproductive system are
sensitive to the proliferative stimuli of endogenous estrogens. These agents are
rarely administered in the neo-adjuvant setting, but are widely used as adjuvant
treatment.
Table 1.3 Anti-endocrine therapies used in breast cancer
treatment
preferred drug/
method
dosage &
administration
Indications
drug-related
toxicities, risks
SERMs tamoxifen
20 mg daily, oral
duration of 5-10
years
pre-/post-
menopausal
women
slightly increased risk
of stroke, increased
risk of uterine cancer,
menstrual
irregularities, hot
flashes
Aromatase
Inhibitors (AIs)
anastraxole or
letrozole or
exemestane
1mg daily
2.5 mg daily
25 mg daily
duration of 5 years
post-
menopausal
women only
musculo-skeletal side
effects, sexual
dysfunction, increased
risk of osteoporosis,
cardiovascular risks
Ovarian
suppression
and ablation
radio-ablation
surgical removal
hormonal
ablation
radiation
N/A
LHRH agonist
goserelin, 3.6 mg SQ
every 28 d
pre-menopausal
women only
symptoms associated
with menopause

More recently, the efficacy of anti-HER2 biologics against HER2 over-expressing
breast cancers in the neoadjuvant and adjuvant settings has been established.
Trastuzamab (Herceptin®) is a humanized monoclonal antibody that recognizes
and binds to the extracellular domain of the HER2 protein, thereby inhibiting cell

20

Figure 1.3 Chemical structures of selective estrogen receptor modulators and
aromatase inhibitors
21
proliferation and stimulating the immune mediated process of antibody-
dependent cellular cytotoxicity (ADCC) (Brenner et al., 2013b).

Early detection, improved surgical techniques and the refinement of systemic
treatment protocols are credited with improving survival rates but there remains
room for improvement as there are still many drawbacks relating to the use of
systemic therapies. The current chemotherapeutic agents used in the treatment
of breast cancers do not differentiate between healthy and diseased tissue which
can lead to potentially life-threatening sequelae and toxicities. The anthracyclines
are amongst the most effective class of chemotherapeutic agents but must be
administered with caution as they carry very serious sequelae. The risk of acute
and chronic heart failure increases with higher cumulative dosing of
anthracyclines. Extravasation injuries can cause extensive tissue necrosis.
Secondary neoplasms have also been reported in patients treated with
anthracyclines. Cardiotoxicity, hypersensitivity reactions and neuropathies are
adverse effects associated with the use of taxanes. Nitrogen mustards, like the
anthracyclines, carry the risk of cardiotoxicity and secondary neoplasms. The
risk of long term sequelae common to these agents is cause for concern given
the high percentage of patients reaching and surpassing the 5-year survival
mark. Thus the discovery of broad spectrum agents with selectivity toward
cancer cells could improve the long-term impact of chemotherapy on quality of
life.

22
In an attempt to limit the toxicities arising from chemotherapies, targeted agents
have been employed for the treatment of many different types of cancer. This
approach has been particularly successful in the treatment of breast cancers with
the exception of the particular subgroups that fail to express the targets at which
the agents are directed. For patients harboring tumors that are estrogen receptor
(ER) / progesterone receptor (PR) negative, and/or non HER2 expressing
treatment options are limited to chemotherapy. Furthermore, the loss of estrogen
sensitivity and triple negative status are traits characteristic of aggressive
disease, poor prognosis and shorter survival. These patients could benefit from
the development of safer broad spectrum agents and well as alternative targeted
agents.

Regardless of hormone receptor status and HER2 expression, patients with
metastatic disease have bleak prognosis. Once breast cancer has metastasized
to distant sites within the body, the current treatment options offer little hope for
cure. At this stage, therapeutic intervention focuses primarily on prolonging
survival and providing palliative care. Agents that prolong survival and even
reverse disease are desperately needed for this small yet significant population
of patients.

Pancreatic Cancer: Age-related Incidence, Prognosis and Current Therapies
The diagnosis of pancreatic cancer carries bleak prognosis, in any given year
the number of persons diagnosed is roughly equivalent to the number of persons
23
dying from the disease (Howlader 2013). As with most cancers, advanced age
increases the risk of developing pancreatic cancer (Figure 1.1-B). Given that the
early symptoms of pancreatic cancer are vague, and that no predictive screening
methods have been implemented, the majority of pancreatic cancer patients
present with advanced or late stage disease. Regardless, even those cases
diagnosed as localized and surgically resectable generally have less than 25% 5-
year survival rates (Howlader 2013) (Table 1.4). The survival rates for
pancreatic cancer have been stagnant and low for decades. (Figure 1.1-E)
Table 1.4 Pancreatic Cancer Stage Distribution and 5-year Relative Survival by Stage at Diagnosis
Stage Percent distribution 5-year relative survival
local 8 23.3
regional 27 8.9
distant 53 1.8
unknown 12 3.9

Current treatment options consist of surgery, radiation and systemic therapies.
Surgery is the best option to prolong survival but is limited to localized disease.
Systemic therapies can be used in the neoadjuvant and adjuvant setting alike.
Cases that are borderline-resectable often employ neoadjuvant chemotherapy
with the intent to shrink tumor size prior to surgical resection. At later stages
systemic therapies are employed to reduce symptoms and prolong survival
(Hidalgo, 2010; Mossner, 2010). The most commonly used drugs for the
treatment of pancreatic cancers are gemcitabine, 5-fluoruracil, leucovorin,
oxaliplatin, irinotecan and erlotinib (Figure 1.2) (Brenner et al., 2013a). Both
gemcitabine and 5-fluorouracil are nucleoside analogs and have dual action as
anti-metabolite drugs where they can interfere with RNA and DNA replication and
24
inhibit nucleotide biosynthesis. Leucovorin synergizes with 5-FU to enhance its
anti-metabolite activity. Oxaliplatin is a platinum-based DNA crosslinking agent.
Irinotecan is a DNA-damaging drug targeting topoisomerase I. Erlotinib is the
newest drug introduced into the pancreatic cancer treatment arsenal. It is
considered a molecular-targeted agent, in particular, a tyrosine kinase inhibitor
(Conradt et al., 2011). These agents are generally administered as combination
treatments according to the protocols listed in table 1.5.
Table 1.5 Pancreatic cancer treatment regimens
drug
dosage &
administration
schedule drug-related toxicities
gemcitabine
1000 mg/m2,
30-60 min IV
Adjuvant: Weekly 3
wks/1 wk rest
Metastatic: Weekly 7
wks/ 1 wk rest,
then 3 wks/ 1 wk for 7
cycles
myelotoxicity,hepatotoxi
city, hemolytic-uremic
syndrome, pulmonary
toxicity
FOLFIRINOX
-oxaliplatin
-leucovorin
-irinotecan
-5-fluorouracil

85 mg/m2 IV infusion
400 mg/m2 IV infusion
180 mg/m2 IV infusion
400 mg/mg2 IV bolus &
2400 mg/m2 IV infusion
All drugs given on Day
1 of 14 day cycle
myelotoxicity,
neutropenia,
thrombocytopenia,
diarrhea, mucositis or
hand-foot syndrome
FOLFOX6
-oxaliplatin
-irinotecan
-5-fluorouracil

85 mg/m2 IV infusion
400 mg/m2 IV infusion
400 mg/mg2 IV bolus &
2400 mg/m2 IV infusion
All drugs given on Day
1 of 14 day cycle
myelotoxicity,
neurologic toxicities,
diarrhea

Colon Cancer: Age-related Incidence, Prognosis and Current Therapies
Nearly 150,000 people per year receive a diagnosis of colorectal cancer (CRC).
Similar to breast and pancreatic cancer the incidence of CRC increases with age
(Figure 1.1-D). Due to the implementation of screening methods and early
detection, nearly 80% of CRCs cases are detected when disease is still relatively
25
localized. Detection at this stage carries reasonable prognosis and 5-year
survival rates (Table 1.6). The remaining 20% of patients present with late-stage
disease and for them the curative potential of treatments are very low. Only
slightly more than a third of these patients will survive five years from the time of
diagnosis. (Howlader 2013)
Table 1.6 Colorectal cancer stage distribution and 5-year relative survival by stage at diagnosis for 2002-
2008, all races, both sexes
stage at diagnosis
stage
distribution (%)
5-year
relative survival (%)
Localized (confined to primary site) 39 89.9
Regional (spread to regional lymph nodes) 36 69.6
Distant (cancer has metastasized) 20 11.9
Unknown (unstaged) 5 33.9

The current treatment modalities for CRC include surgery, radiotherapy and
systemic chemotherapies. Surgery is the preferred modality for localized, locally
advanced and in some circumstances even for metastatic disease confined to
the liver. Radiation therapy is generally used as an adjuvant. Systemic
chemotherapies have been applied in both neoadjuvant as well as the adjuvant
settings. Patients receiving systemic chemotherapy are generally given drug
combinations, such as FOLFOX, (oxaliplatin, 5-fluorouracil, and leucovorin) or
CAPOX (oxaliplatin and oral capecitabine), FOLFOXIRI (oxaliplatin, 5-
fluorouracil, leucovorin and irinotecan), FOLFIRI (5-fluoruracil, leucovorin, and
irinotecan). These drugs are generally administered on day 1 of a 14 day
treatment cycle. More recently, biologics such as the anti-VEGF monoclonal
antibody or anti-EGFR monoclonal antibodies, cetuximab and panitumumab,
have been introduced, either as part of FOLFOX and FOLFIRI regimens or in the
case of the anti-EGFR antibodies as single agents (Rodriguez-Bigas et al.,
26
2013). Many of the same drugs used in the treatment of pancreatic cancer are
also indicated for the treatment of CRC, and as such the side-effects and
treatment related toxicities are similar to those listed in table 1.5. The biologics
have better toxicity profiles but carry the risk of infusion reactions and
anaphylaxis. In addition, dermatologic rash frequently occurs with anti-EGFR
treatment and is considered a surrogate indicator of treatment efficacy.

Current status and future directions of anti-cancer drug design
Proliferation under selective pressures requires tumor cells employ adaptive
mechanisms that confer growth advantage. Adaptation under adverse conditions
promotes cellular phenotypes unique to neoplastic transformation that in turn
offer opportunities for selective targeting of cancer cells (Pathania et al., 2009b).
Selective targeting exploits the intrinsic differences between healthy and
cancerous tissue and therefore, forms the basis for the discovery and
development of safe, effective anti-cancer drugs. The ideal therapeutic should
act preferentially on tumor cells to induce apoptotic cell death while sparing
normal cells. To this end recent drug discovery efforts have adopted the
rationale that reduction in systemic toxicity and increased efficacy is best
achieved through the careful generation and validation of novel lead molecules
having selectivity toward cancer-enriched targets. Numerous agents developed
under this premise have advanced to clinical use in recent years, with mixed
results. It may yet be too soon to fully weigh the clinical success of this paradigm
but key lessons emerging suggest that refinement in this approach to drug
27
discovery is warranted. Although the precision of targeting that has been
achieved speaks to the success of medicinal chemistry, it may also be an
Achilles’ heel for cancer therapeutics. Selectivity in this setting may favor the
development of drug resistance. For example, a common mechanism of
resistance to EGFR inhibitors occurs through the evolution of tumor populations
harboring mutations that impede drug binding (Ji et al., 2006). A recent meta-
analysis of randomized clinical trials evaluating new molecularly targeted agents
and chemotherapeutics sheds doubt on the assertion that selectivity reduces
toxicity. In this study, newer agents trumped standard therapies in terms of
efficacy, but were in fact more likely to cause: Grade 3/4 adverse events,
discontinuation of treatment and toxic death (Niraula et al., 2012). Thus
refinements in targeting that reduce drug resistance and the risk of drug-related
toxicities would be a welcome improvement. From a practical standpoint, the
introduction of new agents with broad spectrum activity would be beneficial for
patient populations that are unable to benefit from molecularly targeted agents
(i.e. triple negative breast cancer patients, Philadelphia Chromosome negative
leukemia patients).
28
Targeting metabolism to treat cancers
An up and coming approach for the discovery of new cancer drugs involves
targeting metabolic pathways supporting cancer growth. Intense research over
the last decades has provided an unprecedented understanding of the nature
and cause of the metabolic reprogramming that supports tumor cell proliferation.
Application of this knowledge has returned a number of promising small
molecules aimed at disrupting tumor cell metabolism which are currently in
various stages of pre-clinical development as anti-cancer agents (Pathania et al.,
2009a).

A change in cellular bioenergetics is one of the key hallmarks of cancer.
The first observations of altered metabolism in cancer cells were reported by Otto
Warburg a early as 1924 (Warburg, 1956b). Warburg noted that despite an
ample supply of oxygen, tumor cells often relied upon glycolysis as a means to
produce energy. Since these early observations we have come to learn that
aerobic glycolysis provides cancer cells with several growth advantages. These
advantages sustain growth of cells in adverse microenvironments, provide for the
generation of substrates for glycosylation reactions, and supply precursors for
biosynthetic reactions (RA Gatenby et al., 2006; DeBerardinis et al., 2008).
Moreover metabolism independent functions of the glycolytic enzymes provide
additional benefits such as resistance to apoptosis, invasiveness, malignancy
and transcriptional regulation (Kim and Dang, 2005). In addition to aerobic
glycolysis, the malignant phenotype is characterized by additional metabolic
29
changes in de novo lipid and nucleotide biosynthesis, and glutamine dependent
anapleurosis. These metabolic changes are necessary in order to produce the
biomass required for the rapid cell proliferation. The increased dependence on
glycolysis has been attributed to nuclear DNA mutations, abnormal expression of
metabolic enzymes (fumarate hydratase and succinate dehydrogenase),
adaptation to the tumor microenvironment (HIF-1α), and disruption in
oncogenic/tumor suppressor signaling (Ras, Akt, p53, pVHL).

Oncogenes act in the regulation of cancer cell metabolism
The PI3K/Akt/mTOR(PI3K, phosphoinositide 3-kinase; mTOR, mammalian target
of rapamycin) signaling axis, HIF-1α (Hypoxia-inducible factor-1α), and c-Myc
have been identified as key regulators of cancer cell metabolism. Activation of
the PI3K pathway increases glucose and nutrient uptake, and increases the
expression of enzymes involved in glycolysis and lipid synthesis. c-Myc is known
to regulate glutamine uptake, components of the transcription machinery and
protein synthesis. The activity of c-Myc modulates nucleotide, amino acid, fatty
acid and polyamine synthesis. The expression of HIF-1α inhibits translation
initiation in addition to induction of glycolysis. HIF-1α inhibits the entry of
glycolytic end products into the TCA cycle via its effect on pyruvate
dehydrogenase kinases (PDKs). The metabolic changes in cancer cells
complement tumor cell requirements for increased biogenesis and energy
production and an adaptive response to tumor microenvironment (DeBerardinis
et al., 2008).
30

Loss of tumor suppressor p53 and deregulation of metabolism
Tumor suppressor p53 affects mitochondrial respiration as well. p53 can
transactivate SCO2 (synthesis of cytochrome c oxidase2). SCO2 is required for
the assembly of the mitochondrial cytochrome c oxidase complex. Moreover p53
stimulates TIGAR (tp53 induced glycolysis and apoptosis regulator), which
decreases fructose-2,6-biphosphate and hence suppresses glycolysis
(Hatzivassiliou et al., 2005). Mutation and/or loss of p53 occurs in nearly half of
all neoplasms, therefore alterations in p53-mediated metabolic regulation are
likely to be a prevalent phenotype in most cancers.

The role of hypoxia inducible factors in altering cancer cell metabolism
The adaptive response to tumor hypoxia leads to stabilization of HIF-1 and/or
HIF-2. HIF activation leads to transcriptional activation of the genes for glucose
transporters, glycolytic enzymes, and angiogenic factors. HIF-1 may also be
elevated under normoxic conditions in tumor cells (Harris, 2002). Several factors
such as pVHL (von Hippel-Lindau protein) inactivation or Ras, Src, or PI3K/Akt
activation can lead to stabilization of HIF-1 under normoxic conditions.
Oncogenes can also directly activate glycolysis independent of HIF-1. c-Myc
binds the promoters of several glycolytic genes (Hexokinase II, Enolase1 and
Lactate dehydrogenase A), glucose transporters, and activates them even under
normoxia. Constitutive activation of Akt increases surface expression of the Glut1
glucose transporter. Akt also maintains the mitochondrial association of
31
hexokinase. This association prevents changes in the permeability of the outer
mitochondrial membrane and cytochrome c release upon apoptotic stimulation.
Akt phosphorylates BAD (Bcl-2 antagonist of cell death) and inhibits the
association of pro-apoptotic BAX (Bcl-2 associated protein X) and BAK (Bcl-2
homologous antagonist/killer) (Harris, 2002; Kim et al., 2005).

The other side of coin: biosynthetic drive
Tumor cells must increase their cellular biomass prior to cell division. The
metabolic changes that evolve during neoplastic transformation are necessary to
provide tumor cells with sufficient substrates for the biosynthetic machinery.
Moreover, oncogene products, tumor suppressor gene products and HIF-1
participate in the regulation of the enzymatic machinery and in doing so control
the availability of substrates for the biosynthesis of cellular macromolecules.
These biosynthetic precursors derived and/or diverted from the metabolism of
glucose and glutamine (DeBerardinis et al., 2008) are critical to the production of
nucleotides, fatty acids and proteins.

Biosynthesis of nucleotides
The synthesis of purines and pyrimidines requires ribose-5-phosphate, which is
derived from the oxidative and non-oxidative branches of the pentose phosphate
pathway. Some of the glycolytic intermediates feed the pentose phosphate
pathway leading to the synthesis of ribose-5-phosphate. Moreover the non-
32
essential amino acids obtained from glucose and glutamine metabolism are also
required for the synthesis of nucleotides (DeBerardinis et al., 2008a).

Oncogene and tumor suppressor gene products play an important role in
diverting glycolytic metabolites into the branches of the pentose phosphate
pathway. Several enzymes of the nucleotide biosynthetic pathway are the target
of c-Myc (Tong et al., 2009). TIGAR suppresses glycolysis by decreasing the
levels of PFK-1 and PGM. TIGAR decreases the expression of PFK1 activator
(fructose-2,6-biphosphate) and thus leads to the accumulation of fructose-6-
phosphate, which is retained for ribose-5-phosphate synthesis by the pentose
phosphate pathway (DeBerardinis et al., 2008a). c-Myc and Ras are also known
to activate PFK1(Tong et al., 2009). In the majority of tumors, pyruvate kinase-
M2 exists as a dimer (less active form of the enzyme), leading to accumulation of
upstream glycolytic intermediates that are subsequently used by the pentose
phosphate pathway (DeBerardinis et al., 2008a). Increased expression of HIF-1α
regulates the entry of glycolytic intermediates into the pentose phosphate
pathway. HIF-1α potentiates the expression of transketolase and pyruvate
kinase-M2, which promote the synthesis of ribose-5-phosphate via the pentose
phosphate pathway (Tong et al., 2009).

33
Biosynthesis of fatty acids
The biosynthesis of fatty acids is required for the formation of lipids which are
incorporated into the membranes of cells and organelles. Furthermore, these
lipids can also function to modify proteins destined to be associated with
membranes. Lipogenic enzymes are over-expressed in tumor cells. The
expression of ATP citrate lyase, a lipogenic enzyme, favors the Warburg effect
by preventing citrate build up in the cytosol (increased citrate can suppress
glycolysis).

Oncogenic mutations play an important role in fatty acid synthesis in tumor cells.
Activation of the PI3K/Akt pathway promotes the expression of lipogenic
enzymes, while suppressing the β-oxidation of fatty acids. Additionally the
PI3K/Akt pathway increases the expression of glucose transporters providing the
substrates for the reactions (DeBerardinis et al., 2008a).

Biosynthesis of proteins
Both glucose and glutamine metabolism are involved in generating amino acids,
tRNAs and ribosomes required for the protein synthesis. The ribose-5-phosphate
synthesized as a result of shunting of glycolytic intermediates to the pentose
phosphate pathway is used in the synthesis of nucleotides. These nucleotides
are then used as constituents of the protein synthesis machinery of the cells
(tRNA, ribosomes). Increased glutaminolysis also adds to the cellular pool of
ribose-5-phosphate and contributes to protein synthesis. Metabolism of glucose
34
and glutamine is involved in increasing cellular supply of amino acids for
synthesis of proteins (DeBerardinis et al., 2008a).

Anaplerosis and NADPH production: role of glutaminolysis in biosynthesis
The increased demand for biosynthetic precursors depletes the intermediates of
glycolysis and the TCA cycle. Citrate produced by TCA cycle is transferred out of
the mitochondria into the cytosol, where it is used in fatty acid synthesis. As the
need to replenish these metabolic intermediates arises, glutamine anapluerosis
plays a key role in compensating for the depletion of these metabolites.
Glutaminolysis provides a source of reducing power, supplying NADPH required
for nucleotide and fatty acid biosynthesis (DeBerardinis et al., 2008a). The
uptake of glutamine by tumor cells is regulated by c-Myc. It also induces the
expression and activity of enzymes involved in biosynthetic processes (Tong et
al., 2009).

The role of mitochondria in cancer cell metabolism
The mitochondrion is the powerhouse of the cell and serves as the major energy
source. Mitochondria are implicated in the regulation of programmed cell death
(PCD, the intrinsic pathway of apoptosis), reactive oxygen species (ROS)
generation, and maintenance of calcium homeostasis (Babcock et al., 1997;
D.Gutterman, 2005; Wang, 2001). Mitochondria play an important role in cell
survival and cell death, and dysregulation of any form leads to diseases.
Mitochondrial dysfunction has been implicated in neurodegenerative and
35
neuromuscular disorders, ischemia-reperfusion injury, diabetes, obesity, inherited
mitochondrial diseases and most importantly, cancer (Chan, 2006; Kurt Højlund
et al., 2008; Wallace, 2005).

Metabolic reprogramming consistent with a proliferative phenotype has been
noted for nearly all types of cancers including breast, pancreatic and colorectal
cancers. Agents that target tumor cell mitochondria with high selectivity hold
clinical significance due to the adaptive and modulatory role of this organelle in
cancer cell energy production, metabolism and apoptosis. Mitochondria carry out
critical functions that determine cell fate, including energy production,
biosynthesis, and commitment to apoptosis. During the process of tumorigenesis,
oncogene-driven reprogramming of metabolic function occurs to meet the
massive biosynthetic demands of deregulated replication. Contrary to widely held
belief, mitochondrial function remains critical for survival during the shift to a
proliferative phenotype. Under selective pressures and by varied mechanisms,
mitochondria adapt their function to increase production of biosynthetic
intermediates while decreasing ATP production. As a result, cancer cell
mitochondria are markedly different than their normal counter parts (Figure 1.4).
These differences include alterations in mitochondrial DNA content, changes in
energy and nutrient production and utilization, and increased mitochondrial
membrane potential.

36

Figure 1.4 Metabolic reprogramming results in key differences between cancer and
normal cell mitochondria. In normal cells glucose is converted to pyruvate that subsequently
enters the TCA cycle to support energy production by oxidative phosphorylation. ATP
produced this way in the mitochondria is transported to the cytosol by the adenine nucleotide
transporter in conjunction with cyclophilin D and the voltage dependent ion channel where it is
utilized for energy-requiring biologic functions. 1) In cancer cell mitochondria hexokinase II is
over-expressed and bound tightly to VDAC, where it co-opts ATP produced by OXPHOS for to
fuel glycolytic reactions that divert pyruvate away from TCA cycle. 2) Transforming mutations
in KRAS cause decrease in Complex I expression to compromise respiratory chain activity
decreasing ATP production by mitochondria. 3) Cancer cell mitochondria have increased
mitochondrial membrane potential compared to normal cells. 4) Citrate produced in TCA
reactions is co-opted for nucleotide biosynthesis. As a compensatory mechanism, cancer cells
increase glutamine uptake and anapleurosis.

The metabolic changes that accompany neoplastic transformation offer unique
and as of yet, not fully realized opportunities for drug development. To this end,
this dissertation describes recent efforts in the design, discovery and
37
development of novel small molecules targeting cancer cell mitochondria and the
glycolytic enzyme PKM2.

38
Chapter 2
Preclinical evaluation of novel triphenylphosphonium salts
with broad-spectrum activity

Introduction

Phosphonium salts have broad utility, with applications in chemistry, biology and
pharmacology. Triphenylphosphine can easily react with alcohols, alkyl halides,
and carboxylic acids giving rise to a large variety of chemical entities, which
supports their wide applicability. Initially, phosphonium salts were used in
preparation of phosphorus ylides, an essential component in the Wittig method of
alkene synthesis (Wittig, 1980). As a reagent for biological research, the
lipophilic, cationic properties of tetraphenylphosphonium were first utilized to
demonstrate the existence of electrochemical potential across the mitochondrial
membrane (Grinius et al., 1970). Charged molecules are generally unable to
traverse cell membranes without the aid of transporter proteins due to the large
activation energies associated with removal of associated water molecules. The
distribution of charge across the large lipophilic surface of the phosphonium ion
significantly lowers this energy requirement facilitating passage across lipid
membranes (Murphy 2008). Thus phosphonium salts accumulate in energized
mitochondria due to their highly negative membrane potential. Based on this
observation, the triphenylphosphonium ion has been used as a targeting moiety
for delivery of agents such as spin traps, fluorescent dyes, and antioxidants to
isolated mitochondria, as well as the mitochondria of intact cells and whole
39
organisms. As pharmacological agents, certain phosphonium salts have
demonstrated anti-microbial activity against gram negative and positive bacteria
and the parasite T. cruzi., antiglycemic properties in animal models and anti-
proliferative activity in cell- and animal-based systems (Bergeron et al., 2009;
Blank et al., 1975; Kinnamon et al., 1977; Rideout et al., 1989). As anti-cancer
agents, phosphonium salts show great promise for the diagnosis and treatment
of neoplasms. Increased mitochondrial membrane potential is a unique property
that distinguishes cancer cell mitochondria, offering opportunity for selective
targeting of the most malignant cells within a tumor mass (Chen, 1988b; Heerdt
et al., 2005b, 2006b). Delocalized lipophilic cations (DLC) which include
compounds bearing the triphenylphosphonium ion are ideal agents for this type
of targeting since they accrue selectively in the mitochondria of experimental
tumors based on the differences in mitochondrial membrane potential between
normal and transformed cells (Figure 2.1) (Modica-Napolitano and Aprille, 2001).

Figure 2.1 Triphenylphosphonium cation
accrues in mitochondria based on
mitochondrial membrane potential.

The mitochondria-specific character of TPP cations is well-studied in model
systems ranging from isolated mitochondria, to cell cultures and even whole
organisms such as mice, rats and humans (Agapova et al., 2008; Asin-Cayuela
40
et al., 2004; Dubois et al., 1978b; Kim et al., 2008b; Li et al., 2009; Madar et al.,
2002; Manetta et al., 1996b; Millard et al., 2010a; Min et al., 2004; Porteous et
al., 2010; Ross et al., 2006; Ross et al., 2005b; Ross et al., 2008; Wang et al.,
2007b; Yang et al., 2008; Zhou and Liu, 2011). This unique property results from
the distribution of the cationic charge across the planar, lipophilic phenyl rings
which decreases the activation energy required to pass through the hydrophobic
regions of the plasma and mitochondrial membranes. Uptake is dependent upon
mitochondrial membrane potential, without the need for ion transport proteins
and thus follows the Nernst equation such that a 10-fold increase in TPP
concentration is noted for every 60 mV difference in membrane potential (Ross et
al., 2005b). At the cellular level it has been established that cationic charge
provides the driving force while lipophilicity is the primary determinant of the
kinetics of mitochondrial uptake. Thus compounds with greater lipophilicity
accumulate more rapidly in mitochondria compared to more hydrophilic TPP
analogs (Ross et al., 2008).

The first report of a TPP salt having anti-proliferative activity was published in
1978 following routine screens of synthetic intermediates.(Dubois et al., 1978a)
In these screens, isoindolylalkyl phosphonium salts showed potent anti-leukemic
activity. More recently, some phosphonium salts have been reported to show
anti-proliferative activity in several cancer cell lines and a xenograft model of
ovarian cancer based on their ability to disrupt mitochondrial ultrastructure and
alter cellular lipid content (Cooper et al., 2001; Manetta et al., 1996a; Rideout et
41
al., 1994; Rideout et al., 1989). Studies of phosphonium salts as contrast agents
for diagnostic imaging (Kim et al., 2008a; Wang et al., 2007a) have elucidated
two key points concerning the selectivity of this class of compounds: 1) these
agents are capable of preferentially accumulating within tumor cells, 2) that
phosphonium cation itself does not impart cytotoxicity.

When developing agents targeting cancer cell mitochondria a primary concern is
that organs with higher densities of mitochondria such as heart, muscle and liver
may through a similar mechanism accumulate sufficient drug to cause toxicity.
For this reason, tumor selectivity is of great importance. The experimental basis
for tumor cell selectivity of the DLC, Rhodamine 123 and several close analogs
has been demonstrated to be related to the overall lipophilicity of the compound
(Belostotsky et al., 2011). Compared to analogs with increased LogP,
Rhodamine 123 and its hydrophilic analogs showed greater selectivity for cancer
cells in vitro. Studies of TPP cations as in vivo contrast agents for PET imaging
provide valuable insight for the development of anti-cancer agents highly specific
for tumor cell mitochondria as the background caused by accumulation within
mitochondria-dense organs of the chest cavity limits probe sensitivity. To achieve
increased tumor specificity and retention times for use in molecular PET imaging
studies, TPP cations bearing substituents of a more hydrophilic nature have been
employed with good results (Kim et al., 2008b; Li et al., 2009; Wang et al.,
2007b). The rationale for this approach is rooted in the knowledge that tumor
cells may at times have decreased mitochondrial density but in general maintain
42
increased mitochondrial membrane potential compared to normal tissue. Much
like in cell-based assays, the kinetics of cellular uptake in vivo appears to be a
function of the lipophilicity. Similarly, TPP cation uptake is driven by
mitochondrial membrane potential but compounds with greater lipophilicity will
target rapidly to mitochondria-dense organs while those of more hydrophilic
character will have slower kinetics of uptake, circulate longer and accumulate on
the basis of increased mitochondrial membrane potential (Zhou and Liu, 2011).
Thus, modulation of lipophilicity is a simple and effective means by which to tailor
the pharmacokinetic properties of TPP cations.

In vivo, TPP cations have been explored most extensively as delivery agents for
the ubquinone–derived antioxidant, Mito-Q10 and as tumor-specific PET imaging
probes. To this end, administration via intravenous, intraperitoneal and oral
routes resulted in pharmacologically significant accumulation within the
mitochondria of organs and tumors (Kim et al., 2008b; Li et al., 2009; Madar et
al., 2002; Min et al., 2004; Porteous et al., 2010; Smith et al., 2003; Wang et al.,
2007b; Yang et al., 2008; Zhou and Liu, 2011). In non-tumor bearing mice and
rats, uptake of Mito-Q10 into organs was rapid, disappearing from the blood
within 5 minutes of administration (Porteous et al., 2010; Smith et al., 2003). The
extent of distribution varied by tissue type with the greatest accumulation in the
kidney and liver and to a much lower extent in the heart, muscle, fat and brain
(Porteous et al., 2010; Smith et al., 2003). Upon equilibration with the
extracellular space, compound is steadily released from the mitochondria as
43
clearance from the circulation progresses, usually within 2-15 hours. Steady state
levels were achievable within 5 days through oral administration of Mito-Q10
(Porteous et al., 2010). Pharmacodynamic studies demonstrated no significant
modifications to the TPP moiety, although modification by glucorinidation and
sulfation on the substituents were noted (Porteous et al., 2010; Smith et al.,
2003). Several clinical trials are indicative of the potential for clinical translation of
phosphonium salts as therapeutic agents. In a Phase II trial, the DLC, MKT-077
was successfully targeted to tumor cell mitochondria, providing strong proof-of-
concept evidence supporting translation of this biological observation to the
clinical realm (Propper et al., 1999). In the PROTECT and CLEAR clinical trials,
MitoQ, a triphenlyphosphonium derivative of ubiquinone was found to be
amenable to oral formulation and dosing with bioavalability around 10% and peak
plasma conentration of 35 ng/mL at one hour following 80 mg dose (Gane et al.,
2010; Snow et al., 2010).

Delivery of small molecule inhibitors to the inner matrix of cancer cell
mitochondria using the TPP ion is advantageous because: a) it achieves
selectivity over non-targeted compounds. b) through conjugation to relevant
mitochondrial effectors it can circumvent upstream activators of the intrinsic
pathway of apoptosis which are complex and often dysfunctional in tumors. c)
uptake is rapid and doesn’t require channels or drug transporters which are
subject to saturation, drug interactions, genetic heterogeneity among patient
44
populations and are frequently mutated under selection for drug resistance. d)
phosphonium salts can easily be synthesized in large quantity in 4 or less steps.

Recently our laboratory discovered a series of novel compounds containing a
triphenylphosphine moiety that showed remarkable activity in a panel of cancer
cell lines. Two of these compounds were tested in a mouse xenograft model and
showed significant in vivo efficacy with no apparent toxicity. Further molecular
characterization of these compounds in cell-based models suggest a mechanism
of action that includes mitochondrial localization, decreased oxygen
consumption, increased superoxide production and attenuated growth factor
signaling.

Results

TP compounds are cytotoxic in a panel of human cancer cell lines

We have in place a highly diverse library of small molecules comprised of
approximately 40,000 chemical entities. For the present study, a subset of
10,000 compounds predicted in silico to have favorable drug-like properties was
selected for in vitro studies. Initial screening to identify active compounds was
performed using a high-throughput, 96-well format MTT-based cytotoxicity assay

45

Table 2.1 IC 50 values of TP187 and TP197 in MTT assay
IC
50
, µM± S.D.
cancer origin cell line TP187 TP197
breast MDA MB 435 0.8 ± 0.1 0.8 ± 0.1
MCF-7 0.8 ± 0.1 1.3 ± 0.1
MDA MB 231 1.7 ± 0.3 3.7 ± 4.3
MDA MB 468 0.1 ± 0.1 0.5 ± 0.3
CAMA-1 4.2 ± 5.1 1.7± 1.2
HS578T 2.9 ± 2.2 4.2 ± 3.4
SKBR-3 1.2 ± 0.9 1.5 ± 0.7
BT549 0.9 ± 0.2 0.6 ± 0.2
BT-474 3.4 ± 0.9 1.6 ± 1.0
MDA MB 361 12.0 ± 0.1 0.8 ± 0.1
prostate PC-3 1.7 ± 0.1 0.7 ± 0.1
DU145 11.0 ± 0.1 1.2 ± 0.1
ovarian OVCAR 8 0.1 1.1 ± 1.2
NC ADR-RES >20 ± 0.1 2.0 ± 0.1
HEY 0.1 0.1
colon HCT116 p53 +/+ 0.4 ± 0.1 0.4 ± 0.1
HCT116 p53 -/- 1.3 ± 0.1 1.3 ± 0.1
pancreatic MIA PaCa-2 1.2 ± 0.5 0.2 ± 0.2
Panc-1 4.5 ± 1.1 1.7 ± 1.1
BxPC3 2.3 ± 0.8 1.1 ± 0.4
glioblastoma U87MG 1.5 ± 0.6 0.6 ± 0.6
lung HOP92 6.0 ± 0.1 1.5 ± 0.1
primary
endothelial
HUVEC 10.5 ± 3.5 5.8 ± 4.6

in a panel of cancer cell lines of varied origin. This screening method identified
lead compounds, TP187 and TP197, having cytotoxicity values in the low
micromolar range. IC
50
values obtained in MTT assay are listed in Table 2.1.
These results were further confirmed in colony formation assay using HCT116
p53 +/+ cells treated with increasing concentrations of TP187, 197 and the close
analogue TP421 (Figures 2.2 and 2.3). All three TP compounds exhibited IC
50

values in the low micromolar range across most cell lines tested in MTT as well
as in HCT116 p53 +/+ colony assays and were, therefore, selected for further
analysis in cell- and animal-based models.
46

Figure 2.2 Chemical structures of lead TP compounds

Figure 2.3 Colony formation assay HCT116
p53 +/+ treated with TP compounds 24 h
and allowed 10 days to form colonies.

A B C
Figure 2.4 TP compounds arrest cell cycle progression. Distribution of DNA content in (A)
MDA MB 435, (B) HCT116 p53 +/+ and (C) HCT116 p53 -/- cells treated 24-72 h with 1 µM
TP187.
47
TP compounds arrest cell cycle progression in human cancer cell lines

Flow cytometry was performed on ethanol-fixed propidium iodide stained tumor
cell lines treated with 1µM TP187 for 24-72 h to investigate the effect of TP
compounds on cell cycle progression and DNA content. TP187 arrested cell
cycle progression in all cell lines, starting at 24 h treatment. (Figure 2.4 A-C)
MDA-MB-435, a p53 mutant cell line, arrested in the S-Phase of the cell cycle in
response to treatment with TP187 (Figure 2.4-A). HCT116 p53 +/+ (Figure 2.4-
B) and HCT116 p53 -/- (Figure 2.4-C), p53 competent and deficient cell lines,
respectively, both exhibited cell cycle arrest in the G2/M phase of the cell cycle
when treated with TP187. These effects on cell cycle progression were sustained
through the 72 h timepoint. Based on these results, the mechanism of action was
concluded to be independent of p53 status.

TP compounds inhibit tumor growth in vivo

Next, we tested the TP analogues 187, 197 and 449 in a nude mouse xenograft
model to determine the in vivo efficacy of these compounds. Treatment of MDA-
MB-435 xenografts with TP compounds as single agents significantly inhibited
tumor growth in both TP187 and TP197 treatment groups compared to vehicle
controls. (Figure 2.5-A) Starting at day 13, significant differences in tumor volume
were noted between TP187 and 197 treated mice compared to controls. (TP187;
p < 0.05, TP197; p < 0.05, Figure 2.5-B) The suppressive effects of TP187 and
48

A

D
B

C

E

Figure 2.5 TP compounds are effective as single agents in delaying tumor growth. (A)
Mean tumor volumes ± SEM of TP187 and TP197 compared to vehicle or paclitaxel treated
mice. Treatments were administered intraperitoneally in 5% DMSO/ 95% peanut oil at 10
mg/kg over a 33-day period. (B) p-values at end of treatment. (C) Mean weights in control and
treatment groups does not reveal any significant differences between treated and untreated
mice (D) H&E stained tissue sections from vehicle and TP-treated mice show no overt sigh of
drug-related toxicity following 33-day treatment schedule. (E) IHC stained tumor sections
showing decrease of Ki-67 positive cells in TP187 treated tumor indicative of proliferative
arrest. Cleavage of caspase 3 was also observed in tumor sections of TP187 treated mice.
49
197 were maintained throughout the course of treatment (at day 33 TP187; p =
0.003 and TP197; p = 0.04). Upon conclusion of treatment at day 33, the
average tumor volume of vehicle treated tumors increased by 1149%, whereas
the average tumor volume in TP187 and 197 treated groups increased less than
373 and 573%, respectively (Figure 2.5 A & B).

TP treated mice exhibit no sign of drug related toxicity

Mice were monitored daily by visual inspection and weighed three times per
week to detect symptoms of drug related toxicity. TP-treated mice showed no
outward signs of drug related toxicity such as malaise, weakness or lethargy
throughout the entire 33-day treatment course. TP197 was initially administered
at similar dosing but due to slight decrease in weight (<10% of body weight)
within the first week, treatment was modified to dosing at 10 mg/kg, three times
weekly. Soon after the dose adjustment TP197 treated mice resumed gaining
weight, and by the end of the experiment weights were similar to that of the
controls (Figure 2.5-C). Despite the reduction in number of doses of drug
administered, TP197 retained its efficacy in suppression of tumor growth.

In addition to daily health checks, organ samples were collected at the time of
sacrifice (day 33) to evaluate for potential drug related toxicities at the cellular
level. Tissue samples derived from brain, heart, lung, liver, kidney, pancreas and
spleen were harvested from mice in all treatment groups. These samples were
50
subsequently fixed in 10% neutral buffered formalin, paraffin embedded and
stained with H&E for histological analysis. Careful examination of these tissue
sections demonstrated no histological abnormalities were present in tissue
samples taken from TP-treated compared to control mice. Representative
micrographs comparing tissue sections of liver, kidney, brain and spleen taken
from control and TP187 treated mice are presented in Figure 2.5-D.

TP187 decreases the number of proliferating cells and induces caspase-3
cleavage in tumor xenografts

The suppression of tumor growth in response to TP187 treatment prompted us to
examine tumor sections for histological markers that could validate our in vivo
results. Immunohistochemical staining (IHC) was performed on formalin-fixed,
paraffin-embedded tumor sections collected from vehicle- and TP187-treated
treated mice to evaluate the effect of treatment on cell proliferation and
apoptosis. Tumor sections were processed and probed using antibodies against
Ki-67 and cleaved caspase-3.

Ki-67 is a cell-cycle related protein. Its presence solely in actively cycling cells
makes it an ideal marker to identify the fraction of proliferating cells within a
tissue sample. Ki-67 expression is increased in rapidly dividing cell populations
and the degree and intensity of Ki-67 staining can be correlated with prognosis
and response to treatment in many solid tumors.(Isola et al., 1990; Veronese et
51
al.,1993). Ki67 expression was nearly absent in all TP187 treated tumor sections.
A representative micrograph of the decrease in Ki-67 staining is shown in Figure
2.5-E. This result correlates well with the observed suppression of tumor growth
in vivo.

Next we sought to determine whether our compound could induce cell death in
vivo. Caspases are cysteine-aspartate specific proteases that, upon cleavage by
upstream proteases, function in apoptotic cell death pathways. Caspase 3 is a
downstream executioner caspase common to both the intrinsic and extrinsic
pathways of apoptosis. Cleavage of caspase 3 is a late and irreversible event in
the process of apoptosis and therefore serves as a marker for both major
apoptotic pathways leading to cell death (Li and Yuan, 2008). Compared to
vehicle-treated, TP187-treated tumor sections showed increased cytoplasmic
staining with cleaved caspase 3 antibody (Figure 2.5-E). Our
immunohistochemical studies showed a marked decrease in Ki-67 staining and
an increase in caspase 3 cleavage in response to treatment with TP compounds.
Taken together, these results suggest a mechanism, acting on effectors of cell
proliferation and death pathways that together, are capable of suppressing tumor
growth in vivo.

52
TP compounds localize to the mitochondria

The favorable results obtained in vivo prompted us to further characterize the
molecular mechanisms of TP mediated tumor suppression and identify potential
targets of TP action. To this end, we chose to first investigate the sub-cellular
localization of TP compounds in an effort to narrow the range of possible targets.
The triphenylphosphonium moiety common to the TP compounds imparts a
delocalized charge and lipophilic character that favors mitochondrial
accumulation (Murphy 2004; Ross et al., 2005a). Therefore, we tested whether
our compounds could accumulate preferentially in the mitochondria by exploiting
the fluorescent properties of the analog TP421 (Figure 2.6-A). We performed
fluorescence spectroscopy to determine the optimal excitation and emission
wavelengths using a steady state spectrofluorimeter. TP421 has an optimal
excitation wavelength of 396 nm. Excitation peaks of nearly similar intensity were
observed at 450 nm and 573 nm (Figure 2.6-B) with 450 nm having slightly
higher peak intensity.

Based on the excitation / emission spectra we were able to measure the uptake
of TP421 using fluorescence-based assays. MDA-MB-435 cells were treated with
53

Figure 2.6 TP compounds rapidly accumulate in energized mitochondria. (A) Coumarin
moiety (blue) imparts fluorescent properties to TP421 (B) TP421 Fluorescence emission
spectra, inset excitation spectra (C) Histogram TP421 fluorescence versus time, shift in
fluorescence follows addition of 5 µM TP421 to cells. (D) TP421 co-localizes with MitoTracker
Red, a fluorescent dye specific for mitochondria.

5µM TP421 then analyzed by flow cytometry using a 355 nm UV-laser as the
excitation source and optical filters capable of capturing emission in the 450 nm
range. MDA-MB-435 cells treated with 5 µM TP421 displayed a sharp increase
in fluorescence intensity (Figure 2.6-C) compared to cells treated with a
comparable volume of DMSO. The intensity of fluorescence increased
immediately upon addition of TP421, leveling off to a steady state with 15
minutes suggesting that TP uptake is rapid and likely due to the lipophilic nature
and delocalized charge of the triphenylphosphonium moiety. The sub-cellular
localization of TP421 has been examined in breast and pancreatic cancer cell
54
lines using fluorescent microscopy. In agreement with the results of our flow
cytometry experiments we found that TP421 was rapidly taken up into the cell. At
the earliest time point observed (5 minutes, data not shown) TP421 was visible
within the cytosolic compartments of cells. Co-staining with a mitochondrial
specific dye, Mitotracker Red (MTR) revealed similar staining pattern (Figure 2.6-
D). Composite images of the micrographs of TP421 and MTR fluorescence
showed significant overlap in staining. These results taken together indicate that
TP421 and likely, the close analogs, TP187 and TP197 do indeed enter the
mitochondria.

TP compounds increase mitochondrial superoxide production

Energy production through oxidative phosphorylation is the primary function of
the mitochondria. TP compounds decrease the cellular rate of oxygen

A B
Figure 2.7. TP compounds inhibit mitochondrial function. (A) Treatment of MDA MB 435
cells with 1 µM TP197 caused rapid and sustained production of superoxide ion in
mitochondria as measured using the mitochondria specific superoxide probe, MitoSOX red. (B)
MIA PaCa-2 were stained with JC-1 dye and treated with 1 µM TP187 or TP197 for 6 and 24 h
exhibit decreased mitochondrial membrane potential.
55
consumption (Millard et al., 2010b). Inhibition of components of the oxidative
phosphorylation machinery is known to result in the increased production of
superoxide ion. The mitochondrial localization of TP compounds therefore, led us
to investigate superoxide production as a possible mechanism of action for our
compounds.

To examine the effect of TP treatment on mitochondrial superoxide production,
we treated MDA-MB-435 cells with 5 µM of TP197 for varying lengths of time. At
the end of treatment, cells were incubated with 5 µM of MitoSOX red and change
in fluorescence intensity corresponding to production of superoxide was
measured by flow cytometry. In order to rule out the possibility that our
observations could be the result of non-specific ROS production in response to
xenobiotic treatment, we also included MDA-MB-435 cells treated with paclitaxel.
Using this fluorogenic mitochondrial superoxide indicator, we found that TP197
treatment caused increased production of mitochondrial superoxide. MDA-MB-
435 cells were treated with 5 µM TP197 and collected at various time points
ranging from 10 minutes up to 24 hours. The surge in superoxide production was
observed at time points as early as 10 minutes post treatment, increased to
maximal intensity by 3 h post treatment and remained increased for 24 h
following treatment with TP197 (Figure 2.7-A). Similar results were obtained
measuring superoxide production in TP187 and TP421 treated MDA-MB-435
cells at 0.5 and 1, 6 and 24 h time points (data not shown). In contrast, paclitaxel
56
treatment had no effect on the production of mitochondrial superoxide, as
fluorescence intensity was unchanged compared to the vehicle control.

TP compounds decreased mitochondrial membrane potential

Mitochondrial membrane potential was measured in MIA PaCa-2 cells treated
with TP187 and TP197 for 6 and 24 h . Compared to vehicle treated cells TP
compounds caused a slight decrease in mitochondrial membrane potential at 6 h,
but required 24 h treatment before significant differences were noted (Figure 2.7-
B). Treatment with gemcitabine had no effect on mitochondrial membrane
potential.
Discussion

The tumor microenvironment is characterized by alterations in oxygen and
nutrient content based, in part, on the inadequacies of tumor vasculature. The
tortuous, disorganized tumor vascular bed produces hypoxic, nutrient poor
regions within growing tumors. Survival in these sub-optimal conditions requires
tumor cells invoke adaptive strategies to circumvent nutrient deprivation and
hypoxia. Alterations in energy metabolic pathways and the adaptive responses to
hypoxia and nutrient deprivation are emerging as hallmarks of cellular
transformation (Pathania et al., 2009b). Midway into the 20
th
century, Warburg
postulated that even with adequate oxygenation, cancer cells rely on glycolytic
metabolism as an adaptive mechanism to compensate for dysfunction at the
57
level of mitochondrial respiratory chain (Warburg, 1956a). More recent
knowledge demonstrates results to the contrary in some tumor types suggesting
the Warburg effect may not apply to all cancers. Poor tumor cell perfusion limits
glucose supply considerably making glycolysis energetically unsustainable.
Measured oxygen levels in hypoxic tumor tissues are higher than that at which
respiratory dysfunction is thought to occur, precluding hypoxia as a cause of
mitochondrial dysfunction (Moreno-Sánchez et al., 2007). Through examining the
contributions of each metabolic pathway to cell proliferation it has been observed
that oxidative phosphorylation supplies a large portion of the ATP produced in
many tumor types, particularly in the absence of glucose. Furthermore, cell-
based studies demonstrate an increase in cancer cells’ oxidative phosphorylation
capacity and oxygen affinity upon prolonged culture in glucose free media and
the requirement of functional mitochondria to activate survival mechanisms such
as the unfolded protein response (UPR) pathway under glucose starved
conditions (Haga et al., 2010; Smolková et al., 2010). All of which are not
observed in non-transformed cell lines. Tumors exhibit additional metabolic
changes, including de novo lipid and nucleotide biosynthesis and glutamine
dependent anapleurosis. These alterations allow for growth under adverse
conditions, generation of substrates for glycosylation reactions, and supply of
precursors for biosynthetic reactions (Gatenby et al., 2006; DeBerardinis et al.,
2008). Often, TCA cycle intermediates are co-opted as precursors for the
biosynthetic reactions. Citrate derived from the TCA cycle is utilized as a
precursor for fatty acid synthesis. Oxaloacteate and α-ketoglutarate provide
58
nonessential amino acids required for protein and nucleotide synthesis. To
compensate, cancer cells have higher uptake of glutamine for replenishing TCA
cycle intermediates through glutaminolysis (DeBerardinis et al., 2008b). Taken
together, these evidence suggest that drugs targeting the mitochondrial
respiratory capacity should have a profound effect on tumor growth while sparing
normal cells.

Herein we have identified a series novel triphenylphosphine based compounds
exhibiting potent anti-tumor activity. Our preliminary screening of TP analogues,
TP187, TP197 and TP421 demonstrated cytotoxicity in a panel of cancer cell
lines of varied origin and cytogenetic attributes, inhibition of colony formation at
sub-micromolar concentrations and cell cycle arrest independent of p53
competence. TP analogues 187 and 197 significantly suppressed the growth
and proliferation of MDA-MB-435 tumors in a mouse xenograft at clinically
achievable dosing. Daily health monitoring and post-mortem histology revealed
no detectable systemic toxicities or drug related tissue injury. Collectively, these
results demonstrate the potential clinical utility of TP compounds in the treatment
of a wide range of cancer types.

Using the fluorescent analog TP421 we were able to confirm rapid uptake and
mitochondrial localization as is expected for compounds containing the
triphenylphosphine moiety. Early events in TP induced apoptosis included an
increase in superoxide production and reduction in mitochondrial membrane
59
potential. TP treatment of MDA-MB-435 cells caused an immediate and
sustained It is likely these early events contribute significantly, if not wholly, to the
TP mechanism of action.

Superoxide produced in the mitochondria in response to inhibition of oxygen
consumption can be converted to hydrogen peroxide, hydroxyl radical or in the
presence of nitric oxide, peroxynitrite.(Chance and Williams, 1956; Deby and
Goutier, 1990; Fridovich, 1978; Steinbeck et al., 1993) These ROS derivatives
can diffuse or be transported out of the mitochondria to enter the cytosolic and
nuclear compartments. ROS can react with thiols of cysteine and methionine
residues of proteins, in both the cytosol and nucleus, causing either intra- or
extracellular disulfide linkages. Disulfide linkages modify protein structure, which
can directly impact function thorough alterations in protein activity, protein/
protein associations and sub-cellular localization. In the nucleus, ROS can also
act to inhibit transcription factors by altering their redox status while higher
concentrations of ROS have been shown to induce oxidative DNA damage
(Cromlish and Roeder, 1989; Holmgren, 1985).

The cellular response to ROS production is dependent on the cellular redox
buffering potential as well as the degree and duration of ROS production. Abrupt,
intense ROS over production may overwhelm cellular antioxidant responses
causing irreversible oxidative damage to cellular proteins, lipids and DNA leading
to oxidative stress and ultimately cell necrosis. Lower levels of ROS production,
60
on the other hand, produce milder oxidative imbalances that via redox mediated
regulation of signal transduction result in programmed responses such as cell
proliferation, senescence and apoptosis (Filomeni et al., 2005; Holmgren, 1985).

Based on this knowledge, it is plausible that TP compounds, through selective
inhibition of tumor cell oxygen consumption and subsequent superoxide
generation in the mitochondria, contribute to sustained, low levels of ROS
production. The sustained levels of ROS eventually create an imbalance in the
cellular redox status that causes modulation in the function of proteins involved in
cell cycle regulation, growth factor signaling and DNA transcription resulting in
the global protein expression profile observed in the Kinexus data. Dysfunction in
these molecular effectors impinges on pathways critical to cancer cell survival,
proliferation and interaction with the extracellular environment leading to the
induction of apoptosis.
Experimental Procedures

Cell culture
MDA-MB-435 breast, MCF7 breast cancer, Panc-1 pancreatic ductal
adenocarcinoma and PC3 prostate cancer cell lines were purchased from the
American Type Cell Culture. HCT116 p53+/+ and HCT116 p53-/- cells were
kindly provided by Dr. Bert Vogelstein. Cell lines were maintained in the
appropriate growth media (DMEM for MDA-MB-435, MCF7 and PC3, RPMI for
the Panc-1 and HCT116 cell lines) containing 10% heat-inactivated fetal bovine
61
serum and supplemented with 2 mM L-gutamine at 37º C in a humidified
atmosphere of 5% CO2. For subculture and experiments cells were washed with
1x PBS, detached using 0.025% Trypsin-EDTA, collected in growth media and
centrifuged. All experiments were performed in growth media using sub-confluent
cells in the exponential growth phase. For use in tissue culture experiments,
compounds were prepared at 10 mM concentration in sterile dimethylsulfoxide
(DMSO) and stored at -20ºC when not in use.

Cytotoxicity assay
Cytotoxicity was assessed by a 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide (MTT) assay as previously described.(Carmichael et
al., 1987a) Cells were seeded in 96-well tissue culture treated dishes and
allowed to adhere overnight. Cells were subsequently treated with a continuous
exposure to drugs for 72 hours. An MTT solution was added to each well to give
a final concentration of 0.3 mg/mL MTT. Cells were incubated with MTT for 3-4
hours at 37°C. After removal of the supernatant, DMSO was added and the
absorbance was read at 570 nm. All assays were done in triplicate. The IC
50
was
then determined for each drug from a plot of log drug concentration versus
percentage of cell kill.

Colony formation assay
Colony formation assay was performed as previously described (Munshi et al.,
2005) to further assess drug toxicity. To this end, cells were seeded in 96 well
62
tissue culture dishes at a density of 200 cells per well in growth media and
allowed to adhere overnight. Cells were subsequently treated with varying
concentrations of compound for 24 h. Following treatment, monolayers were
washed with 1x PBS and incubated in growth media for a period of 7-10 days,
allowing sufficient time for colonies to form in control wells. To visualize the
extent of colony formation, cells were fixed and stained in a 2% solution of crystal
violet containing 1% glutaraldehyde. Excess stain was removed through multiple
washes in distilled water and allowed to air dry. Stained plates were imaged
using Quantity One software running on the VersaDoc imaging platform.
Quantification by measurement of optical density at 570 nm was performed after
solublization in a 2% solution of sodium dodecyl sulfate accompanied by 2 h
shaking on a platform rocker.

Cell cycle analysis
Cells were seeded in 100 mm tissue culture dishes at a density of 1x10
6

cells/plate in growth media and allowed to adhere overnight. The following day
cells were treated with IC
80
concentrations of TP compounds or DMSO alone as
vehicle control for 24–72 h. Upon completion of treatment, cells were detached
with trypsin and both media and cells were collected by centrifugation. Cells were
washed and resuspended in 1x PBS prior to fixation in ethanol overnight at -20
ºC. Fixed cells were treated with 10 µg/mL RNase A and stained in a 50 µg/mL
solution of propidium iodide. DNA content was determined by flow cytometry
63
using the BD LSR II equipped with a 488 nM Sapphire™ argon-ion laser and PE
emission detector.

Xenograft studies
1.5x10
6
MDA-MB435 cells were implanted subcutaneously into the rear flank of
6-week old female nu/nu mice. When tumor volumes reached 100 mm
3
, mice
were separated into one of six treatment groups consisting of four to eight mice
per group. Treatments were administered by intraperitoneal injection in a 50 µL
suspension of 5% DMSO/ 95% peanut oil (v/v). Group 1 (n=8) received vehicle
control of DMSO/peanut oil. Group 2 (n=4) received 10 mg/kg body weight
paclitaxel every other day for a total of seven doses. Group 3 (n=8) received 10
mg/kg body weight TP187, five times weekly. Group 4 (n=8) received 10 mg/kg
body weight TP197 every other day. Tumor volumes and weights were
measured three times weekly. Tumors were measured using calipers. Tumor
volume was calculated using the following equation: V= d
2
x D/2 , where d = the
width or smaller measure and D= the length or larger measure. Data collected
was plotted and analyzed to determine average tumor volumes and weights,
SEM values, and p-values using Microsoft Excel. Health checks were performed
daily. Mice exhibiting toxicities or excessive tumor burden (> 2.0 cm
3
) were
sacrificed using CO
2
gas, necropsies were performed and tumor samples and
organs were harvested and fixed in 10% neutral buffered formalin prior to
processing for histological analysis. Upon completion of the experiment
remaining mice were sacrificed and following necropsy, the organs harvested,
64
fixed and processed for histological analysis. The animal studies were approved
by the USC Animal Care and Use Committee under protocol numbers 10766 and
11458. Animal care and manipulation were in agreement with the USC
institutional guidelines, which were in accordance with the Guidelines for the
Care and Use of Laboratory Animals.

Tissue handling
Surgically excised tissues or organs were washed in 1x PBS to remove blood
and bodily fluids prior to fixation in 10% neutral buffered formalin. Samples were
fixed for 24-48 hour after which time the organs were stored in 1x PBS until
ready to process for analysis. Fixed samples were placed in cassettes and
processed for histological analysis using the Microm STP 120 spin tissue
processor. At the completion of the processing, tissues/organs were embedded
in molds containing hot paraffin and allowed to solidify on the Microm EC 350-2
refrigerated cooling tray. Paraffin blocks were cooled on ice and 4-micron
sections were cut using a Microm HM 310 microtome. Paraffin embedded
sections were floated in a Thermo Scientific Tissue Flotation bath filled with water
heated to 44 °C prior to mounting on pre-cleaned, positively-charged glass
slides. Slides were placed upright and allowed sufficient time to air dry.

Hematoxylin & eosin staining
H&E staining was performed using Thermo Scientific Shandon Varistain Gemini
automated staining system. Slides were deparaffinized in the heater block for 20
65
minutes. The program then continued with incubation of slides in 3 changes of
Clear-Rite 3 for three minutes followed by two changes of FLEX100 for one
minute each. The slides were then incubated in FLEX 95 for one minute before a
running water wash. After the water step, slides were stained with Hematoxylin
7211 for two minutes, thirty seconds followed by a one minute running water
wash. Next, the slides were incubated one minute with Clarifier 2 to remove
background hematoxylin staining. Clarifier 2 treatment was followed with a one-
minute running water wash prior to a one-minute incubation with bluing reagent.
After the bluing reagent, the slides were washed one minute in running water and
then incubated for thirty seconds in FLEX 95. The slides were then stained with
Eosin Y. Eosin Y staining was followed with three consecutive one minute
washes in 100% FLEX and finally three consecutive changes of Clear-rite 3. The
slides were then removed from the Gemini stainer and coverslipped using 1-2
drops of mounting media and air dried several hours. Specimens were examined
by light microscopy. Slides were visualized using a Ziess axioscope light
microscope equipped with 10 x eyepiece and 5, 20, 40 and 100 x objectives.
Light micrographs were obtained using Moticam 2300 microscope camera.

Immunoperoxidase staining of formalin-fixed paraffin-embedded tissue sections
Tissue sections four microns thick were mounted on pre-cleaned positively
charged glass slides. Tissue sections were deparaffinized using three changes of
xylenes for five minutes each. Sections were hydrated, first in two washes of
100% ethanol for 10 minutes each, then two washes in 95% ethanol for 10
66
minutes each followed by immersion in double distilled water (ddH
2
O) for one
minute. Antigen retrieval was performed by boiling slides for ten minutes in 10
mM sodium citrate pH 6.0. Immunohistochemical staining was performed using
the UltraVision ONE detection system according to the manufacturer’s protocol.
MDM2 (clone SMP14) was obtained from Biosource, Invitrogen, p53 antibody
(clone DO-1) was obtained from Santa Cruz Antibodies and used at dilutions of
1:500. Ki67 antibody (clone SP6) was obtained from Thermo Scientific and was
used at a dilution of 1:400. Anti-cleaved Caspase 3 antibody was purchased from
Cell Signaling and was used at a 1:400 dilution. IgG isotype controls for rabbit
and mouse were purchased from Santa Cruz antibodies and used at dilutions of
1:400 and 1:500 as negative controls in all staining procedures. Immunolabelled
sections were counterstained for 10 seconds with hematoxylin 7211 and rinsed in
ddH
2
O three to four times to remove excess stain. Tissue sections were then
dehydrated through two ten-second washes in 95% and 100% FLEX alcohol,
followed by three five-second changes of Clear-rite 3. Excess clearite was blotted
and slides were mounted using clarion mounting medium and glass coverslips.
Slides were air-dried overnight prior to microscopy.

Statistical analysis
Statistical analysis was performed for tumor volumes in Microsoft Excel using the
Student’s t-test, assuming unequal variances. P-values less than 0.05, obtained
by this method were considered to be significant.

67
Flow cytometric analysis of TP421 uptake
The fluorescent properties of TP421 were exploited to investigate the cellular
uptake of TP compounds. PC3 prostate cancer cell lines were seeded at a
density of 5.0 x 10
5
cells/dish in 33 mm tissue culture-treated dishes and allowed
to adhere overnight in 2 mL growth media. The following day cells were collected
by trypsinization, washed and resuspended in 500 µL 1x PBS. Three minutes
prior to, and immediately following addition of 10 µM TP421 or DMSO, as vehicle
control, fluorescence versus time was recorded for emission wavelengths
between 425-475 nM in response to excitation with the 355 nM UV solid state
laser of the BD LSR II flow cytometer.

Fluorescence spectroscopy
Fluorescence spectroscopy was performed on TP compounds to determine
optimal excitation and emission wavelengths using the Fluorolog 3, steady state
spectrofluorimeter. Spectroscopy was performed using 5-10 µM compound
dissolved in dimethylsulfoxide (DMSO) resuspended in 2 mL ddH
2
O. Readings
were corrected to remove background fluorescence corresponding to DMSO and
ddH
2
O.

Fluorescent Microscopy of TP421 subcellular localization
PANC-1 cells were seeded in double chambered cover glass at a density of
50,000 cells / chamber and allowed overnight to adhere. The following day,
68
PANC-1 were treated with 2 µM TP421 for 30 minutes. Prior to imaging cells
were stained for 15 minutes at 37 ºC in humidified atmosphere containing 5 %
CO
2
using either 200 nM Mitotracker red CMXRos. Cells were visualized using a
Nikon DAIPHOT 300 inverted microscope equipped with DAPI and Cy3 filter
blocks, 10 x eye piece and 100 x / 1.3 Nikon oil immersion lens and super high
pressure mercury lamp. Images were captured using a Photometrics CoolSNAP
9 CCD camera and processed using Q-capture Pro v 5.1.1.14 imaging software.

Flow cytometric analysis of ROS formation
MDA-MB-435 cancer cells were treated with 2 µM TP compounds for various
time periods prior to incubation with 5 µM MitoSOX Red Mitochondrial
Superoxide Indicator at 37 º C for 10 minutes. Cells were trypsinized, washed
three times with Hank’s Balanced Salt Solution (HBSS) to remove residual
MitoSOX before resuspending in 750 µL 1x PBS. Fluorescence intensity
corresponding to oxidation of MitoSOX Red by superoxide radicals was recorded
for emission wavelengths between 562-588 nM in response to excitation with the
488 nM Sapphire™ argon-ion laser of the BD LSR II flow cytometer.

Acknowledgements
This study was supported in part by funds from the CDMRP Breast Cancer
Concept Award and the Sharon and William Cockrell Endowed Cancer Research
Fund.
69
Chapter 3
TP compounds in the treatment of pancreatic ductal
adenocarcinoma

Introduction

The primary obstacle to successful treatment of pancreatic ductal
adenocarcinoma (PDAC) is the lack of efficacious therapeutics. Failure of
treatment is attributable to a characteristically aggressive phenotype driven by
successive mutations in several key intracellular signaling pathways coupled with
poor tumor profusion caused by extensive desmoplasia within the tumor stroma.
For this reason, the discovery and development of novel agents that elicit clinical
response are of paramount importance. Targeting cancer cell metabolism
through inhibition of mitochondrial function is a novel approach for successful
PDAC treatment because of the essential role mitochondria have in cancer cell
energy production, metabolism and apoptosis. PDAC harboring mutations in the
KRAS oncogene exhibit significant metabolic reprogramming and thus are
expected to be particularly sensitive to mitochondrial inhibition. Inhibition of
cancer cell mitochondria in this manner will have the added benefit of interrupting
the metabolic coupling of tumor-stromal cell interactions which contribute to the
desmoplastic reaction that impedes drug delivery.

70

KRAS in pancreatic ductal adenocarcinoma
KRAS activating point mutations are observed in up to 95% of PDAC cases (Bos,
1989). This type of mutation confers unfavorable prognosis, aggressive
phenotype, and is associated with resistance to chemotherapy and radiation.
Similar to the other members of the Ras family of guanine nucleotide
transferases, K-ras is a critical point of integration between extracellular cues
originating at cell surface receptors and the intracellular signaling cascades that
drive proliferation, differentiation, migration, and survival, the most well
characterized being the Raf/Mek/Erk and and PI3K pathways. It is not surprising
then that constitutive activation at this key signal transduction hub is implicated
as a potent initiator in the pathogenesis of PDAC. The process of oncogenesis is
further stimulated by the loss of oncogenes p16, p53 and SMAD4, but ultimately
requires KRAS activation (Hruban et al., 2000) (Figure 3.1). Although, KRAS
mutations are highly prevalent in PDAC and constitute 32% of all solid tumors,
there are no effective treatments to serve this large subset of patients. Direct
targeting is hampered by a lack of binding pockets or active sites amenable to
interaction with small molecules, while the use of nucleotide analogs is
unsuitable due to picomolar affinity and high intracellular concentrations of GTP.
Recent efforts to develop therapeutic agents have focused on Ras localization or
downstream targets in the Ras/Raf/MEK/Erk pathway but have resulted in
compounds that lack efficacy or are overtly toxic due to compensatory signaling
mechanisms and lack of selectivity for cancer cells (Gysin et al., 2011).
71
Identification of agents that target multiple pathways to confer tumor-selective
lethality in the presence of KRAS mutation are considered a better approach and
are desperately needed.

Figure 3.1 Progression Model of Pancreatic Cancer

K-RAS mutation predicts PDAC sensitivity to mitochondrial-targeted agents.
It has long been known that Ras proteins interact with and regulate mitochondria
function. In the yeast, S.cerivisae, Ras2 proteins control mitochondrial biogenesis
and stimulate oxidative phosphorylation in response to changes in nutrient
availability and composition (Hlavata and Nystrom, 2003). Oncogenic Ras
signaling increases mitochondrial density in human Schwann cells more potently
72
than growth factor and mitogen stimulation (Echave et al., 2009). Sch66
mediated inhibition of Ras signaling prevents mitochondrial DNA replication in
transformed fibroblasts (Trinei et al., 2006). Ablation of activated KRAS in human
colon cancer cell lines inhibits cardiolipin synthesis compromising the efficiency
of cellular respiration and decreasing oxygen consumption (Chun et al., 2010).
Ras family members have also been reported to associate with Bcl2 and Bcl-XL
at the mitochondrial outer membrane where they function in the inhibition or
induction of apoptosis respectively (Rebollo et al., 1999).

More recently a growing body of evidence indicates an interdependent
relationship between KRAS and mitochondria in which oncogenic KRAS alters
mitochondrial function to support tumor growth such that inhibition of
mitochondrial function abrogates tumor growth (Figure 3.2). In an inducible
system meant to recapitulate KRAS driven tumorigenesis, cells stably transduced
with mutant KRAS exhibit significant, sustained alterations in mitochondrial
function as early as 12 h following oncogene induction (Hu et al., 2012) (Figure
73

Figure 3.2 Mitochondrial function supports, and is required for KRAS-dependent cell
proliferation. (A) Mutation in KRAS is considered an initiating event in PDAC and is
accompanied by mitochondrial dysfunction. In the context of the human PDAC progression
model, mutations in mitochondrial DNA have been noted as early as the PanIN1A stage of
hyperplasia and remain highly prevalent in PDAC. Changes in mitochondrial function are noted at
all stages of progression in various PDAC tissue samples and cell lines.

Figure 3.3 Timeline of bioenergetic changes occurring during KRAS driven oncogenic
transformation. In an inducible model, mitochondrial dysfunction is noted as early as 12 hours
following the induction of mutant KRAS and is sustained throughout the process of cellular
transformation.
74
3.3). In cell lines harboring endogenous or exogenous oncogenic KRAS,
inhibition of glutamine catabolism in the TCA cycle, detoxification of reactive
oxygen species derived from Complex III of mitochondria, or inhibition of
mitochondrial respiration have all been shown to limit KRAS anchorage
independent growth in soft agar (Weinberg et al., 2010). Inhibition of the mtDNA
transcription factor TFAM prevents mitochondrial DNA replication which ablates
in vivo tumorigenesis in a murine oncogenic K-ras background.

Based on this rationale it is reasonable to hypothesize that compounds
selectively targeting tumor cell mitochondria will exploit KRAS dependence on
mitochondrial function and can be developed as clinical agents to effectively kill
pancreatic cancer cells in primary as well as metastatic tumors.

Results

TP187-, TP197- and TP421- dependent modulation of core processes and
signaling pathways commonly altered in PDAC pathogenesis.

A recently published in-depth genetic analysis to detect and validate mutations
commonly occurring in pancreatic tumors identified 12 core intracellular signaling
pathways or processes universally affected (Jones et al., 2008). The 12 core
pathways identified were found to be mutated in 67-100% of all the PDAC tumor
samples examined and therefore are considered highly significant to the PDAC
75

Table 3.1 Core signaling pathways altered in PDAC tumors

Statistical significance in treated samples
TP187 TP197 TP421
Apoptosis   
DNA damage control   
Regulation of G1/S phase transition   
Hedgehog Signaling   
Homophilic cell adhesion - - -
Integrin Signaling   
c-Jun N-terminal kinase signaling   
KRAS signaling   
Regulation of invasion   
Small GTPase-dependent signaling - - -
TGF-β signaling   
Wnt/Notch signaling  - 
= -log(p-value) 1-5.9; = -log(p-value) 6-10.9; = -log(p-value) 11-15.9; = -log(p-value) 16-20;

Figure 3.4 TP compounds impinge on cellular functions required for pancreatic cancer
progression. Proteomics analysis of global protein expression demonstrated significant
alterations to pathways involved in cancer cell survival and death, growth and proliferation,
morphology and movement. Histograms for TP187, TP197 and TP421 showing statistical
significance as –log (p-value).

phenotype. Interestingly, pathway analysis of the kinexus proteomics data
indicated 10 of the 12 core pathways mentioned above as being significantly
affected by TP-treatment in MiaPaCa-2 cells (Table 3.1). Relation of these critical
effector pathways to our proteomics results substantiates the significance of TP-
76
compounds for the management of PDAC and demonstrates pleiotropic effects
necessary for efficacy against this highly mutated cancer type.

PDAC cell lines are sensitive to TP compounds

TP187, TP197 and TP421 and several close structural analogs previously
identified as having activity were screened in a panel of PDAC cell lines. In the
MIA PaCa cell line, IC
50
concentrations ranged from sub- to low-micromolar in
value. Panc-1 and BxPC3 cell lines were sensitive to micromolar concentrations
of TP compounds. Although the cell line panel was too small for statistical
significance, no distinct difference in sensitivity was noted between KRAS wild-
type and mutant cell lines in this panel (Table 3.2).

Table 3.2 IC 50 values of TP compounds in PDAC cell lines
IC 50, µM ± S.D.
Compound
MIA PaCa-2 BxPC3 Panc1
TP187 1.2±0.5 2.3±0.8 4.5±1.1
TP197 0.2±0.2 1.1±0.4 1.7±1.1
TP421 0.3±0.1 4.9±2.4 2.8±0.3
TP736 1.9±0.9 3.5±1.5 3.0±1.4
TP737 0.8±0.3 5.6±0.4 2.5±0.3
TP768 1.1±0.6 5.3±0.4 3.5±1.4
TP772 1.4±0.5 5.4±1.6 3.4±1.2
TP801 1.1±0.6 5.3±1.1 4.5±0.8
TP821 1.1±0.1 4.7±0.6 8.4±9.3

77

Figure 3.5 TP197 and its close analogs suppress tube formation a in 3-D model of
endothelial cell differentiation. A) Structures of TP197 and analogs. B) Micrographs showing
effects on tube forming capacity following treatment with 5 µM TP compounds.

Activity of TP197 and related analogs in PDAC and HUVEC cell lines
Cytotoxicity in PDAC cell
lines,
IC 50, µM± S.D.
IC 50, µM± S.D. in HUVEC
% dcrs @
5 µM

MiaPaCa BxPC3 Panc-1 Cytotoxicity
Adhesion
to VN
Adhesion
to FN
TFA Migration
TP197 0.2±0.2 1.1±0.4 1.7±1.1 5.8±4.6 5 4 0.5 42.1
TP736 1.9±0.9 3.5±1.5 3.0±1.4 6.3±0.4 15 10 5 26.2
TP737 0.8±03 5.6±0.4 2.5±0.3 5.5±1.4 15 12 5 33.3
TP768 0.6 5.5 4.5 8.8±4.6 >20 >20 5 nt
TP772 1.4±0.5 5.4±1.6 3.4±1.2 16.0±8.5 >20 >20 5 nt

TP197 and its close analogs inhibit integrin-mediated cell attachment and
exhibit anti-angiogenic activity

Proteomics analysis identified integrin ligation as one of the processes
significantly affected by TP-treatment. Visual observation of cell-rounding at early
time points following TP197 treatment further prompted the exploration of this
prediction. In cell-based adhesion assays, TP197 and several close analogs
inhibited integrin-dependent MDA-MB-435, MiaPaCa-2 and HUVEC cell
78
attachment to the extracellular matrix components vitronectin, fibronectin or
laminin at concentrations near or below that which caused cytotoxicity in 50% of
cells (IC
50
). Tumor cell haptotaxis on vitronectin was inhibited following treatment
with comparable concentrations of TP197 or close analogs. 3-D models of
vascular differentiation assess the ability of a given compound to alter endothelial
cell tube formation in gelled basement membrane extract (Figure 3.5). This type
of functional assay is intended to recapitulate the complex process of
angiogenesis. Therefore compounds showing activity in this model are expected
to show similar anti-angiogenic activity in vivo. When applied to this model,
TP197 and some of its analogs suppressed HUVEC tube formation at doses
between 0.5 and 1 µM. The emergence of a structure activity relationship
between TP197 analogs as well as a distinct lack of activity in TP187 and TP421
and their related compounds suggests a novel and previously unreported
mechanism for integrin inhibition unique to TP197 analogs.

Differential sensitivities to TP compounds in syngeneic cell lines harboring
wild-type or mutant KRAS

Table 3.3 IC 50 values of TP187, TP197 in KRAS cell lines
IC 50, µM
TP187 TP197 Cell line
glucose galactose glucose galactose
HEK293 WT-KRAS 4.0 7.0 0.5 2.5
HEK293 MUT-KRAS 0.5 5.5 0.3 1.0

Considering that additional phenotypic or genotypic differences between the cell
lines in the PDAC panel could be contributing to the response to treatment, we
79
opted to screen TP187 and TP197 in a pair of synegenic HEK293 cell lines, that
differed in KRAS status. The results of MTT screening are listed in Table 3.3.
Under typical cell culture conditions, only TP187 exhibited marked selectivity for
the KRAS mutant cell line. TP197 was active in both cell lines regardless of the
KRAS status. Tissue culture cell lines sub-cultured in glucose containing growth
media exhibit the “Crabtree” effect, meaning that they adopt a glyoclytic
phenotype irrespective of the functionality of their mitochondria. In essence, they
are behaving as cancer cells. To reverse the “Crabtree effect”, we sub-cultured
HEK293 cells using media in which the glucose was replaced with galactose.
Culturing cells in this manner, requires cells to increase respiration as they can
generate ATP solely through oxidative phosphorylation in the mitochondria, as
galactose oxidation via glycolysis produces no net ATP. Interestingly, when
treated in the presence of galactose a marked difference was noted in the IC
50

concentrations for both TP197 and TP187 in each of the cell lines. HEK293 WT-
KRAS were less sensitive to TP compounds when they adopted an oxidative
phenotype compared to their sensitivity when cultured in glucose. HEK293 MUT-
KRAS cultured in galactose were 11 times less sensitive to TP187 and 3 times
less sensitive to TP197. Comparing the response to treatment in galactose
between the cell lines showed little difference for TP187 but TP197 treated WT-
KRAS cells were about 2.5 times less sensitive than their MUT-KRAS
counterparts.

80
Mitochondrial attributes of HEK293 WT-KRAS and MUT KRAS cell lines

To provide more context in which to interpret these results, assessment of the
mitochondrial attributes of the syngeneic HEK293 cells was performed under
glucose and galactose growth conditions. To this end, we used flow cytometry to
measure parameters such as mitochondrial membrane potential, mitochondrial
density, superoxide production and NADH levels (Figure 3.6 A-D). Mitochondrial
membrane potential was unchanged when WT- KRAS cells were cultured in
galactose or glucose. In contrast, the mitochondrial membrane potential of MUT-
KRAS cells was decreased when cultured in glucose but returned to levels
81

A B
C

D

Figure 3.6 Mitochondrial attributes of KRAS wild-type and mutant cell lines grown in
glucose or galactose containing media. (A) JC-1 corresponding to mitochondrial membrane
potential. (B) MitoTracker Green FM staining intensity as a measure of mitochondrial density.
(C) Superoxide production as measured by MitoSOX Red superoxide indicator, (D) NADH
autofluorescence. Histograms are representative of three separate experiments.

observed in the wild-type cells in galactose containing media. Similar trends were
noted when mitochondrial density was measured. Superoxide production
increased when cells were cultured in galactose, this is in line with the
observation that respiration increases to furnish sufficient ATP. The increase
occurring when cultured in galactose was less pronounced for MUT-KRAS cells.
NADH changed relatively little when either cell line was switched from glucose to
82
galactose containing media, but was slightly increased in MUT-KRAS compared
to WT cells.

Long-term retention of TP421 in mitochondria

Initial experiments to determine the sub-cellular localization of TP421 were
carried out at short time points following treatment and thus provided a static
measure. Further microscopy experiments were carried out to characterize the
sub-cellular location(s) of TP421 throughout a longer time course of treatment.
TP421 fluorescence was compared to that of MitoTracker Red over a time period
of 72 h in the BxPC3 cell line (Figure 3.7). Co-localization of staining was
apparent at the earliest time point observed but required approximately 8 h for
complete uptake as evidenced by the high levels of TP421 background
fluorescence seen at earlier time points. Over the entire course of treatment,
TP421 localized to the mitochondria, At no time was TP421 present in the
nucleus of the cell. At early time points mitochondrial morphology appeared
normal, but beginning at 24-48 h mitochondria appeared rounded and swollen.
83

Figure 3.7 TP 421 mitchondrial uptake is rapid and sustained. BxPC3 cells were treated with
5 µM TP421 and imaged at various timepoints. MitoTracker Red staining colocalized with TP421
fluorescence indicating mitochondrial localization.

Figure 3.8 TP421 parent coumarin does not enter mitochondria. BxPC3 cells were treated
with 10 µM TP421 parent coumarin and imaged at various timepoints. MitoTracker Red
staining did not colocalize with TP421 fluorescence indicating mitochondrial localization.
84
A non-targeted TP421 analog does not enter mitochondria.

A non-targeted TP421 analog (i.e lacking the triphenylphosphonium moiety) was
tested in MTT assay and reported to be devoid of anti-proliferative activity
(Shabaik et al. 2013) As an extension of this work, we performed fluorescent
imaging using this compound. The parent coumarin had similar fluorescence
emission, and therefore a time course similar to the previous experiment was
carried out in BxPC3 cells, substituting the TP421 parent compound for TP421
(Figure 3.8). Over the time course some mitochondrial staining was noted but
the majority of the coumarin fluorescence was cytosolic. No significant
morphological changes were noted, even at late timepoints.

In vitro screening of a TP187 analog and novel allyl-phosphonium salts in a
panel of PDAC and breast cancer cell lines

A series of TP187 analogs were recently synthesized. The chemical structures
of these allyl-phosphonium salts are shown in Figure 3.9. The allyl
triphenylphosphonium was the closest in structure to TP187, the double-bond
being terminal in this compound. The other allyl-phosphonium salts were
tricyclohexyl- , triethyl-phosphonium, and one additional allyl, a triethyl-aminium
salt was also prepared. These newly synthesized analogs were screened in
MTT assay in order to establish a structure-activity relationship (Table 3.4) . Ally-
TPP retained activity comparable to TP187 in each of the cell lines tested.
85

Figure 3.9 Chemical Structures of TP187 analogs

Table 3.4 Anti-proliferative properties of Allyl -TPP and analogs
IC 50, µM
Compound Panc-1 BxPC3 MIA PaCa-2 MDA MB 231 BT549
TP187 1.5 2.5 1.5 0.7 1.5
Allyl TP 2.0 2.5 1.2 0.8 2.0
Allyl-Cy3 > 50 > 50 > 50 > 50 > 50
Allyl-Et3 > 50 > 50 > 50 > 50 > 50
Allyl-N-Et3 > 50 > 50 > 50 > 50 > 50
P
Br
P
Br
N
Br
P
Br
allyltriphenylphosphonium bromide allyltricyclohexylphosphonium bromide
allyltriethylphosphonium bromide N,N,N-triethylprop-2-en-1-aminium bromide
86
Replacement of the phenyl rings with cyclohexyl, or ethyl groups abolished
activity as did replacement of the phosphonium cation with a nitrogen cation.
These results underscore the importance of the TPP moiety in the activity of TP
compounds.

Rapid activation of AMPK in TP197-treated PDAC cells.

AMPK is a central node in energy sensing within the cell. AMPK is activated in
response to various stimuli including nutrient and oxygen depletion or increases
in the ADP to ATP ratio. Once activated, it prompts the cell to shift from energy
expending activity to energy conserving processes in an attempt to restore
homeostasis. TP compounds cause a time- and dose –decrease in oxygen
consumption accompanied by similarly inverse changes in glycolytic capacity().
The change in bioenergetics is accompanied by increased superoxide production
and decreased mitochondrial membrane potential. Given these energetically
unfavorable consequences, AMPK expression and phosphorylation levels were
measured in MIA PaCa-2 cells treated with TP compounds for 5m up to 2 h
(Figure 3.10-A). TP187 did not have much effect on AMPK phosphorylation.
TP197–treated MIA PaCa-2 exhibited an increase in AMPK phosphorylation
within 15 minutes and continuing through the 2h timepoint. Acetyl Co-A
carboxylase (ACC) catalyzes the first and rate-limiting step of fatty acid
synthesis. ACC is a direct target of AMPK and its phosphorylation inhibits its
87
activity. In MIA PaCa-2 treated with TP197, ACC phosphorylation peaked at 5
min post-treatment and declined

A

B

C
Figure 3.10 TP197 caused rapid
phosphorylation of AMPK and ACC.
Western blots showing AMPK
phosphorylation in (A) MIA PaCa-2
treated for short periods with TP187 or
TP197, (B) MIA PaCa-2 treated with
5 or 2.5 µM TP197 for up to 72 h , (C)
MDA MB 468 treated 0-24 h 1 µM with
TP197, 10 mM metformin or
combination.

Figure 3.11 TP197 caused rapid turnover
of LC3B and DNA damage response.
Western blot in MIA PaCa-2 treated with
increasing concentrations TP197 0-5 h.
88
to baseline within the 2 h treatment frame. With longer treatment at lower
dosing, AMPK phosphorylation was sustained up to 72 h following treatment
(Figure 3.10-B). Phosphorylation of ACC reached its height at 18 hours
treatment and returned to baseline by 48 h treatment. In MDA MB 468 breast
cancer cells, TP197 induced AMPK phosphorylation occurred in a time
dependent manner, peaking at 18 h treatment and returning to baseline by 24 h.
Phosphorylation of ACC followed similar kinetics. Combining TP197 with the
AMPK-activating diabetes drug, metformin resulted in phosphorylation at earlier
time point and for longer duration (Figure 3.10-C). Growth curves for MDA MB
468 treated with TP197 in the presence or absence of metformin suggests that at
lower concentrations of TP197, synergistic effects on cell proliferation could be
achieved.

Rapid Induction LC3B conversion following TP197 treatment

As a means to conserve energy and restore homeostasis, AMPK can activate
catabolic processes such as autophagy. Light chain 3B-I (LC3B-I) is converted
during the process of autophagy via cleaveage and lipidation to give LC3B-II
which is then incorporated into autophagic vesicles. In this capacity it serves as a
marker of autphagy. Given the rapid kinetics of AMPK activation in MIA PaCa-2
following TP197 treatment, LC3B-I and -II expression were assayed at short time
points over a series of three doses (Figure 3.11). For each dose, TP197 caused
89
a rapid increase of LC3B-II levels at 5 min that gradually returned to baseline
over the 5 h time course.

TP197 stimulates DNA-damage response

TP 421 has been reported to induce DNA damage response in pancreatic cancer
cell lines. Similar to the rapid induction of AMPK phosphorylation and LC3B-II
accumulation, TP197 treatment also caused a rapid induction of Histone H2A.X
phosphorylation (Figure 3.11).

Discussion

Pancreatic cancers are characteristically aggressive and treatment options
capable of significantly prolonging survival are desperately needed. Proteomics
analysis performed on kinexus antibody microarray data predicted that TP
compounds had the potential to be suitable drugs for the treatment of pancreatic
cancers. Mechanistic studies carried out with TP421 indeed confirmed the
involvement of several of these pathways in TP421 mechanism of action
(Shabaik et al., 2013). In line with these predictions and previous data, TP421,
TP187, TP197 and its close analogs were active in the PDAC panel.

Integrins influence tumor growth by both direct and indirect mechanisms.
Integrins are required for tumor angiogenesis, the growth of new vessels from
90
existing ones. Angiogenesis is essential for tumor progression, as the
vasculature routes nutrients and oxygen, required for growth, to the tumor bed. In
the course of tumor progression, alterations in surface expression of integrins is
common and facilitates migration, invasion, anchorage-independent growth and
metastasis. The PDAC microenvironment is characterized by excessive
deposition of extracellular matrix, therefore agents that interfere with or inhibit
integrin mediated signaling are of clinical importance (Desgrosellier et al., 2009;
Grzesiak et al., 2007; Mahadevan and Von Hoff, 2007). TP197 and several
close structural analogs inhibited integrin mediated cell adhesion and migration
and were active to varying degrees in inhibiting tube formation. Angiogenesis is a
complex, multi-step process that is recapitulated well in the 3-D Matrigel tube
formation assay. Agents that are active in this type of assay have high potential
for clinical translation.

The selectivity for KRAS mutant cell lines was not distinctly evident when our
PDAC cells were treated with TP compounds which could be possibly due to the
small size of the panel. These differences could also be attributed to additional
mutations or physiologic differences between the cell lines. To control for the
differences imparted by genetic variations between cell lines the lead TP
compounds were also screened in the syngeneic HEK293 cell lines WT-KRAS
and MUT- KRAS. In this model, MUT-KRAS cells were 8 times more sensitive to
TP187 than WT-KRAS cells. No differences in sensitivities were noted in
HEK293 WT-KRAS and MUT-KRAS cell lines treated with TP197. When we
91
attempted to recapitulate the 'oxidative phenotype' by culturing cells in the
presence of galactose, sensitivity to TP187 was attenuated in the MUT-KRAS
cell lines. Increases in IC
50
values were seen for both cell lines treated with
TP197in the presence of galactose. Overall, when cultured in galactose and
treated with TP compounds, HEK293 WT-KRAS cells were less sensitive than
MUT KRAS cells. When placed in context of the predicted in vivo phenotypes of
normal and transformed cells, (i.e. WT-KRAS/galactose and MUT-
KRAS/glucose) the fold change in IC
50
concentrations for each of the TP
compounds is quite impressive (TP187 7.0 vs 0.5 µM TP197 2.5 vs 0.3 µM).
These results provide a possible mechanism for the lack of toxicity and selectivity
in our mouse xenograft model.

By replacing media glucose with galactose cells are forced to use mitochondria
as the sole source of ATP production. Therefore, cell lines cultured in the
presence of galactose respond by increasing oxygen consumption and
respiration (Marroquin et al., 2007). Based on the results shown here, for the
HEK293 MUT-KRAS cell lines, changes in key mitochondrial attributes
accompanied the switch to galactose containing media. Although in WT-KRAS
cells the mitochondrial membrane potential and mitochondrial density were
maintained regardless of the growth conditions, the concomitant uptick in
superoxide production does suggest that growth in galactose was indeed
increasing mitochondrial output. MUT-KRAS cell lines had lowered mitochondrial
membrane potential and mitochondrial density when adopting the glycolytic
92
phenotype but showed remarkable plasticity when cultured in galactose. This
suggests that KRAS mutation may have a suppressive effect on mitochondrial
function when glycolytic enzymes are stimulated such as in the presence of high
glucose concentrations but when nutrients are limiting, the reserve capacity of
the mitochondria is maintained. Interestingly, despite their lowered mitochondrial
membrane potential in glucose containing media, compared to WT-KRAS cells,
MUT-KRAS cells were more sensitive to TP187 and equally sensitive to TP197 .
These results provide support for the hypothesis that KRAS mutation sensitizes
cells to TP compounds.

In agreement with the early SAR performed on isoindolalkylphosphosphonium
salts, our studies on the relationship between the sub-cellular localization of TP
compounds and cytotoxicity demonstrated that the cytotoxicity is dependent upon
the triphenylphosphonium cation. Using fluorescence microscopy, we compared
the time to uptake, retention and sub-cellular localization of TP421 and its parent
coumarin compound lacking the triphenylphosphonium cation. Both TP421 and
the parent coumarin entered cells rapidly and were retained through out the 72 h
time period. By 72 h, TP421 treated BxPC3 cells exhibited morphological
changes and swollen mitochondria indicative of cytotoxic effects of the
compounds. Despite its long retention time , the 421 parent coumarin exhibited
no cytotoxicity in cells (Shabaik et al., 2013) likely due to the fact that it did not
localize within mitochondria. The extended retention time of both molecules,
93
although in different sub-cellular compartments may be due to the high affinity of
coumarins for lipids membranes (Sarpietro et al., 2011).

In another approach to understanding the relationship between mitochondrial
localization and cytotoxic activity a series of allylphosphonium salts were
synthesized and screened in a panel of cell lines comprised of breast and
pancreatic cancer cell lines. The prototype allylphosphonium salt, was a close
analog to TP187. The remainder of the series was derived by replacing the
phenyl groups with less hydrophobic constituents such as cyclohexyl and ethyl
groups. In addition an iminium bromide salt in which the phosphine was replaced
with a nitrogen cation was also prepared and tested. The allylphosphonium salt
showed comparable activity to TP187. Replacement of the phenyl rings
completely abolished activity. Taken together these results raise an important
distinction regarding the TP cation. Most, if not all TPP cations enter
mitochondria, but not all are cytotoxic. Conversely, not all molecules that are
cytotoxic when delivered to the mitochondria will retain activity outside of the
mitochondria.

In our previous studies, TP compounds were found to have dose and time
dependent effects on cellular bioenergetics that preceded induction of stress and
DNA damage responses (Millard et al., 2010b; Shabaik et al., 2013). These
same responses were noted for TP197, albeit with more rapid kinetics than
TP421. An increase in AMPK phosphorylation concomitant with ACC
94
phosphorylation, not previously reported was observed in PDAC and breast
cancer cell lines treated wit TP197. Given the similar kinetics it is likely that
AMPK activation is occurring upstream of LC3B-II induction. The increase in
phosphorylated Histone H2A.X is likely due to the increase in ROS production
that follows TP treatment. The relationship between TP treatment, ROS and DNA
damage is explored in more detail in the next chapter.
Experimental Procedures

Kinexus antibody microarray
MIA PaCa-2 cells were treated with 1 µM of TP187, TP 197 or TP421 for 24 h.
Upon completion of treatment, cells were washed in ice-cold PBS to remove
residual medium, then lysed in 200 µL of lysis buffer (20 mM MOPS, pH 7.0, 2
mM EGTA, 5 mM EDTA, 30 mM sodium fluoride, 60 mM β-glycerophosphate, pH
7.2, 20 mM sodium pyrophosphate, 1 mM sodium orthovanadate, 1 mM
phenylmethylsulfonylfluoride, 3 mM benzamidine, 5 µM pepstatin A, 10 µM
leupeptin, 1% Triton X-100, 1 mM dithiothreitol) and collected. The cell lysates
were sonicated on ice, four times for 10 s each pausing for 15 s intervals
between pulses, to rupture the cells and shear DNA. After sonication, the
homogenates were cleared by centrifugation at 90,000 x g for 30 min at 4 °C.
The supernatants were transferred to a clean microcentrifuge tube and the
protein concentrations were measured using the BCA protein assay. Whole cell
lysates in a final volume of 250 µL were submitted to Kinexus for a 628-antibody
microarray analysis using the Kinex™ KAM-1.1 Antibody Microarray.
95

Ingenuity Pathway Analysis
Potential intracellular signaling pathways or molecules affected by TP compound
treatment were identified using the Ingenuity Pathway Analysis (IPA) software to
analyze the Kinexus™ antibody microarray results. The significantly up-regulated
or down-regulated pan-specific or phopsho-proteins with their Swiss-Prot
accession numbers and the ratio changes were uploaded as an Excel
spreadsheet file to the IPA server. TP-mediated signaling pathways were
analyzed by core analysis.

Cell Culture
The syngeneic human embryonic kidney HEK293 WT-KRAS and HEK293 MUT-
KRAS were previously generated in our lab using HEK293 cells purchased from
ATCC. PDAC cell lines, including MIA PaCa-2, Panc-1 and BxPC3 were also
purchased from the ATCC. The breast cancer cell lines MDA MB 231 and BT549
were also purchased from ATCC. Cell lines were maintained in growth media
containing 10 % FBS and 2 mM L-glutamine. HEK293 cells were also cultured in
the presence of 10 µg/mL hygromycin selection reagents. All experiments were
carried out sub-confluent cells in exponential growth phase, in the absence of
antibiotics or selection reagents. For experiments requiring the use of galactose,
cells were cultured in the appropriate growth media prepared with out glucose
and supplemented with 0.9 g/L galactose and 2 mM L-glutamine.

96
MTT Proliferation Assay
MTT assays were carried out as previously described (Carmichael et al., 1987a).
For cells incubated in galactose, media was removed at the end of 72 h
treatment and replaced with glucose containing media. MTT reduction was found
to be similar between controls treated in the presence of glucose containing
media and galactose containing media. MTT assay was performed as usual.

3-dimensional endothelial cell tube formation assay
Growth-factor reduced basement membrane extract (BME) was thawed
overnight at 4º C on ice. The next day, 30 µL ice-cold BME was added to the
wells of a chilled, 384-well, black-walled imaging plate. The plates were then
incubated at 37º C for thirty minutes to allow the BME to solidify. Vehicle
(DMSO), sulphorafane, and the individual TP compounds were prepared as
serial dilutions at 2 x final concentration in Endothelial Cell Basal Medium (EBM-
2). Next, 25 µL of each dilution was added to the respective BME-coated wells.
Sub-confluent cultures of human umbilical vein endothelial cells, (HUVEC) were
collected with 0.25% trypsin/EDTA and re-suspended in 2x endothelial cell
growth medium (EGM-2) consisting of EBM-2, 4 % FBS and 2x concentration of
the growth factors and supplements contained within the Endothelial Growth
Medium SingleQuots Kit to give a cell density of 2.0 x 10
5
cells/ mL. To each well
containing treatment or vehicle, 25 µL of cell suspension was added to give a
final volume of 50 µL and 1x concentrations of media and compound. Cells were
incubated at 37º C for 6-8 h to allow sufficient time for tube formation. When
97
tubular networks had formed in the control wells, 5 µL of EBM-2 containing 10 x
calcien AM (final concentration 6 µM) was added to each well. Imaging was
performed on a BD-Pathway 435 high-content bioimager equipped with calcien
AM filter and 4x objective.

Measurement of Mitochondrial Membrane Potential
JC-1 staining for Δψ
mt
: HEK293 cells WT-KRAS and HEK293 cells MUT-KRAS
and were collected with trypsin followed by centrifugation at 1200 RPM for 5
minutes at RT. Cell pellets were washed with 1x PBS, centrifuged and
resuspended in basal media containing 10 µg / mL JC-1 dye. Cells were
incubated for 15 minutes in a 37 °C water bath. In preparation for analysis cells
were centrifuged at 5000 RPM for 5 minutes at RT in a microcentrifuge. Pelleted
cells were resuspended in 1x PBS and kept on ice. Mitochondrial membrane
potential was determined as the ratio of mean red fluorescence intensity to green
fluorescence intensity. Mean fluorescence intensity was measured on a BD
LSRII flow cytometer using detectors equipped to capture light emitted in the 525
and 590 nM range following excitation at 488 nM.

Measurement of Mitochondrial Density
JC-1 staining for Δψ
mt
: Human embryonic kidney HEK293 cells WT-KRAS and
HEK293 cells WT-KRAS and were collected with 0.25% trypsin EDTA followed
by centrifugation at 1200 RPM for 5 minutes at RT. Cell pellets were washed
with 1x PBS, centrifuged and resuspended in basal media containing 50 ng/mL
98
Mitotracker Green FM dye. Cells were incubated for 15 minutes in a 37 °C water
bath. In preparation for analysis cells were centrifuged at 5000 RPM for 5
minutes at RT in a microcentrifuge. Pelleted cells were resuspended in 1x PBS
and kept on ice. Mitochondrial density was determined as the mean of green
fluorescence intensity. Mean fluorescence intensity was measured on a BD
LSRII flow cytometer using detectors equipped to capture light emitted in the 525
range following excitation at 488 nM.

Measurement of Superoxide Production
MitoSOX red staining: Human embryonic kidney HEK293 cells WT-KRAS and
HEK293 cells WT-KRAS and were collected with trypsin followed by
centrifugation at 1200 RPM for 5 minutes at RT. Cell pellets were washed with
1x PBS, centrifuged and resuspended in basal media containing 50 ng/mL
MitoSOX Red dye. Cells were incubated for 15 minutes in a 37 C water bath. In
preparation for analysis cells were centrifuged at 5000 RPM for 5 minutes at RT
in a microcentrifuge. Pelleted cells were resuspended in 1x PBS and kept on ice.
Mitochondrial membrane potential was determined as the ratio of mean red
fluorescence intensity to green fluorescence intensity. Mean fluorescence
intensity was measured on a BD LSRII flow cytometer using detectors equipped
to capture light emitted in the 590 nM range following excitation at 488 nM.

99
Measurement of NADH levels
NADH auto-fluorescence: Human embryonic kidney HEK293 cells WT-KRAS
and HEK293 cells WT-KRAS and were collected with trypsin followed by
centrifugation at 1200 RPM for 5 minutes at RT. Pelleted cells were resuspended
in 1x PBS and kept on ice. NADH auto-fluorescence was determined as the
mean fluorescence intensity. Mean fluorescence intensity was measured on a
BD LSRII flow cytometer using detectors equipped to capture light emitted in the
460 nM range following excitation at 355 nM.

Synthesis of Novel Phosphonium Salts
Phosphonium salts were designed and synthesized by the Olenyuk laboratory
and provided as lyophilized powders. For use in tissue culture experiments, a
small quantity of powder was weighed and resuspended in sufficient volume of
reagent grade DMSO to give a concentration of 10 mM. Compounds were
aliquoted into single use tubes and stored at -20 C prior to use.

Western Blotting
Western blotting experiments were carried out as previously described in chapter
2.

Fluorescent Microscopy of TP421 subcellular localization
MIA PaCa-2, PANC-1 and BxPC-3 cell lines were seeded in double chambered
cover glass at a density of 50,000 cells / chamber and allowed overnight to
100
adhere. The following day, MIA PaCa-2 and PANC-1 were treated with 2 µM
TP421 while BxPC-3 cells were treated with 5 µM TP 421 or 10 µM of the parent
TP coumarin for time periods up to 72 h. Prior to imaging cells were stained for
15 minutes at 37 ºC in humidified atmosphere containing 5 % CO
2
using either
200 nM Mitotracker red CMXRos or 50 nM Lysotracker Red DND 99 live cell
organelle stains prepared as a 10x solution in warmed HBSS in order to
visualize mitochondria and lysosomes, respectively. Cells were visualized using
a Nikon DIAPHOT 300 inverted microscope equipped with DAPI and Cy3 filter
blocks, 10 x eye piece and 100 x / 1.3 Nikon oil immersion lens and super high
pressure mercury lamp. Images were captured using a Photometrics CoolSNAP
9 CCD camera and processed using Q-capture Pro v 5.1.1.14 imaging software

Immunofluorescent staining of fixed cells
MIA PaCa-2 and BxPC-3 cell lines were seeded on glass coverslips at a density
of 50,000 cells and allowed overnight to adhere. The following day, MIA-PaCa-2
and BxPC3 cells were treated with 2 µM TP421 in the presence or absence of 5
µM rapamycin or 10 µM chloroquine for 18 h for LC3B staining or 1 µM TP197 or
1 µM Captothecin for mnSOD and pH2A.X assessment. At the end of treatment,
media was removed and cells were washed with 500 µL 1x PBS prior to fixation
with 3.7% formaldehyde for 15 m at RT. Fixed cells were washed with 500 µL 1x
PBS prior and subsequent to permeablization with ice-cold acetone for 5 minutes
at -20C. Coverslips were blocked for 30 m with 1 % bovine serum albumin (BSA)
in PBS to inhibit non-specific antibody binding prior to incubation overnight at 4C
101
with antibodies raised against LC3B and Complex I or pH2A.X and mnSOD
diluted 1:1000 in 1% BSA/PBS. Antibodies were removed and coverslips were
washed 500 µL 1x PBS with gentle agitation. Goat-anti-rabbit -Cy5 conjugated
and Goat-anti-mouse Cy3-conjugated antibodies were diluted 1:200 in 1%
BSA/PBS and incubated with coverslips for 2 h. Coverslips were again washed
with PBS500 µL 1x PBS with gentle agitation, air-dried and mounted on pre-
cleaned glass slides using Prolong Gold anti-fade mounting media. Images were
obtained using a Lieca SP2 scanning confocal microscope equipped with 488 nm
argon and 633 nM krypton lasers and Leica Confocal software v 2.61

102
Chapter 4
A selective, mitochondrial-targeted chlorambucil with
remarkable cytotoxicity in breast and pancreatic cancers

Introduction

Chlorambucil, a member of the nitrogen mustard class of DNA alkylators, is used
primarily for the treatment of chronic lymphocytic leukemia (CLL), lymphomas
such as Hodgkin’s Disease and non-Hodgkin lymphoma, Waldenström’s
Macroglobulinemia and some solid tumors (Chabner BA, 2011). The nitrogen
mustards share a common mechanism of action stemming from the presence of
an N,N-bis(2-chloroethyl)amine group. This moiety is highly reactive with
intracellular components such as proteins, phospholipids and nucleic acids.
Covalent modification of proteins and phospholipids can have modest inhibition
of cellular function, but alkylation of genomic DNA is the primary mechanism of
cytotoxicity. Chlorambucil modifies DNA through the formation of either mono- or
bi-functional adducts. Mono-adducts interfere with gene transcription and
promote mismatched base-pairing, while bi-functional alkylation creates intra-
and inter-strand crosslinks that inhibit DNA synthesis and cause double strand
breaks. In this way, exposure to chlorambucil leads to apoptotic cell death via the
accumulation of persistent DNA damage (Hansson et al., 1987).

Issues of drug stability, selectivity and resistance are common to this class of
alkylating agent. The high reactivity and short half-life of nitrogen mustards can
103
attenuate drug potency. Covalent interactions with proteins and phospholipids
can sequester active drug away from the site of action such that < 10% of the
original dose may be available to interact with DNA. These consequences of
instability and reactivity are compounded by the slow rate of passive drug uptake.
In the end, overcoming ‘off-target’ binding and drug instability requires higher
dosages to produce a therapeutic response.

The lack of selectivity for tumor cells is another concern. In the short term,
myelosuppression, the primary dose-limiting toxicity, increases the risk of serious
infections. Over the longer term, cumulative damage to normal tissues
significantly increases the probability of developing secondary neoplasms
(Chabner BA, 2011). In addition to toxicities, repeat administration eventually
leads to drug resistance for most patients (Panasci et al., 2001). In a recent
analysis of several Phase III clinical trials, dose and duration of treatment were
identified as two important determinants of response to chlorambucil (Catovsky
et al.). Thus, the toxicity and drug resistance associated with nitrogen mustards
should be considered as serious impediments to therapeutic efficacy.
Conversely, enhancement of chlorambucil tissue selectivity, drug accumulation
and tumor sensitivity could significantly benefit treatment outcomes.

Chlorambucil was one of the first DNA damaging agents used for chemotherapy
and for decades was considered the standard of care for CLL because it
produced fewer dose-limiting toxicities than other alkylating agents in this class.
104
Although newer drugs have taken the place of chlorambucil in current CLL
chemotherapeutic regimens, the fact that it remains a first-line treatment for CLL
in the elderly and immune-suppressed patients underscores its continued
relevance as an anti-cancer agent (Catovsky et al.; Knauf et al.). Certain
advanced cancers such as hormone refractory or hormone receptor negative
breast cancers and pancreatic adenocarcinomas often fail to respond to current
regimens. The repurposing of ‘tried and true’ agents such as chlorambucil could
offer a much needed treatment option for these patients. Thus from a medicinal
chemistry standpoint, tissue selectivity and drug resistance are key areas of
focus that could expedite the repurposing of chlorambucil.

With this in mind, we hypothesized that directing chlorambucil to cancer cell
mitochondria might be an effective approach to combat issues of safety and drug
resistance. Linkage to a lipophilic cation to exploit the higher mitochondrial
membrane potentials of solid tumors, we reasoned, would improve tissue
selectivity and hasten drug uptake. Meanwhile the lack of a nucleotide excision
repair mechanism (NER) in mitochondria might sensitize resistant cells to the
DNA damaging effects of chlorambucil (Cullinane and Bohr, 1998; LeDoux et al.,
1992).

105

Chlorambucil Mito-Chlor
Figure 4.1 Chemical Structures of Chlorambucil and Mitochlor.

To this end, we designed and synthesized Mito-Chlor, a phosphonium salt
derivative of chlorambucil. When screened in a panel of human carcinoma cell
lines, Mito-Chlor exhibited remarkable potency compared to the parent drug. The
sensitivity of these previously resistant cell lines was dependent on the ability of
Mito-Chlor to enter mitochondria and bind mtDNA. Once in the mitochondria,
Mito-Chlor caused S-phase specific cell cycle arrest consistent with mtDNA
damage and acted to increase nuclear Histone H2A.X phosphorylation (γH2A.X).
Interestingly, Mito-Chlor-induced γH2A.X, but not Mito-Chlor cytotoxicity, was
reversible upon treatment with antioxidant.
106
Results

A two-step synthetic method for preparation of Mito-chlor

Using the carboxylic acid functional group as a point of attachment, we designed
a facile, 2-step method of TPP conjugation suitable for large scale production of
Mito-chlor.

Conjugation to triphenylphosphonium cation reversed chlorambucil
resistance.

The newly synthesized TPP-derivative, Mito-Chlor and the parent drug
Chlorambucil were screened for anti-proliferative activity in a panel of breast
cancer cell lines using the colorimetric MTT assay. The cell lines comprising this
panel were representative of typical molecular subtypes encountered in clinical
practice and as such varied considerably in hormone receptor and HER2/neu
status (Table 4.1). To determine the concentration that could inhibit cell viability
and/or proliferation by 50 % (IC
50
concentration), cells were treated for 72 h with
increasing concentrations of compound, ranging over three to four logs. The
resulting IC
50
values for Chlorambucil and Mito-Chlor in each cell line are
presented in Table 4.2 along with the sensitivity ratio. The sensitivity ratio
represents the fold change in IC
50
value between the parent drug and its TPP
analog. As shown, none of the cell lines were sensitive to chlorambucil, their IC
50

107
concentrations ranging from 44 µM to over 500 µM. In contrast, the panel of cell
lines was highly sensitive to Mito-Chlor. For each of the cell lines tested, the IC
50

concentration of Mito-Chlor was significantly lower than that of Chlorambucil, all
ranging from 1.7 - 11.4 µM. Among the panel, MBA-MB-468 was most sensitive
Table 4.1 Molecular attributes of selected breast cancer cell line panel.
cell line molecular subtype

known mutations

MDA MB 435 ER -, PR - basal B IDC BRAF, CDKN2A,TP53,
MCF-7 ER +, PR +, luminal IDC CDKN2A, PIK3CA
MDA MB 231 triple negative, mesenchymal-like IDC
BRAF, CDKN2A,KRAS,
NF2,TP53,PDGFRA
MDA MB 468 triple negative, basal-like DC PTEN, RB1, SMAD4, TP53
CAMA-1 ER +, PR -, HER2 -, luminal AC CDH1, PTEN, TP53
BT-549 triple negative, Basal B IDC PTEN, RB1, TP53
Hs578T triple negative, Basal B IDC CDK2NA, HRAS, PIK3R1,TP53
SKBR-3 ER -, PR -, HER2 +, luminal AC
CDH1 deletion, decreased BRCA-1
expression
BT-474 ER +, PR +, HER2 +, luminal IDC PIK3CA, TP53

IC 50 values represent the average of three experiments, each treatment performed in triplicate ± standard deviation. Sensitivity ratio is
the ratio of the IC 50 values of chlorambucil : Mito-Chlor, which indicate the increase in sensitivity to Mito-Chlor.

Table 4.2 IC 50 values (µM) of chlorambucil (CMBC) and Mtio-Chlor in a panel of breast cancer cell lines
Compound
MDA
MB 435
MCF-7
MDA
MB 231
MDA
MB 468
CAMA-1 Hs578T SKBR-3 BT549 BT-474
CMBC
150 ±
5.0
73.3 ±
14.4
521.3 ±
249.9
43.8 ±
8.1
191.7 ±
62.9
162.5 ±
88.4
247.5 ±
3.5
150,
>200,
>200
237.5 ±
17.7
Mito-Chlor
3.0 ±
1.9
9.5 ±
5.7
9.5 ±
5.7
1.7 ±
0.6
2.4 ± 3.0
11.4 ±
4.3
5.7 ±
5.9
7.3 ±
2.9
6.5 ±
3.6
sensitivity
ratio
50.0 7.7 54.8 25.8 79.9 14.3 43.4 27.4 36.5
108
Table 4.3 Molecular attributes of selected PDAC cell line panel
cell line type known mutations and deletions
MIA PaCa-2
pancreatic ductal
adennocarcinoma
KRAS, TP53, KDM6A, CDK2NA, CDK2Na(p14)
BxPC-3
pancreatic ductal
adennocarcinoma
CDKN2A, CDKN2a(p14), MAP2K4, SMAD4, TP53
Panc-1
pancreatic ductal
adennocarcinoma
KRAS, TP53, CDKN2a(p14), HER2/neu +++, ER -,
MUC4 -

Table 4.4 IC 50 values (µM) of chlorambucil and Mito-Chlor in a panel of pancreatic ductal adenocarcinoma
cell lines
compound MIA PaCa-2 BxPC3 Panc-1
chlorambucil 78.3 ± 0.3 >100 110 ± 13.2
Mito-Chlor 1.6 ± 0.9 2.5 ± 2.8 50
sensitivity ratio 48.9 > 40 2.2
IC 50 values represent the average of three experiments, each treatment performed in triplicate ± standard deviation.

to Mito-Chlor, having the lowest IC
50
(1.7 µM), whereas CAMA-1 with a sensitivity
ratio of 79.9, demonstrated the greatest increase in potency in response to Mito-
Chlor. Additional anti-proliferative screening was carried out in a small panel of
pancreatic cancer cell lines (Table 4.3). A similar increase in potency was
observed in two of the three pancreatic cancer cell lines treated with Mito-Chlor
(Table 4.4). The IC
50
values for MIA PaCa-2 and BxPC-3 treated with Mito-Chlor
were 1.6 and 2.5 µM, respectively. Each of these values represent a greater
than 40-fold increase in potency over the parent drug Chlorambucil. For unknown
reasons, the Panc-1 cell line was insensitive to both Chlorambucil and Mito-
Chlor. Overall, these results demonstrate a significant increase in chlorambucil
potency as a result of conjugation to the phosphonium cation.

109
Mito-Chlor localized to mitochondria.

In order to determine if this increase in potency was due to mitochondrial-
targeting, the subcellular localization of Mito-Chlor was observed by confocal
microscopy (Figure 4.2). MIA PaCa-2 cells treated with Mito-Chlor were fixed
and co-stained with antisera specific for the TPP moiety and an antibody raised
against Complex I of the electron transport chain in combination with fluorophore-
conjugated secondary antibodies. The mitochondrial membrane potential-
dependent nature of TPP accumulation dictates that upon loss of mitochondrial
integrity, any unbound compound will diffuse out of the mitochondria. Therefore if
Mito-Chlor is able to enter mitochondria and alkylate mtDNA, it should be
retained during the fixation and permeablization process and be readily

Figure 4.2 Confocal images of MIA PaCa-2 treated 18 h, with 100 µM chlorambucil or 10
µM mitochlor, stained with anti-Complex I antibody and anti-TPP antisera.
110

detectable within mitochondria upon immunofluorescent staining with anti-TPP
anti-sera. Consistent with mitochondrial uptake and mtDNA alkylation, TPP-
associated fluorescence showed a distinctive staining pattern, similar to that
observed for Complex I which is known to reside within mitochondria.
Cytoplasmic portions of the cell stained diffusely for TPP, whereas the nuclear
compartment was void of any TPP related fluorescence. These results
demonstrate localization and retention of mito-chlor which is consistent with a
mechanism of action that includes mitochondrial uptake and mtDNA alkylation.
Positive staining with the TPP antibody also indicates that Mito-Chlor does not
undergo significant β-oxidation within mitochondria but rather interacts with its
target as an intact molecule.

Figure 4.3 Mitochondrial Membrane
Potential in MDA MB 231 Rho wild-type
and MDA MB 231 Rho θ cell lines. Graph
indicating the ratio of red: green JC-1 mean
fluorescence intensity as a measure of
intrinsic mitochondrial membrane potential
Δψ
mt
for MDA MB 231 Rho wild-type and
MDA MB 231 Rho θ cell lines. Error bars
represent ± standard deviation of samples
analyzed in triplicate.

111
Mito-Chlor activity is dependent on the presence of mitochondrial DNA.

Building upon these results, we sought to determine whether covalent
modification of mtDNA was necessary to the mechanism of action of Mito-Chlor.
To this end, we utilized an MDA MB 231 Rho θ cell line, comparing its sensitivity
to Chlorambucil and Mito-Chlor with that of the MDA MB 231 Rho wild-type cells
from which they were derived. Despite being devoid of mtDNA, Rho θ cells
maintain mitochondrial membrane potential similar to that of their parental
counterpart, (Figure 4.3) making this a suitable model in which to investigate the
mito-chlor mechanism of action. Having equivalent mitochondrial membrane
potential, we can expect similar levels of exposure to treatments in these
syngeneic cell lines, therefore any differences in sensitivity observed in MDA MB
Table 4.5 IC 50 values (µM) of chlorambucil and Mito-Chlor MDA MB 231 Rho wild-type and paired Rho θ
cell lines
compound MDA MB 231 MDA 231 Rho θ
chlorambucil 521.3 ± 249.9 237.5 ± 17.7
Mito-Chlor 9.5 ± 5.7 55.0
chlorambucil wt : rho 0.5
Mito-Chlor wt : rho 5.8

231 Rho θ would likely be due to the inability of Mito-Chlor to interact with
mtDNA. With this in mind, MTT-based cell viability screening was carried out in
the MDA MB 231 Rho θ cell line just as previously described for IC
50

determinations in the breast and pancreatic cancer cell panels. A comparison of
the response to treatment with Chlorambucil shows that sensitivities between
both cell lines was similar but, when MDA MB 231 Rho θ were treated with Mito-
112
Chlor, there was a nearly 6-fold increase in the IC
50
concentration, as compared
to with IC
50
obtained for the MDA MB 231 Rho wild-type parental cell line (Table
4.4). These results demonstrate the requirement of mtDNA alkylation in the
mechanism of action of Mito-Chlor.

Mito-Chlor causes cell cycle arrest consistent with mtDNA damage.

To further understand the mechanisms of action of the non-specific parent drug
and its mitochondria-targeted derivative we compared cell cycle progression in
response to Chlorambucil and Mito-Chlor. MIA PaCa-2 cells were treated with
each compound at their respective IC
50
concentrations for 24, 48 and 72 h and
DNA content was analyzed by flow cytometry (Figure 4.4). At 24 h post-
treatment, cells treated with chlorambucil arrested at the G2/M phase, whereas
treatment with

Figure 4.4 Cell cycle distrubtion of MIA PaCa-2 at various timepoints (0-72 h) following
treatment with 20 µM chlorambucil or 2 µM Mito-Chlor as determined by flow cytometry.
113

Mito-Chlor increased the number of cells in S-phase accompanied by a
secondary block in G2/M progression. By 48 h, the cell cycle distribution was
similar for each of the treatments, most notably the marked arrest in G2/M phase.
The percentage of hypodiploid cells increased throughout the duration of
treatment and by 72 h accounted for the highest fraction of stained cells in both
treatment groups, resulting from loss of cells in G2/M. Although chlorambucil is
not considered a schedule dependent drug, as a DNA alkylator it does activate
the DNA damage checkpoint causing arrest of cells in the G2 phase of the cell
cycle(Amrein et al., 2007). Damage to mtDNA via oxidative stress or adduct
formation results in S-phase arrest accompanied by a moderate G2/M blockade
and disruption of mitochondrial membrane potential with CCCP has been
demonstrated to cause G2/M arrest(Koczor et al., 2009; Mart√≠nez-Diez et al.,
2006). The cell cycle perturbations induced by Mito-Chlor at 24 h are consistent
with previous reports on the role of mtDNA damage in cell cycle arrest, lending
additional support to our hypothesis that mtDNA damage is central to the
mechanism of action of Mito-Chlor. The primarily G2/M arrest noted at 48 h is
consistent with the loss of mitochondrial membrane potential that is known to
precede mitochondrial-mediated apoptosis.

114
Mito-Chlor induced nDNA damage is reversible upon ROS scavenging.

We examined γH2A.X expression as a marker of nDNA damage, expecting that
treatment with Chlorambucil, but not Mito-Chlor would cause an increase in
γH2A.X. Histone H2A.X is rapidly phosphorylated at serine 139 (γH2A.X)
following the induction of double strand breaks, while the lack of histones in
mtDNA allows discrimination of the subcellular location of DNA damage(KUO
and YANG, 2008). We performed western blotting for γH2A.X expression in MIA
PaCa-2 and MDA MB 468 cell lines at various time points following treatment
with Chlorambucil and Mito-Chlor. Contrary to our expectations, both treatments
caused a time dependent increase in γH2A.X (FIgure 4.5 A, B) in each cell line.
Modest increases were seen at early time points (t < 5 h) with more significant
increase occurring later at 5, 8 and 24 h treatment. Mitochondria are a major
source of free radicals within the cell. Dysfunctional mitochondria produce higher
115

Figure 4.5 Western blot for γH2A.X expression (A) in MDA MB 468 treated with 20 µM
chlorambucil (1) or 2 µM Mito-Chlor (2) for indicated periods of time. β- Tubulin expression was
probed as a loading control. B. Western blot for γH2A.X expression in MIA PaCa-2 treated with
75 µM 1 or 2 µM 2 for indicated periods of time. β-Tubulin expression was probed as a loading
control.
116

Figure 4.6 Western blot MIA PaCa-2 treated with chlorambucil (1) and Mito-Chlor (2) for
24 h in the presence and absence of N-acetyl cysteine. Values represent fold change in
γH2A.X expression normalized to GAPDH as calculated by densitometry.

levels of reactive oxygen species (ROS) than their normal counterparts.
Oxidative stress can be linked to genomic instability via ROS induced-damage to
nDNA in the form of single- and double-strand breaks that when inadequately
repaired are capable of triggering apoptosis. For this reason, we sought to
investigate whether DNA damage induced by Mito-Chlor at the level of
mitochondria could produce ROS damage to nDNA and if the resulting nDNA
damage was involved in or required for cell death. In order to determine if ROS
were responsible for the increase in γH2A.X, we treated cells with Chlorambucil
and Mito-Chlor in the presence of the antioxidant N-acetyl-cysteine (NAC) for 24
h . Results obtained by western blot showed a 2.5 and 2.9 fold increase in H2A.X
following Chlorambucil and Mito-Chlor, respectively but only in cells treated with
Mito-Chlor were the phosphorylation levels returned to baseline by the addition of
NAC (Figure 4.6). Despite the ability to reverse signs of nDNA damage, NAC
treatment did not prevent cell death induced by Mito-Chlor (Table 4.5). These
117
Table 4.6 Effect of ROS scavenging on IC 50 values (µM) of chlorambucil and Mito-Chlor in MDA MB 468
cell line
chlorambucil Mito-Chlor
- NAC 50.0 ± 8.1 1.9 ± 0.9
+ NAC 67.5 ± 3.5 0.6 ± 0.3

results indicate that Mito-Chlor is capable of inducing nDNA damage, but it is not
central to the mechanism of cell death, nor is it by the same mechanism as the
parent drug.

Discussion

In recent years, intense efforts have focused on improving the therapeutic index
of chlorambucil by exploiting tumor specific mechanisms of drug delivery.
Generally these attempts have been successful often with the added benefits of
enhancing drug stability and producing modest to substantial gains in
chlorambucil potency (Beyer et al., 1998; Bielawski and Bielawska, 2008;
Bielawski et al.; Clavel et al.; Descoteaux et al.; Descoteaux et al.; Goff and
Thorson; Guaragna et al.; Gupta et al.; Myrberg et al., 2008; Pedersen et al.,
2009; Stark et al., 1992). While in vivo drug stability, selectivity and potency are
desirable, the issue of drug resistance remains a challenge. In an attempt to
evade drug resistance through the selection of sensitive populations, there has
been a resurgence of interest in chlorambucil for the treatment of certain solid
tumors of the breast and pancreas. This renewed attention follows studies
indicating that tumor cell lines derived from patients harboring germ-line
mutations in BRCA1, BRCA2, FANCC, or FANCG gene loci exhibit in vivo
118
hypersensitivity to DNA crosslinking agents due to the defects in the homologous
recombination pathway of DNA repair conferred by these mutations(Evers et al.,
2010; van der Heijden et al., 2005). But without some measure of selectivity, this
approach to treatment stratification would likely serve to enhance the already
high risk of secondary tumors in this DNA repair deficient population. Based on
these observations we concluded that a chlorambucil derivative which could both
selectively target tumor cells and evade common resistance mechanisms would
be of greater overall benefit.

Exploiting phenotypic differences is an effective pharmacologic approach to
impart selectivity between normal and cancerous tissues. In our efforts enhance
selectivity and increase the concentration of chlorambucil at the site of action we
focused on the differences in mitochondrial membrane potential between normal
and cancerous tissue. It is well known that carcinoma cells maintain higher
intrinsic mitochondrial membrane potential (Δψ
mt
) compared to normal epithelial
cells(Chen, 1988a). Furthermore, within a tumor mass there can be
heterogeneity of Δψ
mt
. Tumor cells with higher Δψ
mt
have shown: greater
propensity for tumor formation, increased motility and invasive behavior,
anchorage-independent growth, the ability to survive under low oxygen
conditions and resistance to apoptosis (Heerdt et al., 2005a, 2006a; Heerdt et al.,
2003). Thus agents such as mito-chlor, that accumulate based on Δψ
mt
can offer
selectivity in targeting the most malignant cells within a tumor, while sparing
healthy, normal tissue.
119

Recent efforts at mitochondrial targeting of chlorambucil by Fonseca et al. using
a mitochondrial penetrating peptide composed of unnatural amino acids have
shown enhanced activity in leukemia cell lines compared to the parent drug
(Fonseca et al.). This approach establishes a strong proof of concept for
mitochondrial targeting of DNA alkylators, but the clinical applicability may be
limited by the lack of in vivo knowledge surrounding this type of delivery
mechanism. As a means to direct chlorambucil to the mitochondria we chose the
triphenylphosphonium cation. The efficacy with which triphenylphosphonium salts
(TPP) accumulate selectively within energized mitochondria based on Δψ
mt
is
well documented. TPP are readily concentrated within cells, initially driven by the
potential of the plasma membrane and then further concentrated within the
mitochondria(Murphy, 2008). As a result, differences in concentration between
this organelle and the interstitial space can be up to 1000-fold. The variation in
uptake between cell populations of differing Δψ
mt
have been estimated to be
between 10 and 50 fold. Upon equilibration, TPP cations that do not form strong
bonds with molecules within mitochondria, are released into the cytoplasm and
interstitial space along a concentration gradient. When mitochondria remain
energized, the rate of release is slower than the rate of uptake because the
concentration gradient is opposed by the Δψ
mt
. Upon dissipation of Δψ
mt
, any
unbound TPP is rapidly released from mitochondria(Chen, 1988a).

120
The direct translational feasibility of using the TPP targeting moiety is supported
by numerous in vivo studies demonstrating the potential of TPP based
compounds as mitochondrial targeted therapeutics. Phase I and II studies of
Mito-Q, a mitochondrial targeted antioxidant, validate the in vivo capacity of TPP-
based compounds to accumulate within mitochondria and exert therapeutic
effects without systemic toxicity (Gane et al., 2010; Smith and Murphy, 2010;
Snow et al., 2010). TPP-based PET diagnostic imaging probes have confirmed in
vivo tumor selectivity (Li et al., 2009; Madar et al., 2002; Wang et al., 2007b;
Yang et al., 2008; Zhou and Liu, 2011). Preclinical studies of novel phosphonium
salts including those performed in our own laboratory have demonstrated in vivo
safety and efficacy in animal models of human cancer (Manetta et al., 1996b;
Millard et al., 2010b).

In our current study we found the TPP cation to be an effective means of
delivering chlorambucil to cancer cell mitochondria. Immunohistochemical
staining in breast and pancreatic cancer cells lines both demonstrated uptake
and long-term retention of intact Mito-Chlor within the mitochondria. These
results clearly illustrate the subcellular localization of Mito-Chlor and rule out
direct interaction with nDNA in the mechanism of action. Beyond this, our results
also suggest that TPP conjugation may inhibit the formation of toxic metabolites.
In vivo, the butyric acid side chain of chlorambucil undergoes rapid and extensive
β-oxidation to produce phenylacetic acid mustard (PAAM) (McLean et al., 1980).
PAAM shows similar anti-proliferative properties in cell culture, but has a lower
121
therapeutic index and greater acute toxicity and teratogenicity than that of
chlorambucil (Lee et al., 1986; McLean et al., 1980; Mirkes and Greenaway,
1982). Positive staining of fixed cells with antibody specific for the TPP moiety is
contingent upon both the presence of TPP cation and formation of chloroethyl
covalent adducts. β-oxidation of Mito-Chlor, would liberate the TPP cation,
resulting in loss of signal. Given the robust staining observed, it is possible to
conclude that, unlike chlorambucil, Mito-Chlor is not a substrate for mitochondrial
β-oxidation. Thus if Mito-Chlor cannot be metabolized to PAAM, it may have a
more favorable therapeutic profile in vivo.

The lack of sensitivity to chlorambucil treatment observed in our panels of cancer
cell lines is typical, as the majority of pancreatic cancers are intrinsically resistant
and most advanced breast cancers eventually become refractory to genotoxic
agents (Hidalgo, 2010; Mathews et al., 2011). In both breast and pancreatic
cancers, DSB repair mechanisms have been linked with resistance to genotoxic
therapies (Li et al., 2012; Mao et al., 2009). Chlorambucil damages DNA in
multiple ways, but the primary mechanism of chlorambucil cytotoxicty involves
the formation of N7G : N7G crosslinks (Kondo et al., 2010). The extent and
duration of DNA adducts is a direct determinant of chlorambucil potency
(Hansson et al., 1987). Repair of DNA damage of this type is complex, requiring
the engagement and coordination of multiple DNA repair mechanisms including
Nucleotide Excision Repair (NER), Homologous Recombination (HR) and Non-
Homologous End Joining (NHEJ) (Kondo et al., 2010). In terms of drug
122
resistance, sensitivity to chlorambucil is tempered by intrinsic DNA repair
mechanisms which are often up-regulated in breast and pancreatic cancer
tumors(Kondo et al., 2010; Shaheen et al., 2011). The over-expression of key
DNA repair factors involved in HR and/ or NHEJ enhances the ability to detect
and repair DNA crosslinks and adducts which attenuates the effect of DNA
damage on tumor cell survival. In the case of nitrogen mustards this is evidenced
by 1) the observation that breast and pancreatic tumors competent for relevant
DNA repair factors are significantly less sensitive to chlorambucil compared to
incompetent cell lines (Evers et al., 2010; van der Heijden et al., 2005) and 2)
significant gains in therapeutic efficacy are achieved when combining
chlorambucil with agents that prevent DNA repair such as PARP inhibitors (Evers
et al., 2010).

Mitochondrial targeting of chlorambucil produced substantial gains of potency in
chlorambucil-resistant breast and pancreatic cancer cell lines. The reversal in
resistance caused by mitochondrial targeting was dependent on the presence of
mtDNA as evidenced by the loss of activity in the MDA MB 231 Rho θ cell line.
The marked variation in drug potency between the parent and Mito-Chlor could
be attributed to differences in the rate and extent of DNA repair occurring in the
respective cell compartments. The integrity of mtDNA does not appear to be as
highly guarded as that of the nuclear genome and therefore, targeting mtDNA
may offer a key to overcoming DNA-repair mediated drug resistance. Although
subject to the same genetic insults, compared to the nucleus, mitochondria
123
appear limited in their capacity to repair DNA and thus may be more sensitive to
the broad spectrum damage caused by bi-functional alkylating agents (Cullinane
and Bohr, 1998; LeDoux et al., 1992). Mitochondria lack a functional NER
mechanism and are unable to repair DNA crosslinks, as a result intra- and inter-
strand chlorambucil adducts are long lived within mitochondrial DNA (LeDoux et
al., 1992). Crosslinks interfere with mtDNA synthesis and transcription, induce
DSBs, and cause mtDNA degradation, all of which negatively impact
mitochondrial function. Persistent mtDNA damage in the form of oxidative base
damage and single strand breaks is also sufficient to cause mitochondrial
dysfunction and trigger apoptosis(Li et al., 2012).

Analysis of DNA content revealed distinct differences in cell cycle distribution
between Mito-Chlor and the parent drug. Proliferating cells that sustain damage
to nuclear DNA generally arrest in the G2-phase of the cell cycle. Activation of
the G2-M checkpoint allows for repair of genetic damage prior to entry into
mitosis, but can also trigger apoptosis in the event that nDNA damage exceeds
the cell’s repair capability. In agreement with previous reports in CLL cells,
chlorambucil treated MIA PaCa-2 cells showed early arrest and continued
accumulation in the G2/M phase of the cell cycle. Loss of cells from G2/M phase
was concomitant with an increase in the percentage of hypodiploid cells and
comparable to the observed IC
50
concentration. Perturbation of mitochondrial
function can also result in cell cycle disruption. Mitochondrial dynamics can
influence G1-S phase transition and imbalances in fusion and fission have been
124
shown to drive lung cancer cell proliferation (Owusu-Ansah et al., 2008; Rehman
et al., 2012). Damage to mtDNA has been shown to activate the cell cycle
regulatory kinase, Chk-2, trigger S-phase arrest and cause a secondary block in
G2/M phase (Koczor et al., 2009). In synchronized cells populations, Δψ
mt

increased with progression through G2/M phase (Martinez-Diez et al., 2006).
Conversely, loss of Δψ
mt
causes a G2/M block. MIA PaCa-2 cells treated with
Mito-Chlor exhibited a cell cycle profile that was consistent with a mechanism of
action that involved mtDNA damage at an earlier time point, followed by a decline
in Δψ
mt
. Similar to the chlorambucil treated cells, the percentage of hypodiploid
cells observed at the end of 72 h was in accord with the IC
50
concentration.
Given that changes in cell cycle distribution are related to the mechanism of
action for a compound or drug, these results indicate that altering subcellular
localization can influence mechanism of action.

Despite its mitochondrial localization, Mito-Chlor showed effects on nDNA
damage as indicated by the time and dose dependent increase in Histone H2A.X
phosphorylation. Based on the reversibility observed in the presence of
antioxidant, N-Acetyl cysteine, we concluded that nDNA damage would be the
result of oxidative stress. Several possible explanations for increased ROS
production by mitochondria in response to Mito-Chlor are feasible. ROS
production may be related to the presence of the phosphonium cation. Novel
phosphonium salts discovered in our laboratory have been shown to cause rapid
and sustained increases in mitochondrial superoxide and cytosolic hydrogen
125
peroxide levels along with increased γH2A.X expression (Millard et al., 2010b;
Shabaik et al., 2013). It is also possible that mtDNA damage itself may be the
cause of enhanced ROS. mtDNA damage can directly stimulate production of
reactive oxygen species (ROS) (Tann et al., 2011). In an indirect manner, the
loss of oxidative phosphorylation as a result of mtDNA depletion can also
increase the formation of oxygen radicals capable of inducing nDNA
damage(Tann et al., 2011). The increase in γH2A.X expression could be due to
incomplete repair of non-lethal, non-pathological levels of ROS as mtDNA
depletion can inhibit the repair of oxidative damage in nDNA (Delsite et al.,
2003). It is important to note that the reversal of γH2A.X expression by NAC had
no significant impact on the growth inhibitory properties of Mito-Chlor. These
results indicate that ROS-induced nDNA damage is non-lethal and secondary to
the mechanism of action of Mito-Chlor.

In addition to the pharmacological advantages gained, mitochondrial targeting of
chlorambucil will be an important tool to study the role of mitochondrial DNA
damage in nitrogen mustard- induced cell death. It has long been known that
both the nuclear and mitochondrial genomes are targets of nitrogen mustards,
but the contribution of mtDNA damage in promoting cell death remained unclear
due to an inability to selectively promote damage within this cell compartment.
To this end our study provides strong evidence that mtDNA damage is sufficient
to cause cell death and that this cell death occurs by a mechanism distinct from
that occurring with non-targeted chlorambucil.
126
Conclusions

Nitrogen mustards such as chlorambucil have a long history of safety and
efficacy in chemotherapeutic regimens but drug resistance and risk of secondary
neoplasms remain serious sequelae. Using a novel synthetic method, we
designed a triphenylphosphonium derivative of chlorambucil, Mito-Chlor that
targets cancer cell mitochondria based on Δψ
mt
. This approach to targeting
reversed drug resistance in panels of breast and pancreatic cancer cell lines and
may impart tissue selectivity to reduce both chronic and acute toxicities.
Preliminary mechanistic studies demonstrate Mito-Chlor’s mechanism of action
to be distinct from that of chlorambucil. Taken together, these results provide the
basis for further preclinical evaluation as potential therapeutic for treatment
refractory breast and pancreatic tumors. In addition to its clinical benefits, Mito-
Chlor is also a valuable tool to study nitrogen mustard induced- mtDNA damage.

Experimental Procedures

MTT assay for cell proliferation
Cytotoxicity was assessed by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl
tetrazoliumbromide (MTT) assay (Carmichael et al., 1987b). Cells were seeded
in 96-well tissue culture treated dishes at a density of 4 x 10
3
cells per well and
allowed to adhere overnight. Cells were subsequently treated with a continuous
exposure to drugs for 72 hours. Media was refreshed and an MTT solution was
127
then added to each well to give a final concentration of 0.3 mg/mL MTT. Cells
were incubated with MTT for 4 hours at 37°C. After removal of the supernatant,
DMSO was added and the ABS
570
was determined. All assays were done in
triplicate. The IC
50
was calculated for each treatment from a plot of log drug
concentration versus percentage of cell kill.

Materials
Chlorambucil was purchased from Oakwood Products, Incorporated. Anti-
triphenylphosphonium anti-sera was a generous gift from Dr. Mike Murphy.

Cell Culture
The cell lines used in this study were purchased from the American Type Culture
Collection (ATCC) with the exception of the MDA MB 231 Rho θ and MDA MB
435 Rho θ cell lines which were generated in-house using the method described
by Hashigushi (Hashiguchi and Zhang-Akiyama, 2009). Cell lines were
maintained in the appropriate growth media (DMEM for MDA-MB-435, MDA-MB-
435 Rho θ, MDA-MB-468, BT549, MCF7, SKBR3, MIA-PaCa-2 and Panc-1 cell
lines, and RPMI for the BxPC3, MDA-MB-231, MDA-MB-231 Rho θ, Hs578T,
CAMA-1, cell lines) supplemented with 10% fetal bovine serum and 2 mM L-
glutamine at 37°C in a humidified atmosphere of 5% CO2. MDA-MB-435 Rho θ
and MDA-MB-231 Rho θ cell lines were additionally supplemented with 50 µg/mL
uridine. All experiments were performed in growth media using sub-confluent
cells in the exponential growth phase. For use in tissue culture experiments,
128
chlorambucil and Mito-Chlor were prepared at 100 mM and 10 mM
concentrations respectively, in sterile dimethylsulfoxide (DMSO) and stored at
−20°C in single-use aliquots.

Analysis of DNA content by flow cytometry
MIA PaCa-2 grown in 6 well culture dishes were treated with IC
50
concentrations
of chlorambucil and Mito-Chlor. At the indicated time points, cells were harvested
with trypsin and washed with 1x PBS prior to fixation and permeablization with
70% ethanol. Fixed cells were stained in a 10 µg/mL solution of propidium iodide
containing 100 µg/mL RNase A prior to analysis for DNA content. Sample
readings were performed using a BD-LSR II flow cytometer equipped with a 488
nM diode-pulse solid state laser and 550 nM dichroic mirror and 575/26
bandpass filter.

Confocal microscopy of fixed cells
For determination of the subcellular localization of Mito-Chlor, MIA PaCa-2 were
seeded on glass coverslips at a density of 50,000 cells and allowed overnight to
adhere. The following day, MIA-PaCa-2 were treated with 100 µM chlorambucil
or 10 µM Mito-Chlor for 18 h. At the end of treatment, media was removed and
cells were washed with 500 µL 1x PBS prior to fixation with 3.7% formaldehyde
for 15 m at RT. Fixed cells were washed with 500 µL 1x PBS prior and
subsequent to permeablization with ice-cold acetone for 5 minutes at -20C.
Coverslips (were blocked for 30 m with 1 % bovine serum albumin (BSA) in PBS
129
to inhibit non-specific antibody binding prior to incubation overnight at 4C with
immuno-capture antibody raised against Complex I or anti-sera recognizing the
triphenylphosphonium moiety, each diluted 1:1000 in 1% BSA/PBS. Antibodies
were removed and coverslips were washed 500 µL 1x PBS with gentle agitation.
Goat-anti-rabbit -Cy5 conjugated and Goat-anti-mouse Cy3-conjugated
antibodies were diluted 1:200 in 1% BSA/PBS and incubated with coverslips for 2
h. Coverslips were again washed with 1x PBS under gentle agitation, air-dried
and mounted on pre-cleaned glass slides using Prolong Gold anti-fade mounting
media. Images were obtained using a Lieca SP2 scanning confocal microscope
equipped with 488 nm argon and 633 nM krypton lasers and Leica Confocal
software v 2.61.

SDS-PAGE and Western blotting
For molecular studies, whole cell lysates were prepared from MIA PaCa-2 and
MDA-MB-468 cells treated for varying durations with their respective IC
50

concentrations of chlorambucil or Mito-Chlor in the presence or absence of 10
mM N-acetylcysteine. To this end, cell monolayers were washed with 1x PBS
and scraped into ice-cold RIPA buffer [50 mM Tris HCl pH 8.0, 150 mM sodium
chloride, 1 % Nonidet P-40, 0.5 % sodium deoxycholate, 0.1 % sodium dodecyl
sulfate] containing 1x sigmafast protease inhibitor cocktail. Lysates were cleared
by centrifugation at 14,000 rpm for 10 minutes at 4 °C. For SDS-PAGE, 25 µg
protein prepared in SDS loading buffer containing 10 mM DTT was loaded onto a
15 % polyacrylamide gel and resolved by electrophoresis for 2 hours at 100 volts.
130
Proteins were transferred to PVDF membrane. Following electrophoretic transfer,
membranes were blocked 1 h with 5 % non-fat dry milk prepared in Tris-buffered
saline / 0.1 % Tween-20 (NFDM/ TBST). Primary antibodies were prepared in
either 5 % bovine serum albumin (BSA/ TBST) or 5 % NFDM / TBST according
to the manufacturer’s guidelines. Secondary antibodies were prepared in
NFDM/TBST. Blots were incubated with primary antibodies (1 :1000 dilution)
overnight at 4 °C and with horseradish peroxidase-conjugated (HRP) secondary
antibodies (1:5000) for 2-4 h at RT. Membranes were washed 3 x 5 minutes with
TBST following both primary and secondary antibody incubations. HRP-
catalyzed chemiluminescence was generated using Durawest supersignal
enhanced chemiluminescence substrate. Signals corresponding to levels of
protein expression were visualized using Biorad Chemidoc XRS + CCD camera
and QuantityOne detection software

JC-1 staining for Δψ
mt
MDA MB 231 Rho wild-type and MDA MB 231 Rho θ cells were collected with
trypsin followed by centrifugation at 1200 RPM for 5 minutes at RT. Cell pellets
were washed with 1x PBS, centrifuged and resuspended in basal media
containing 10 µg/mL JC-1 dye. Cells were incubated for 15 minutes in a 37 C
water bath. In preparation for analysis cells were centrifuged at 5000 RPM for 5
minutes at RT in a microcentrifuge. Pelleted cells were resuspended in 1x PBS
and kept on ice. Mitochondrial membrane potential was determined as the ratio
of mean red fluorescence intensity to green fluorescence intensity. Mean
131
fluorescence intensity was measured on a BD LSRII flow cytometer using
detectors equipped to capture light emitted in the 525 and 590 nM range
following excitation at 488 nM.
132
Chapter 5
A mitochondrial-targeted doxorubicin derivative, Mito-Dox

Introduction

Doxorubicin is an anthracycline antibiotic used in the treatment of cancers. It is
among the most potent anti-cancer agents and thus, since its introduction in the
1970’s, it has become a mainstay in chemotherapy regimens treating breast
cancer, childhood solid tumors, and lymphomas. Doxorubicin has broad
spectrum activity. Multiple mechanisms of action have been attributed to
doxorubicin including DNA intercalation and damage, Topoisomerase II
inhibition, ROS production and redox cycling, and aberrations in Ca
+2
signaling
(Gewirtz, 1999).

As a caveat to its efficacy, doxorubicin carries the risk of very serious side-
effects. Cardiotoxicity and congestive heart failure can arise during the course of
treatment or may manifest years after cessation. Doxorubicin-induced
cardiotoxicity is not responsive to therapeutic interventions and can lead to heart
failure and death (Minotti et al., 2004). The risk of cardiotoxicity increases with
total drug exposure, so in an effort to balance risk and benefit, cumulative dosing
is limited. Extravasation injury is another complication common among the
anthracyclines. Injury of this type is quite painful and at its most acute, can
133
cause extensive tissue necrosis requiring surgical intervention and skin grafting.
Extravasation injuries are prevented by use of a central administration port.

Chemotherapy is integral to the treatment of breast cancer. Combination or
sequential chemotherapy is used in treatment across all stages of disease, in
both the neoadjuvant and adjuvant setting. Doxorubicin is a mainstay of breast
cancer chemotherapeutic regimens. Given the favorable prognosis and 5-year
survival rates achievable for most breast cancer cases, delayed cardiotoxicity
represents a serious risk. Triple negative breast cancer patients could benefit
greatly from risk-reduction as this cohort is highly reliant upon chemotherapy as a
treatment modality.

Mitochondrial drug-delivery was chosen as an approach to limit anthracycline-
related toxicities. To accomplish this, two novel phosphonium salt derivatives of
doxorubicin were designed, synthesized and screened in a panel of breast
cancer cell lines including the triple negative cell lines, MDA MB 468, MDA MB
231, BT549, and Hs578T.

134
Results

Synthesis of doxorubicin
derivatives Mito-dox and Mito-
dox-hexanoamide

Numerous approaches have been
taken to improve upon the activity
of doxorubicin. Over the years,
attempts at chemical modification
of the anthracycline scaffold have provided value information regarding structure-
activity relationships. From these data, it has become clear that DNA
intercalation and Topoisomerase II inhibition are fundamental to anthracycline
cytoxicity and in turn are closely related to structural attributes (Animati et al.,
1996; Guano et al., 1999; Wasowska et al., 2005; Zunino et al., 2001). Figure
5.1 denotes the domains involved in DNA intercalation in blue and the
topoisomerase II- interacting domains are shown in red. In order to preserve the
activity of doxorubicin, we chose the C14 hydroxyl group as the point of
modification for the TPP linker. As a negative control, we also prepared Mito-
Dox-hexanoamide. Here, linkage to the TPP occurs through the 3’ amino group
of the daunosamine sugar. The daunosamine is critical for activity and the 3’
amino group stabilizes drug intercalation. Replacement with bulky substituents at
O
O
OMe
OH
OH O
OH
O
OH
O
NH
2
OH
Me
minor groove binding moieties
enzyme-interacting domain

Figure 5.1 Doxorubicin drug domains
involved in topoisomerase II inhibition.
135
O
O
OMe
OH
OH O
OH
O
OH
O
NH
2
OH
Me
O
O
OMe
OH
OH O
OH
O
O
O
HN
O
O
NH
2
OH
Me
PPh
3
Br
O
O
OMe
OH
OH O
OH
O
OH
O Me
OH
HN O
Ph
3
P
HCO
2
doxorubicin Mito-Dox Mito-Dox-Hexanoamide

Figure 5.2 Chemical Structures of doxorubicin, Mito-Dox and Mito-Dox-Hexanoamide

this position abolishes topoisomerase II inhibition (Pratesi et al., 1998; Takagi et
al., 1998; Zunino et al., 2001). Therefore TPP linkage via the daunosamine
nitrogen was expected to diminish activitiy via stearic hindrance of DNA binding.

Mito-dox was prepared using a simple, one-step synthetic method beginning with
the known doxorubicin intermediate and (3-aminopropyl) triphenylphosphonium
bromide (Nagy et al., 1996). Use of the doxorubicin intermediate allowed for
protection of the daunosamine sugar and provided a carboxylic acid functional
group suitable for attachment of the TPP linker. The resulting product was
purified by HPLC, lyophilized and resuspended in DMSO prior to use in tissue
culture experiments.

136
Mito-dox-hexanoamide synthesis was carried out in two steps. Initially, the TPP
linker was prepared by reacting triphenylphosphine and bromohexanoic acid.
The newly prepared phosphonium linker was then reacted with doxorubicin. In

Figure 5.3 Micrograph demonstrating mitochondrial specificity of Mito-Dox. Confocal
images of cells treated 10 minutes with either 1 µM doxorubicin, Mito-Dox or Mito-Dox-
Hexanoamide and stained with Mitotracker Green FM.

the absence of a protecting group the daunosamine amine served as the
attachment point. Further purification by HPLC was carried out prior to
lyophilization and resuspension in DMSO.

137
Mitochondrial delivery of doxorubicin using phosphonium salts

Mito-dox and Mito-Dox hexanoamide localization were compared with that of the
parent drug using fluorescent microscopy. Taking advantage of the intrinsic
fluorescent properties of doxorubicin, we compared doxorubicin-related
subcellular staining with that of MitoTracker Green FM. Cells treated with the
parent drug exhibited doxorubicin-related fluorescence primarily in the nuclear
compartment with low levels of staining co-localizing with MitoTracker Green FM
in the mitochondria. Doxorubicin-associated fluorescence showed significant
overlap with MitoGreen FM staining in cells treated with Mito-Dox or Mito-Dox-
hexanoamide. Nuclear staining was absent in both of these samples.
Representative micrographs are shown in Figure 5.3

Table 5.1 IC 50 values of doxorubicin, Mito-Dox and Mito-Dox-hexanoamide
IC 50, µM ± S.D.
cell line Doxo Doxo-TPP Mito-Dox-hexanoamide
MDA MB 435 0.2 ± 0.12 1.5 >100
MCF-7 0.1 ± 0.01 0.1 39.3 ± 5.1
MDA MB 231 0.8 ± 1.0 0.8 ± 1.0 >100
MDA MB 468 0.1 ± 0.01 0.1 ± 0.01 11.3 ± 5.8
HS578T 0.6 ± 0.1 0.6 ± 0.1 >100
BT549 0.2 ± 0.1 0.3 ± 0.1 >50, 100, 100

Mito-Dox has antiproliferative activity comparable to doxorubicin

Having confirmed the mitochondrial localization of both doxorubicin derivatives,
we next screened them for anti-proliferative activity in a panel of breast cancer
cell lines. In MTT assay, Mito-Dox retained activity comparable to doxorubicin.
138
The IC
50
values obtained for doxorubicin and Mito-Dox are shown in table 4.1.
Doxorubicin and Mito-Dox were also screened in colony formation assay. MCF7
cells were treated in triplicate with increasing concentration of doxorubicin or
Mito-Dox for 30 m or 24 h, after which time treatments were removed and

Figure 5.4 Colony formation assay. MCF7 cells treated with doxorubicin or Mito-Dox for 6
and 24 h.

replaced with fresh media. Cells were grown until sufficient numbers of colonies
were formed in control well, then fixed and stained with crystal violet (Figure 5.4).
Interestingly, with shorter treatment time, Mito-Dox inhibited colony formation
more potently than the parent drug. This effect, however was reversed at 24 h
treatment. These results are interesting because the shorter treatment time
implies Mito-dox is taken up by cells more rapidly than the parent drug.

139
Aminosugar linkage to TPP moiety abolishes biologic activity

In a similar fashion to Mito-dox, the hexanoamide derivative was screened for
anti-proliferative activity in the breast cancer cell line panel. As expected, Mito-
Dox-hexanoamide had significantly lowered activity in all of the cell lines tested,
requiring concentrations as high as 500 µM to achieve 50 % cell kill (Table 5.1).

Cell cycle profiling reveals similar pattern of arrest in doxorubicin and Mito-
Dox.

For certain drugs including DNA damaging agents, disturbances in cell cycle
progression can be related to the mechanism of action of a drug. Doxorubicin is a
cell-cycle specific drug and as such inhibits progression through the G2/M phase.
Measurements of DNA content were performed in MDA MB 468 cells treated with
doxorubicin, Mito-Dox and Mito-dox-hexanoamide. A dose response was
performed, in order to determine the optimal treatment conditions. Cells were
treated with 25-500 nM doxorubicin or Mito-Dox for 24h. At this time point, and at
these concentrations cell kill was minimal as evidenced by the small percentage
of cells (~2.6 % of total labeled cells) comprising the hypodiploid populations
(Figure 5.5 A). The pattern of cell cycle distribution among diploid cells followed
similar trends in MDA MB 468 cells treated with lower concentrations of
doxorubicin and Mito-Dox (Figure 5.5-B). At concentrations between 25 and 100
nM a dose-dependent block at the G2/M checkpoint was observed. With 200 nM
140
Mito-dox treatment, a higher percentage of cells was accumulated in the S-phase
of the cell cycle than G2/M and with 500 nM treatment an secondary block in the
G0/G1 phase accompanied the S-phase arrest. With doxorubicin treatment,
dosing of 500 nM was required in order to achieve the shift from G2/M to S-
phase arrest.
A

B

Figure 5.5 Mito-Dox induced cell cycle arrest in MDA MB 468 cells (A) ploidy distribution of
cells following 24 h treatment with 0-500 nM doxorubicin or Mito-Dox. (B) Cell cycle
distribution of diploid cells following 24 h treatment with 0-500 nM doxorubicin or Mito-Dox.
141

A B

Figure 5.6 Mito-Dox hexanoamide causes cells cycle arrest but does not increase
hypodiploid content. (A) Ploidy distribution of MDA MB 468 cells treated with 100 and 500
nM doxorubicin or Mito-Dox or 2.5 and 5 µM Mito-Dox-Hexanoamide. (B) Cell cycle distribution
of diploid cells treated with 100 and 500 nM doxorubicin or Mito-Dox or 2.5 and 5 µM Mito-Dox-
Hexanoamide.

Next, time-response experiments were performed using 100 and 500 nM
doxorubicin or Mito-Dox and 2.5 and 5 µM Mito-Dox-hexanoamide. With longer
treatment, both doxorubicin and Mito-Dox increased the percentage of
hypodiploid cells that accumulated over time (Figure 5.6 A, B). Analysis of the
diploid populations of cells treated with 100 nM doxorubicin or Mito-Dox, showed
an early block in G2/M-phase that was sustained through 48 h treatment. By 72
h, the dipoid population of cells were 100% arrested in the S-phase. At higher
concentration, doxorubicin and Mito-dox treated cells were arrested in the S-
phase, but by 48 h, the majority of cells treated with doxorubicin exhibited DNA
142
content consistent with the G2/M phase of the cell cycle. In contrast, by 48 h,
nearly 100% of the diploid population treated with 0.5 µM Mito-Dox were now
arrested in S-phase. By 72 h treatment at these concentrations, more than half of
the total stained cells were hypodiploid in their DNA content. The remainder of
cells were predominantly distributed within the G0/G1 phase, although this likely
was an artifact produced by the fragmentation of DNA in the S-phase arrested
cell population. Cells treated with Mito-Dox-Hexanoamide did not significantly
increase the percentage of hypodiploid cells (< 20 % at 5 µM, 72 h ). Cells
treated with Mito-dox-hexanoamide exhibited a temporary arrest in G2/M phase
at 24 h, but by 48 h cell cycle distribution returned to normal.

Figure 5.7 Western blot of γH2A.X expression in MDA MB treated with doxorubicin, Mito-
Dox or Mito-Dox-Hexanoamide. Cells were treated 6 and 24 h with IC
50
concentrations of
doxorubicin and Mito-Dox or 0.1 and 10 µM Mito-Dox hexanoamide.

DNA-damage response following doxorubicin and Mito-dox treatment

Doxorubicin can induce DNA damage by more than one mechanism. Doxorubicin
inhibits Topoisomerase II by stabilizing the intermediate form of the enzyme
leading to the formation of covalent DNA-enzyme crosslinks. Via its activity as a
143
redox cycler, doxorubicin increases ROS production, which in turn damages
DNA. Histone H2A.X (γH2A.X) is rapidly phosphorylated following DNA damage
and in this capacity serves as a good marker of DNA damage. Mitochondrial
DNA lack histones, thus γH2A.X also serves as a indicator of damage locale. To
survey for the presence of nDNA damage MDA MB 468 cells were treated 6 and
24 h with doxorubicin, Mito-Dox or Mito-Dox-hexanoamide, camptothecin was
used as a positive control. As expected, phopshorylation of Histone H2A.X was
observed at both 6 and 24 h doxorubicin treatment. Somewhat unexpectedly,
given the mitochondrial localization, Mito-Dox also increased γH2A.X expression.
Mito-Dox-hexanoamide did not induce the DNA damage response.

Discussion

Using the triphenylphosphonium cation we were able to successfully reroute the
doxorubicin to mitochondria. Mitochondria are a known target of anthracyclines,
but it is as yet unclear to what extent damage in the nuclear and mitochondrial
compartments contributes to the total cytotoxic action of doxorubicin (Ashley and
Poulton, 2009).

The molecular mechanisms of anthracycline related cardiotoxicity are complex
and multifactoral. Damage to cardiomyocyte mitochondria has been implicated
as one possible mechanism of drug-related cardiotoxicity (Sawyer et al., 2010).
144
But given the likely role of mitochondria in anthracycline anti-tumor activity,
imparting tumor selective uptake is the most feasible approach to risk reduction.

Conjugation of doxorubicin to the TPP cation via modification of the C14 hydroxyl
group retained the anti-proliferative activity of doxorubicin but prevented nuclear
localization in the time course that we observed. Conversely, linkage of the TPP
cation through the daunosamine also allowed for efficient mitochondrial
localization as evidenced by microscopy, but rendered the compound biologically
inactive in proliferation assays. Without further studies of in vitro DNA binding it
is impossible to be certain, but based on the measurements of DNA content and
nuclear DNA damage it is possible to speculate regarding the underlying
mechanisms accounting for the difference in potency between the two Mito-Dox
analogs.

At 24 h, MDA MB 468 cells treated with Mito-Dox-Hexanoamide arrested in the
G2/M phase of the cell cycle. This seemed to be a temporary block because by
48 h, the distribution of cycling cells returned to levels similar to those observed
in vehicle treated cells. The number of hypodiploid cells never reached greater
than 20 % at the highest dose and longest timepoint. This suggests that once in
the mitochondria, Mito-Dox-Hexanoamide was able to interacalate into
mitochondrial DNA but could not properly align and stably interact with DNA. In
the absence of proper binding, any interactions with mitochondrial topoisomerase
145
I were weakened significantly enough that the cell could repair the minor damage
and resume proliferation following a short block in cell cycle.

Phosphorylation of Histone H2A.X followed treatment with both doxorubicin and
Mito-Dox, but not with the inactive doxorubicin analog. Given the number of
factors capable of stimulating a DNA damage response, further studies are
needed to elucidate the mechanism behind Mito-Dox induced phosporylation of
Histone H2A.X. The lack of phosphorylation even at concentrations of Mito-Dox-
Hexanoamide deemed to be anti-proliferative in MTT assay suggests a
mechanism central to the activity of Mito-Dox.

To our knowledge this is the first report of targeting an anthracycline to
mitochondria by conjugation to a triphenylphosphonium cation. In addition to the
potential clinical benefits that accompany an improved therapeutic profile, Mito-
Dox is also an invaluable tool for in vitro study of the contributions of
mitochondrial DNA damage and repair in anthracycline mechanism of action and
toxicity.
Experimental Procedures

Cell Culture
The cell lines used in this study were purchased from the American Type Culture
Collection (ATCC). Cell lines were maintained in the appropriate growth media
(DMEM for MDA-MB-435, MDA-MB-468, BT549, MCF7, and SKBR3 cell lines,
146
and RPMI for the MDA-MB-231, Hs578T, and CAMA-1 cell lines) supplemented
with 10% fetal bovine serum and 2 mM L-glutamine at 37°C in a humidified
atmosphere of 5% CO2. All experiments were performed in growth media using
sub-confluent cells in the exponential growth phase. For use in tissue culture
experiments, doxorubicin and Mito-Dox and Mito-Dox-hexanoamide were
prepared at 10 mM concentrations respectively, in sterile dimethylsulfoxide
(DMSO) and stored at −20°C in single-use aliquots.

MTT assay for cell proliferation
Cytotoxicity was measured using a high-throughput 3-(4,5-dimethylthiazol-2-yl)-
2,5-diphenyltetrazoliumbromide (MTT) assay. Cells were seeded in 96-well, flat
bottom, tissue culture treated dishes at a density of 4 x 10
3
cells per well and
allowed to adhere overnight. Cells were subsequently treated with a continuous
exposure to drugs for 72 hours. At the end of treatment, media was refreshed
and MTT solution was added to each well to give a final concentration of 0.3
mg/mL MTT. Cells were then incubated with MTT for 4 hours at 37°C. After
removal of the supernatant, DMSO was added and the ABS
570
was determined
using a microplate reader. All assays were done in triplicate. The IC
50
was
calculated for each treatment from a plot of log drug concentration versus
percentage of cell kill.

147
Colony formation assay
Cell survival was evaluated by clonogenic assay. To this end, single-cell
suspensions were seeded at a density of 100 cells per well into 96-well, flat
bottom, tissue culture plates and allowed overnight to adhere. The next day cells
were treated up to 24 h with increasing concentrations of drug or compounds.
Upon completion of treatment, drug containing media was replaced with drug-
free growth media and cells were then incubated at 37 °C for 7-10 days to allow
the formation of colonies in control wells. To visualize clonogenic growth, cells
were simultaneously fixed and stained in a solution containing 0.25 % crystal
violet, 80 % methanol and 10% formaldehyde. Excess dye was removed by
repeated washing in ddH
2
O and air dried prior to imaging with CCD camera.

Analysis of DNA content by flow cytometry
MDA MB 468 cells grown in 6 well culture dishes were treated with doxorubicin,
Mito-Dox and Mito-Dox-Hexanoamide at the indicated concentrations for up to 72
h. At the end of treatment, cells were harvested with trypsin and washed with 1x
PBS prior to fixation and permeablization with 70% ethanol. Fixed cells were
stained in a 10 µg/mL solution of propidium iodide containing 100 µg/mL RNase
A prior to analysis for DNA content. Sample readings were performed using a
BD-LSR II flow cytometer equipped with a 488 nM diode-pulse solid state laser
and 550 nM dichroic mirror and 575/26 bandpass filter.

148
Confocal microscopy of live cells
For determination of the subcellular localization of Mito-Dox and Mito-Dox-
Hexanoamide, MDA MB 435 were seeded on glass coverslips at a density of
50,000 cells and allowed overnight to adhere. The following day, MDA MB 435
were stained with 50 nM Mitotracker green FM for 15 minutes and then treated
with 1 µM doxorubicin, 1 µM Mito-Dox, or 10 µM Mito-Dox -Hexanoamide for 30
minutes. Coverslips were washed 3 times in 1 x PBS and mounted with a drop of
basal media on a clean glass slide. Images were obtained using a Lieca SP2
scanning confocal microscope equipped with 488 nm argon and Leica Confocal
software v 2.61.

SDS-PAGE and Western blotting
For molecular studies, whole cell lysates were prepared from MDA-MB-468 cells
treated for varying durations with their respective IC
50
concentrations of
doxorubicin, Mito-Dox and Mito-Dox -Hexanoamide. To this end, cell monolayers
were washed with 1x PBS and scraped into ice-cold RIPA buffer [50 mM Tris HCl
pH 8.0, 150 mM sodium chloride, 1 % Nonidet P-40, 0.5 % sodium deoxycholate,
0.1 % sodium dodecyl sulfate] containing 1x sigmafast protease inhibitor cocktail.
Lysates were cleared by centrifugation at 14,000 rpm for 10 minutes at 4 °C. For
SDS-PAGE, 25 µg protein prepared in SDS loading buffer containing 10 mM DTT
was loaded onto a 15 % polyacrylamide gel and resolved by electrophoresis for 2
hours at 100 volts. Proteins were transferred to PVDF membrane. Following
electrophoretic transfer, membranes were blocked 1 h with 5 % non-fat dry milk
149
prepared in Tris-buffered saline / 0.1 % Tween-20 (NFDM/ TBST). Primary
antibodies were prepared in either 5 % bovine serum albumin (BSA/ TBST) or 5
% NFDM / TBST according to the manufacturer’s guidelines. Secondary
antibodies were prepared in NFDM/TBST. Blots were incubated with primary
antibodies (1 :1000 dilution) overnight at 4 °C and with horseradish peroxidase-
conjugated (HRP) secondary antibodies (1:5000) for 2-4 h at RT. Membranes
were washed 3 x 5 minutes with TBST following both primary and secondary
antibody incubations. HRP-catalyzed chemiluminescence was generated using
Durawest supersignal enhanced chemiluminescence substrate. Signals
corresponding to levels of protein expression were visualized using Biorad
Chemidoc XRS + CCD camera and QuantityOne detection software.

150
Chapter 6
A mitochondrial-targeted temozolomide, Mito-Tem

Introduction

Temozolomide is a second generation, imidazotetrazine DNA-alkylating agent
used in the treatment of glioblastoma multiforme (GBM), refractory anaplastic
astrocytoma in adults and for recurrent brain stem glioma and high-grade
astrocytoma in pediatric patients in the United States. As of this writing, over 500
clinical trails in various stages of completion are reported for temozolomide
(www.clinicaltrials.gov, 2013). A number of these trials evaluate temozolomide in
combination with standard therapies for primary and metastatic tumors of the
central nervous system (CNS), including gliomas, glioblastomas, metastatic
breast, lung and pancreatic cancer. Temozolomide is also under investigation for
the treatment of advanced melanomas, ovarian and pancreatic cancers
irrespective of the presence of brain metastasis (www.clinicaltrials.gov, 2013).

The intense interest in temozolomide for neoplasms of the CNS is based on its
favorable pharmacologic properties that allow a significant concentration of the
drug to penetrate the blood-brain-barrier (BBB). When given orally,
temozolomide can achieve nearly 100% bioavailability in the plasma and as
much as 40% of the total plasma concentration crosses the BBB. For this reason,
temozolomide is able to deliver potent anti-tumor activity in the CNS.
151

Temozolomide is a pro-drug, meaning that it must undergo chemical conversion
within the body to become pharmacologically active. Unlike dacarbazine, an
earlier triazene derivative from this class, temozolomide does not require
metabolic conversion in the liver. When temozolomide is administered orally, the
acidic environment of the stomach stabilizes the pro-drug form. Upon entry into
compartments with mildly alkaline pH such as the circulation and tissue,
temozolomide spontaneously converts to the active drug form (t
1/2
=1.83 h at 37
°C, pH 7.4). The schema presented in Figure 6.1 illustrates the process of drug
activation (Stevens, 2008).

The methanediazonium ion produced by spontaneous decomposition under
physiologic conditions is responsible for the cytotoxic activity of temozolomide.
Unlike other DNA alkylating drugs such as the nitrogen mustards, platinum
complexes and procarbazine, the methanediazonium ion introduces monomethyl
adducts into cellular DNA. Thus, as a mono-alkylator, temozolomide lacks the
characteristic myelotoxicity common to DNA crosslinking agents. DNA
methylation by the active temozolomide metabolite occurs at the O6 and N7
positions of guanine and the N3 position of adenine. The methyl adducts formed
on these purines trigger DNA damage response and cell cycle arrest. Although
lesions occur at any of these three positions, the primary cytotoxic lesion is
152

Figure 6.1 Schema illustrating temozolomide mechanism of action.

considered to be the O6 position of guanine. Adducts at the N7 position of
guanine and the N3 position of adenine are easily removed by mismatch repair
(MMR) and base excision repair (BER) mechanisms. Methyl adducts at the 06
position of guanine, however, exceed the repair capabilities of the MMR system.
During DNA replication, O6-guanine adducts promote mis-matched base pairing.
temozolomide
N
N N
N
N
H
2
N
O
CH
3
O
O
H H
Base
N
N N
N
H
N
H
2
N
O
CH
3
O HO
Base
N
N O
N
N
H
2
N
O
H
O
N
CH
3
N
NH
N
H
N
H
2
N
O
N
CH
3
MITC
Base
H
N
NH
NH
2
H
2
N
O
N N CH
3
NuH
N N CH
2
H
Nu CH
3
H
AIC
Diazomethane
Methyldiazonium
ion
CO2
153
The cell responds with repeated, futile repair attempts that cause persistent
single and double- strand breaks along the DNA backbone and eventually trigger
apoptosis.

Removal of O6-methyl lesions requires the specialized DNA repair enzyme, O6-
methlyguanine methyltransferase (MGMT). MGMT facilitates extra-helical
rotation that allows transfer of the methyl adduct from the purine base to a
cysteine contained within the enzyme active site. In this manner it is considered a
‘suicide’ enzyme, as it forms a covalent bond with the excised methyl. Both the
potency and tumor selectivity of temozolomide have been related to the
expression levels of MGMT. Tumor cells, in comparison to normal tissue
generally express lower levels of MGMT and are predicted to be more sensitive
to the DNA damaging effects of temozolomide. This may account for the
generally mild to moderate side effects of temozolomide compared to other
alkylating agents. Over-expression and up-regulation of DNA repair enzymes are
a mechanism of resistance common to many DNA damaging agents. The level of
MGMT expression in GBM is correlated with response to treatment. GBM
patients exhibiting DNA hypermethylation in the MGMT promoter region and/or
lower MGMT expression and subsequently are more sensitive to temozolomide
than those with normal methylation patterns in the MGMT promoter region.

Temozolomide is administered on a 5/28 day cycle at dosing between 150-200
mg/m2. This regimen has low acute toxicity and virtually no cumulative toxicity.
154
For harder to treat cases, extended dosing regimens are employed. These are
generally cycles in which drug is administered on 7/14 days, 21/28 days or 6/8
weeks (Agarwala and Kirkwood, 2000). As a caveat to the enhanced response,
extended dosing increases the rate of immunosuppressive lymphopenia and may
increase risk of secondary neoplasms (Spiro et al., 2001; Su et al., 2005). Cases
of treatment related myelodysplastic syndrome, lymphomas and acute leukemias
including Philadelphia positive acute lymphoblastic leukemia have been reported
in patients receiving extended dosing regimens of temozolomide (De Vita et al.,
2005; Spiro et al., 2001; Su et al., 2005). In a mouse model of drug-related
mutagenicity, temozolomide was a potent mutagen in bone marrow cells (Geiger
et al., 2006). MGMT is significantly depleted in peripheral blood monocytes
(PBMC) of patients receiving extended dosing temozolomide. Intratumoral
MGMT levels however, were not as severely depleted and varied between
patients (Spiro et al., 2001). Taken together this evidence suggests that the
mutagenic properties of temozolomide arise from G:C to A:T transitions caused
by incomplete DNA repair resulting from MGMT depletion in PBMC.

Treatment related secondary neoplasms are common sequelae following the use
of DNA alkylating drugs and generally arise 5-10 years post therapy. Thus the
actual mutagenic potential of temozolomide may be obscured by the short life
expectancies of glioma and glioblastoma patients. Given that temozolomide is
currently under evaluation for the treatment of a variety of solid and hematologic
155
malignancies, including those with favorable prognosis and life expectancy, the
potential benefit verses long-term risk must be earnestly considered.

In an effort to circumvent MGMT-related drug resistance and decrease the
mutagenicity of temozolomide we have designed and synthesized Mito-Tem, a
triphenylphosphonium derivative of temozolomide. Mito-Tem localized to
mitochondria and exhibited cytotoxic properties similar to, or greater than the
parent drug. In a panel of breast cancer cell lines Mito-Tem was up to 4 times
more potent than temozolomide.

N
N
N
N
O
Me
N
O
NH
2
N
N
N
N
O
Me
N
O
NH
Ph
3
P
TFA
Temozolomide Mito-Tem

Figure 6.2 Chemical Structures of temozolomide and Mito-Tem

156
Results

Synthesis of a phosphonium salt derivative of temozolomide, Mito-Tem

Mito-Tem was prepared using a simple, two-step synthetic method (Figure 5.1).
In order to retain temozolomide activity, the C7 carboxamide was chosen as the
point of attachment for the triphenylphosphonium linker. Attachment via this
moiety we reasoned, would not affect opening of the triazene ring and
subsequent liberation of the active methylating species, methanediazonium ion.

Table 6.1 IC 50 values of temozolomide, Mito-Tem in a panel of cancer cell lines
IC 50, µM
Cell line
temozolomide Mito-Tem
U87MG
20,
25
110,
150
HCT116 p53 +/+ > 100 > 100
HCT116 p53 -/- > 100 > 100
Jurkat
75,
180
95,
110
MCF7 > 1000 600
MDA MB 231 > 1000 250
MDA MB 468 > 1000 350
CAMA-1 > 1000 250
BxPC3 180 150
MIA PaCa-2 1125±0.53 85±0.7

157
In Vitro Cytotoxicity of Mito-Tem

The IC
50
values obtained for temozolomide and Mito-Tem are shown in table 5.1.
When tested in the human glioma cell line U87MG, Mito-Tem was approximately
5 times less potent than temozolomide. Similar levels of potency were achieved
in the human acute T-cell leukemia cell line, Jurkat and the syngeneic human
colon adenocarcinoma cell lines HCT116 p53 +/+ and HCT116 p53 -/-. Although
the IC
50
concentrations we not achieved in either of the HCT116 cell lines, dose
response curves were similar between cell lines suggesting the p53 is not
involved in the cytotoxic effects of temozolomide or Mito-Tem. The breast cancer
cell line panel we tested was completely insensitive to temozolomide.
Interestingly, this panel showed variable, yet very mild sensitivity to Mito-Tem.
When tested in the PDAC cell line, BxPC3 both Mito-Tem and temozolomide
exhibited modest activity, at similar concentrations. In the MIA PaCa-2 cell line,
Mito-Tem was significantly more potent than the parent drug.

Discussion

Temozolomide has excellent bioavailability and its toxicity profile is superior to
other commonly used DNA alkylating agents. For these reasons, temozolomide
is under consideration as a possible addition to a wide number of cancer
treatment regimens. Combination treatments with temozolomide and PARP
inhibitors are currently under preclinical evaluation as a means to augment drug
158
potency. Extended dosing of temozolomide is associated with increased risk of
secondary malignancies. This is a major caveat that could potentially limit the
use of temozolomide. Considering the already high potential for mutation
associated with temozolomide treatment, further inhibition of DNA damage
repair, while effective in the short-term could have potentially disastrous long-
term effects. Therefore, it is wise to consider alternative means of improving the
efficacy of temozolomide.

In an effort to reduce the mutagenic potential of temozolomide we designed,
synthesized and screened a novel triphenylphosphonium derivative of
temozolomide. Directing temozolomide to the mitochondria, we rationalized might
be a superior means of enhancing drug activity while simultaneously imparting
tumor-selective drug disposition. The basis of enhanced activity stems from the
lack of MGMT within the mitochondria. Without MGMT, O6-methyl guanine
lesions would persist, causing loss of mtDNA and subsequent loss of oxidative
phosphorylation capacity leading to mitochondrial dysfunction and cell death.

Results of our in vitro cytotoxicity screening showed modest activity for
temozolomide and Mito-Tem. Oddly, Mito-Tem showed a loss of activity in the
glioma cell line, U87MG when compared to the parent drug. The most
remarkable difference in activity between the parent and derivative compounds
was observed in the MIA PaCa-2 cell line. These results should be interpreted
with caution. The IC
50
concentrations of Mito-Tem, although two to four times
159
lower than temozolomide, were relatively high. This could be the result of non-
specific disruption of mitochondrial function by the TPP cation.

There results were somewhat unexpected given the evidence that temozolomide
is capable of entering the mitochondria, interacting with mtDNA and eliciting
measurable damage. Furthermore, mitochondrial targeting of MGMT has been
shown to protect cells from apoptosis as well or better than nuclear targeted
MGMT. The disparity between the expected and observed results requires
further investigation.

The unexpected results obtained when evaluating the cytotoxicity of Mito-Tem
suggest that further experimentation is warranted to better understand the lack of
activity. Efforts to quantitate the damage to mtDNA verses nuclear DNA may
indicate whether Mito-Tem is acting as an alkylating agent within the
mitochondria. The spontaneous conversion of temozolomide at physiological pH
may cause release of the methanediazonium ion to occur outside of the
mitochondria, or alternately allow its release from within the mitochondria.

160
Experimental Procedures

Cell Culture
The cell lines used in this study were purchased from the American Type Culture
Collection (ATCC). Cell lines were maintained in the appropriate growth media
(DMEM for U87MG, MDA-MB-468, and MCF7 cell lines, and RPMI for the
HCT116 syngeneic cell lines, BxPC3, MIA PaCa-2 and CAMA-1 cell lines)
supplemented with 10% fetal bovine serum and 2 mM L-glutamine at 37°C in a
humidified atmosphere of 5% CO
2
. All experiments were performed in growth
media using sub-confluent cells in the exponential growth phase. For use in
tissue culture experiments, temozolomide and Mito-Tem were prepared at 100
and 10 mM concentrations respectively, in sterile dimethylsulfoxide (DMSO) and
stored at −20°C in single-use aliquots.

MTT assay
MTT assay was performed as previously described. In brief cells were seeded in
96-well tissue culture treated plates at densities ranging between 2.5-5 x 10
3

cells per well in a volume of 180 µL of the appropriate growth media. The
following day, cells were treated with drugs and/or test compounds prepared as
serial dilutions at 10 x final concentration. Cells were treated in triplicate for each
given concentration. At the end of 72 h incubation, treatment was terminated by
the removal of drug/compound containing media. Cells were subsequently
incubated for 2-4 h in growth media containing 3 µg/mL MTT. At the end of
161
incubation, media was removed and formazan crystals formed by reduction of
MTT were solublized in DMSO. Absorbance at 570 nM was measured in a
multiwell plate reader.

162
Chapter 7
Novel sulfonamides targeting cancer cell metabolism

Introduction

Pyruvate Kinase (PK) catalyzes the final step of the glycolytic pathway, the
conversion of phosphoenolpyruvate to pyruvate, and in the process generates
one molecule of ATP from ADP. Of the four isoforms of PK expressed in human
cells, the PKM2 splice variant is typically expressed at high levels in embryonic
tissue and proliferating cells but expressed at lower levels in differentiated tissue
(Luo and Semenza). The upregulation of PKM2 provides a proliferative
advantage to cancer cells. Unlike PKM1 which is constitutively expressed as a
tetramer and thus retains high rate of activity, PKM2 can adopt a dimeric or
tetrameric form to decrease or increase the rate of enzymatic activity,
respectively. This dynamic capability is advantageous in that it allows the tumor
cell to ‘budget’ metabolic function in response to glucose and nutrient availability
(Tamada et al.). Adopting the dimeric form, PKM2 can slow the flow of carbon
through the glycolytic pathway to allow the shunting of accumulated metabolites
toward the pentose phosphate and serine synthetic pathways . Tetrameric PKM2
increases the flow of glycolytic metabolites into the mitochondrial TCA cycle.

PKM2 multimerization is controlled by allosteric mechanisms that can favor
tetramer formation or dissociation. In a feed forward manner, fructose 1,6,-
163
bisphosphate (FBP) binds to and promotes PKM2 tetramerization (Ashizawa et
al., 1991a; Ashizawa et al., 1991b). Binding to tyrosine phosphorylated growth
factor receptors such as FGFR and EGFR promotes FBP dissociation and
inhibition of PKM2 dimer formation (Christofk et al., 2008). Elevated ROS can
impair PKM2 tetramer formation via oxidation of cysteine 358 (Anastasiou et al.).
Non-glycolytic, pro-survival functions have been attributed to dimeric PKM2.
EGFR-activated Erk 1/2 phosphorylates PKM2 at serine residue 37 promoting
nuclear translocation (Yang et al.). Once in the nucleus PKM2 acts to stimulate
cell proliferation. Via the phosphorylation of STAT3, PKM2 activates MEK5
transcription (Gao et al., 2012). Nuclear PKM2 acts as a transcriptional co-
activator for β-catenin mediated transcription of c-Myc (Yang et al.; Yang et al.,
2012b). It can also promote c-Myc and cyclin D1 transcription via the
phosphorylation of Histone H3 (Lu; Yang et al., 2012a).

Pharmacologic activators that lock PKM2 in the highly active tetrameric
conformation could compromise cancer cells’ capability to adapt to metabolic
fluctuations. Additionally PKM2 activators could be used to inhibit the proliferative
functions of PKM2 within the nucleus. To this end, we recently identified a series
of novel sulfonamides showing anti-proliferative activity in a panel of cancer cell
lines. Target identification returned PKM2 as a potential target for AV220.
Mechanistic studies indicate rapid and sustained phosphorylation of AMPK and
its direct target, Acetyl carboxylase CoA. Furthermore, under nutrient limited
conditions AV220 activated autophagic flux in a time and dose dependent
164
manner in multiple cell lines of varied origin. The increase in autophagy was
concomitant with activation of the UPR including Grp78, CHOP and ATF4. An
increase in levels of phosphorylated Src, Erk and p38 preceded the UPR and
autophagic response. Bioinformatics analysis identified several cancer types in
which PKM2 was elevated compared to normal matched tissue. Increased PKM2
expression in gastric cancers correlated with tumor progression and increases in
microsatellite instability. When tested as a single agent in a mouse xenograft
model of human colon adenocarcinoma, AV220 inhibited tumor growth by as
much as 86% compared vehicle treated tumors. Combined erlotinib and AV220
treatment had increased anti-proliferative effects compared to either drug alone.

Results

Discovery of AV220
Table 7.1 IC 50 values of AV220 in a panel of cancer cell lines.
AV220 IC 50, µM
MDA MB 435

MCF7

MDA MB 468

MDA MB 231

T47D

6.4 ±2.5

12.0

21.3 ± 5.3

25.0

10.0

BT549

U87MG

MIA PaCa-2

BxPC3

22.0

22.0

20.3 ± 1.5

9.6 ± 6.7

165
In the course of screening our in-house chemical library for small-molecules
having anti-proliferative activity, we discovered AV220, a novel sulfonamide.
AV220 exhibited moderate activity in a panel of cancer cells lines (Table 7.1).

Structural Analogs of AV220

We next performed a search of our chemical library to identify possible structural
analogs of AV220. The results of our search retrieved approximately 150
chemical entities with close structural similarity representing 6 unique scaffolds.
These compounds were screened MDA MB 435, MDA MB 468 and BxPC3 cells
lines. Compounds showing activity in MTT assay are listed in Tables 7.2-7.5.

DARTS assay for target identification

The DARTS protease foot-printing assay was performed in hopes of identify the
possible target(s) of AV220 and its close analogs. The principle behind DARTS is
that small molecules, when bound to their target protein will protect portions of
that protein from proteolytic digestion. The protected protein fragments can then
be resolved by SDS-PAGE and compared to the banding patterns produced by
protelysis in the absence of compound. Bands that are distinct in the small
molecules treated samples can be excised and sequenced by Mass
Spectrometry to identify potential targets. BxPC3 were selected for use in target
identification based on their moderate sensitivity to AV220 and several close
166

Table 7.2 Scaffold 1 AV220 analogs

N
N
N
N
N
R
2
OH
N
S
O
O
S
N
O
O
R
1
H

1

IC 50
BxPC3
R 1 R 2 Compound
MDA
MB
435
MDA
MB
468
MDA
MB
231
MCF
7
MTT
IC 50
SA
MIA
PaCa-
2

Cl Cl

AV220
6.4
±2.5
21.3
± 5.3
25 12

9.6 ±
6.7

5
20.3 ±
1.5

57D12 18 10

57D10 16 7
16.0
± 1.0
7.5

Cl Cl

57B1 9
6 ,
18
1.8 11 6 12.5

Cl

57H2 9 >50 >50 12 >50 NA

Cl

58B5 10

Cl

57C5 13 15 20 10 20 NA
167

F

57H1 9.5 3

57E1 13 >50 > 50

Cl

57H4
12,
15
50 >50

F

57C10 12 15

57G11 15

57G7 1.2 >50

57E7 >20 3.5 >50

F
Cl
57C2 >20 15

168

Table 7.3 Scaffold 2 AV220 analogs

N
N
N
N
N
R
2
OH
N
S
O
O
S
O
O
R
1

2

IC 50, µM
BxPC3
R 1 R 2 Compound
MDA
MB
435
MDA
MB 468
MDA
MB
231
MTT
IC 50
SA
MIA
PaCa-2
N

Cl
Cl

57H11 1.8

50
N

57D11 25 3.5

N

Cl

57C4 4.5

N

Cl

58A6 1.8

25
N

F

57F12 10 15

> 50 1.25
N

F

57D6 15 12

N

Cl
Cl

38C6 30 <0.5

169
N
O

Cl Cl

57B3 10.0
1.5
5.0
18 .0

2.8
5.0

> 50.0
N
O

Cl
Cl

57G5 15.0
2.5,
3.0
>50.0

18.0
1.2
MDA
MB 231
12.5
BxPC3
10.0
> 50.0
BT549
15.0
U87MG
20.0
N
O

58B4 1.5

7.5 45.0
N
O

Cl

57C6 12.0 9.0

7.5
N
O

57E2 25.0 2.0

>15 .0
N
O

58B9 9.0

N
O

Cl

57F10 >50 20.0

7.5
N

45D4

3.8 ±
3.8

>50
N

57G8
3.1 ±
0.5

32.0 12.5 28.0
N

F

45C11 3.0

28.0
N

57H9 3.5

N

57F6 10.0
4.5,
12.0

> 50.0 1.25
170
N

Cl Cl

58A12 15.0
0.9,
<0.5

4.5 >50.0
N

F
Cl

57C1 20.0 4.5

N

F

57D5 10.0 3.5

N

Cl

57D7 15.0 3.5

N

Cl

57E12 >50.0 9.5

N

Cl
Cl

58A10 >50.0

171
Table 7.4 Scaffold 3 AV220 analogs
N
N
N
N
N
R
2
OH
N
S
O
R
1
O

3
IC 50 , µM
R 1 R 2 Compound
MDA MB 435
MDA MB
468
(CH 2) 2CH 3
Cl Cl

56A10 >20.0 1.2
F
F
F

Cl

57F1 >20.0 5 .0
O
F
F
F

Cl

55A4 >20.0 2.5
O
O

Cl

55A7 >20.0 11.0
N
O
N
N
N

Cl

57C3 >20.0 6.0
N
N
N
N
O
O

38C8 >20.0 1.5

172
Table 7.5 Scaffolds 5 and 6 AV220 analogs
N
N
N
N
N
R
2
OH
N R
1
O

5

IC 50 , µM
scratch
assay
R 1 R 2 Compound
MDA
MB 435
MDA
MB 468
BxPC3

Cl Cl

58A4
5.8
3.2
25.0

Cl Cl

57A6 > 20.0 > 50.0 5.0
N
N
N
N
N
R
2
OH
N
S
R
1
O
O

6

IC 50 , µM
scratch
assay
R 1 R 2 Compound
MDA
MB 435
MDA
MB 468
BxPC3
Cl

Cl

56D10 15.0 10.0
5.0

3.5
<0.5
2.8

MTT
1.0, 1.8
SA 10.0
S
N
O
O

Cl

56H10
MCF7

9.0

MDA
MB 231

12.0

BT549
12.0
U87MG
25.0

173

Figure 7.1 Target identification using DARTS assay and MS. All marked bands were excised
from gels. Bands 1, 3, and 4 were submitted for analysis by Mass spectrometry.
174

Figure 7.2 Target validation using DARTS-western blotting BxPC3 cell lysates were
incubated with increasing concentrations of AV220, 500 µM paclitaxel (P), dorsomorphin (D) or
17-DMAG (17) and probed for A) Hsc70, B) β-tubulin, C) Grp78 and D) GAPDH.
175

Figure 7.3 Upstream inhibition of the glycolytic pathway counters effects of AV220
treatment on cell proliferation. Percent survival in MDA MB 468 treated with AV220 and 2-
DG in A) galactose containing media B)glucose containing media

analogs. An image of the banding pattern observed following digestion is shown
in Figure 7.1. Excised bands are donated by numbered arrows. Using the
DARTS assay in combination with Mass Spectrometry we were able to identify
several putative targets (Table 7.6). From the list of possible targets, we chose to
examine Grp78, HSC70 and PKM2. In binding assays, AV220 failed to inhibit the
activity of Grp78 and Hsp70 (data not shown). IN DARTS-western blot assay
using Hsc70 antibody a dose response for protein protection was observed
(Figure 7.2). Results of DARTS-Western using PKM2 antibody are pending.
Table 7.6 Putative AV220 Targets
Band # Protein Score Coverage
1 Grp78 score 33.23 coverage 17.28
1 Hsc70 score 15.63 coverage 9.19
3 PKM1/2 score 72.79 coverage 31.64
4 Hsp90 score 76.19 coverage 25.00
4 Hsc70 score 56.27 coverage 46.47

176
Mechanistic Studies for PKM2 target validation

In lieu of a binding assay and due to bioisosteric similarities shared with known
PKM2 activators, we suspected PKM2 to be the true target of AV220. In order to
provide evidence we opted to perform mechanistic studies to confirm the
possibility that PKM2 might be a target of AV220. To this end we carried out a
proliferation assay combining AV220 with the glycolytic inhibitor 2-deoxyglucose
in cells grown in the presence of glucose or galactose. Presumably in the
presence of galactose, flux through the glycolytic pathway would be
compromised, thus addition of 2-DG would counter the activation of PKM-2 and
rescue cell death. Based on the results of our MTT assay, we found that cells
grown in galactose could be rescued from AV220 induced cell death by the
addition of 2-DG (Figure 7.3). These results were not observed in cells grown in

Figure 7.4 Western blot, MDA MB 468 treated for indicated time points with 20 µM AV220

glucose containing media. Given its specialized role in the glycolytic pathway
these results implicate PKM2 as a putative target of AV220.
177

AV220 activates AMPK and inhibits Acetyl carboxylase-CoA

AMPK is considered the master metabolic regulator of the cell. When nutrients
are limiting or the ADP/ATP ratio increases, AMPK activation promotes a
metabolic shift from energy requiring processes to energy consuming processes
in an effort to regulate homeostasis. In this capacity AMPK has tumor
suppressive functions, it has been shown to negatively regulate the Warburg
effect and slow tumor formation in vivo (Faubert et al., 2013). Based its
contribution to metabolic plasticity we hypothesized that activation of PKM2 might
lead to energy imbalances and thus activate AMPK. To this end, we treated
MDA MB 468 cells with AV220 and observed by Western blotting the expression
and phosphorylation levels in AMPK at time points from 5 minutes to 72 h (Figure
7.4). Phosphorylation of AMPK occurred as early as 30 minutes following
treatment and continued through 24 h. Acetyl carboxylase-CoA (ACC) catalyzes
the primary and rate limiting step of fatty acid synthesis and is a direct target of
AMPK (Kemp et al., 2003). Phosphorylation of ACC by AMPK inhibits its
function. Concomitant with AMPK activation, ACC phosphorylation was evident
as early as 30 min but not at its maximal level until 2 h following treatment.
Phosphorylation was sustained over 24 h treatment. Similar changes in AMPK
phosphorylation were obtained in MIA PaCa-2 cells treated with AV220 (data not
shown).

178

AV220- induced growth arrest and changes in cytoplasmic morphology
under nutrient limited conditions

Figure 7.5 AV220 treatment under nutrient restricted conditions induced G0/G1 cell cycle
arrest. MDA MB 435 cells grown in basal media were treated with increasing concentrations of
AV220. Tracings represent the DNA content of propidium iodine stained cells. AV220 caused a
dose-dependent increase in G0/G1 content.

Under conditions of nutrient limitation, activators of PKM2 have been shown to
cause metabolic disturbances in serine biosynthesis leading to auxotrophy and
activation of ATF4 target genes (Kung et al.). Therefore as a means to validate
PKM2 as the in vivo target of AV220, we cultured MBA MB 435 cells in basal
media and evaluated their response to AV220. Cell cycle analysis of MDA MB
435 treated 24h showed dose-dependent accumulation in the G0/G1 phase of
the cell cycle (figure 7.5). Upon visual examination of cells treated o/n with
AV220 we noted intense vacuolization of the cytoplasm (Figure 7.6).
179

Figure 7.6 AV220 treatment under nutrient restricted conditions caused extensive
intracellular vesicle formation. Micrographs of vehicle and AV220 treated MDA MB 435
cells.

LC3B associated with AV220-induced cytoplasmic vesicles

Autophagy is an adaptive response employed by cells under conditions of stress.
We hypothesized that the vacuoles forming in response to AV220 treatment
could be caused by disruption in autophagic flux. To confirm our hypothesis we
treated cells grown in BME with AV220 and stained them with antibody
recognizing LC3B. As shown in Figure 7.7, LC3B appeared to stain the
perimeters of the cytoplasmic vesicles in a pattern distinct to that observed in
vehicle or rapamycin treated cells. Based on these results we concluded that
AV220 may acting on, or stimulating the autophagic machinery.

180

Figure 7.7 LC3B associates with AV220 induced-vesicles Immunofluorescence
Micrographs of vehicle and AV220 treated MDA MB 435 cells stained with LC3B-cy3 and
DAPI.

AV220 stimulates flux through the autophagic pathway

Autophagy is a dynamic process involving the nucleation, elongation and
maturation of double membrane vesicles, autphagosomes and their eventual
fusion with lysosomes to form autolysosomes. This process from beginning to
end is referred to as autophagic flux (Yang and Klionsky, 2010). Conversion of
LC3B-I to LC3B-II is a marker of autophagy but alone it provides only static
measurements. In order to determine if AV220 was acting to disrupt autophagic
flux we combined AV220 treatment with pepstatin A and e64d, inhibitors of
lysosomal fusion and compared the levels of LC3B conversion to that of vehicle
and AV220 cells. If AV220 were acting to increase autophagic flux we would
expect to see an increase in LC3B-II levels when combined with inhibitor.
Conversely, if AV220 were acting to inhibit the flow of autophagy up or down
stream of lysosomal fusion no significant increase in LC3B-II would be seen.
Results of western blotting are shown in Figure 7.8 demonstrate a marked
increase in the levels of both LC3-I and LC3B-II in response to interruption of late
181
stage autophagy. These results suggest that AV220 is stimulating flux through
the autophagic pathway.

Figure 7.8 AV220 increases autophagic flux. MDA-MB-435 were treated 48 h with AV220 in
the presence or absence of pepstatin A, e64d, rapamycin was used as positive control.

The response to AV220 is not cell-line or cancer type specific

The unique morphological changes noted in the MDA MB 435 cell line led us to
question whether these effects might be specific to this particular cell line. To this
end, MIA PaCa-2 pancreatic cancer cells and HCT116 p53 +/+ and HCT116 p53
-/- cell lines were seeded in chambered glass cover slips and observed by light
microscopy following overnight treatment with AV220. In each of these cell lines,
AV220 caused marked cytoplasmic changes indicative of autophagy (Figure 7.9).
Western blotting experiments carried out in HCT116 p53 +/+ cells showed a
dose- and time-dependent increase in LC3B-II expression following treatment
with 10 µM AV220 (Figure 7.10). Similar results were obtained in MIA-PaCa-2
cells (data not
182

Figure 7.9 AV220-induced changes in morphology in a panel of cancer cell lines.

Figure 7.10 AV220 caused a time- and dose-dependent increase in LC3B-II HCT116 p53
+/+, 300 nM Thapsigargin, 18h, 10 µM AV220 various time points
183
shown). The results obtained for colony formation assay in HCT116 p53 +/+
treated with AV220 were similar to those observed for MDA MB 435 in MTT
assay (Figure 7.11).

Figure 7.11 HCT116 Colony Formation Assay

AV220 activates ER stress response pathways

In light of the robust stimulation of autophagic flux observed in response to
AV220 treatment, we opted to delineate the upstream mechanisms responsible.
Induction of Endoplasmic reticulum (ER) stress response proteins is a common
response to glucose and nutrient deprivation and therefore could be involved in
the AV220 mechanism of action. It an attempt restore homeostasis, nutrient
balance and removed mis-folded proteins, cells activate the unfolded protein
response (UPR) (Walter and Ron, 2011). To better understand the role of stress
response proteins in AV220 mechanism of action we performed western blotting
for Grp78 and CHOP. Following 24 h treatment, AV220 induced the expression
of Grp78 and CHOP at levels similar to that observed with the UPR inducer,
thapsigargin (FIgure 7.12-A). In a time dependent fashion, AV220 induced
increases in Grp78, CHOP and ATF4 concomitant to the increase in LC3B-II
conversion (Figure 7.12-B). In cells cultured under normal growth conditions,
AV220 did not activate
184

Figure 7.12 AV220 activates UPR. Western blot in HCT116 p53 +/+ treated (A) with 10 µM
AV220, rapamycin or 300 nM thapsagargin for 24 h. (B) Dose response, HCT116 p53 +/+
treated with 10 µM AV220, 300 nM thapsagargin.

Grp78 or CHOP, nor did it stimulate an increase in LC3B-II conversion despite its
ability to activate AMPK.

AV220 activates signaling molecules involved in the regulation of
autophagy

Akt, Erk and p38 have been implicated in the regulation of various stages on
autophagosome maturation (Lock and Debnath, 2008; Yang and Klionsky, 2010).
Determine what, if any role these signaling events had in the cellular response to
treatment we performed western blotting on MDA MB 435 treated with AV220 for
timepoints between 8 and 48 h (Figure 7.13). AV220 activated p38 and Erk and
185
increased the levels of Src deactivating phosphorylation. Levels of Akt
phosphorylation remained unchanged throughout the time course

Figure 7.13 Erk 1/2, p38 and Src
phosphorylation precede AV220 induced
autophagy. MDA MB 435 cells treated with
10 µM AV220 for 0-24 h, 10 µM rapamycin,
24 h.

PKM2 expression in pancreatic, breast and colon cancer versus matched
normal tissue

AV220 exhibited anti-proliferative activity across multiple cancer types, therefore
in order to determine what types of cancer could be best treated with AV220 in
vivo we used the Oncomine platform to perform bioinformatics analysis of PKM2
gene expression in cancer versus normal tissue. Elevation of PKM2 expression
levels have been noted for most types of cancer, so taking the approach of

186
comparing expression in tumor versus normal tissue we felt might improve
selectivity of targeting AV220 in tumor tissue and reduce any possible untoward
effects on normal tissue. Using stringent search parameters (Thresholds: p-
value < 1E-7, fold change ≥ 2, gene rank 1%) we identified 6 analyses that met
our parameters (Table 7.6). Gene rankings and p-values are listed in Table 7.7
Expanding our search thresholds to include studies in which PKM2 was in the
10% of over-expressed genes (Thresholds: p-value < 1E-7, fold change ≥ 2 ,
gene rank 1%) retrieved an additional 13 analyses (Table 7.8) The gene rank
and p-values are noted in Table 7.9. Of the analyses retrieved by the more
stringent thresholding, pancreatic ductal adenocarcinomas, mucinous breast and
gastric intestinal type adenocarcinoma had the three most significant p-values.
In the two pancreatic cancer analyses, PKM2 expression was 3-5 fold
upregulated compared to normal pancreas (Figure 7.14) (Badea et al., 2008;
Logsdon et al., 2003). Our analysis of the Curtis Breast cancer dataset included
a total of 1781 tissue samples, 147 of which were normal tissue (Figure 7.15)
(Curtis et al., 2012). Here, we found the PKM2 expression was significantly
elevated in the invasive ductal carcinoma, medullary and mucinous breast cancer
tissue samples. When this dataset was sorted based on hormone receptor status
and HER2 expression, some differences were noted in PKM2 expression but the
variability within the sample groups was quite high. PKM2 expression levels were
not significantly different between p53 wild-type and mutant breast cancer tissue
samples in this dataset (data not shown). Although GI-type adenocarcinomas
had the highest fold increase in expression, all of the gastric cancer samples in
187
the D’Errico et al. dataset showed elevated PKM2 expression compared to
normal gastric mucosa (D'Errico et al., 2009). When these samples were sorted
by tumor specific characteristics we found a significant positive correlation
between increasing tumor stage or microsatellite instability (MIN+) and elevated
PKM2 expression (figure 7.16 A & B).

Table 7.7 Analysis Comparison for gene PKM2, upregulation in Tumor vs Normal Tissue
Thresholds: p-value < 1E-7, fold change ≥ 2 , gene rank 1%
Analysis 1: Pancreatic Ductal Adenocarcinoma vs. Normal (Badea Pancreas)
Analysis 2: Leiomyosarcoma vs. Normal (Barretina Sarcoma)
Analysis 3: Mucinous Breast Carcinoma vs. Normal (Curtis Breast)
Analysis 4: Gastric Intestinal Type Adenocarcinoma vs. Normal (Derrico Gastric)
Analysis 5: Pancreatic Adenocarcinoma vs. Normal (Logsdon Pancreas)
Analysis 6: Ovarian Serous Adenocarcinoma vs. Normal (Yoshihara Ovarian)

Table 7.8 Statistical ranking of PKM2 gene analyses

Gene Rank p-value
Analysis 1 103/19574 1.23E-14
Analysis 2 123/12,624 3.58E-12
Analysis 3 66/ 19,273 1.83E-22
Analysis 4 111/ 19,574 4.21E-13
Analysis 5 21/ 5,338 3.09E-09
Analysis 6 97/ 16,724 3.60E-11
Median 100 1.80E-11

188

Table 7.9 Analysis Comparison for gene PKM2, upregulation in Tumor vs Normal Tissue
Thresholds: p-value < 1E-7, fold change ≥ 2 , gene rank 10%
Analysis 1: Pancreatic Ductal Adenocarcinoma vs. Normal (Badea Pancreas)
Analysis 2: Dedifferentiated Liposarcoma vs. Normal (Barretina Sarcoma)
Analysis 3: Leiomyosarcoma vs. Normal (Barretina Sarcoma)
Analysis 4: Myxofibrosarcoma vs. Normal (Barretina Sarcoma)
Analysis 5: Pleomorphic Liposarcoma vs. Normal (Barretina Sarcoma)
Analysis 6: Hereditary Clear Cell Renal Cell Carcinoma vs. Normal (Beroukhim Renal)
Analysis 7: Non-Hereditary Clear Cell Renal Cell Carcinoma vs. Normal (Beroukhim Renal)
Analysis 8: Ovarian Carcinoma vs. Normal (Bonome Ovarian)
Analysis 9: Invasive Ductal Breast Carcinoma vs. Normal (Curtis Breast)
Analysis 10: Medullary Breast Carcinoma vs. Normal (Curtis Breast)
Analysis 11: Mucinous Breast Carcinoma vs. Normal (Curtis Breast)
Analysis 12: Gastric Intestinal Type Adenocarcinoma vs. Normal (DErrico Gastric)
Analysis 13: Squamous Cell Lung Carcinoma vs. Normal (Hou Lung)
Analysis 14: Pancreatic Adenocarcinoma vs. Normal (Logsdon Pancreas)
Analysis 15: Pancreatic Carcinoma vs. Normal (Pei Pancreas)
Analysis 16: Colon Adenoma vs. Normal (Sabates-Bellver Colon)
Analysis 17: Squamous Cell Lung Carcinoma vs. Normal (Talbot Lung)
Analysis 18: Colon Mucinous Adenocarcinoma vs. Normal (TCGA Colorectal)
Analysis 19: Ovarian Serous Adenocarcinoma vs. Normal (Yoshihara Ovarian)

Table 7.10 Statistical ranking of expanded PKM2 gene analyses
Gene Rank p-Value
Analysis 1 103 /19574 1.23E-14
Analysis 2 537 / 12,624 2.66E-08
Analysis 3 123 / 12,624 3.58E-12
Analysis 4 215 / 12,624 4.81E-12
Analysis 5 286 / 12,624 1.24E-08
Analysis 6 458 1.80E-09
Analysis 7 317 6.22E-09
Analysis 8 326 4.57E-12
Analysis 9 567 / 19,273 6.19E-73
Analysis 10 206 / 19,273 3.84E-14
Analysis 11 66 / 19,273 1.83E-22
Analysis 12 111 / 19,574 4.21E-13
Analysis 13 919 3.82E-10
Analysis 14 21 / 5338 3.09E-09
Analysis 15 275 5.43E-08
Analysis 16 740 1.91E-12
Analysis 17 138 4.65E-11
Analysis 18 624 1.03E-10
Analysis 19 97 / 16724 3.60E-11
Median 275 5.43E-08

189

Figure 7.14 PKM2 over-expression in pancreatic cancer tissue samples. Left panel-
Badea et al., right panel Logsdon et al.

Figure 7.15 PKM2 over-expression in breast cancer tissue samples.

190

Figure 7.16 PKM2 over –expression in gastric cancer tissue samples, by type (left), T-
stage (middle) and MIN status (right).

AV220 inhibits the growth of colon adenocarcinoma xenografts in vivo as a
single agent

When embarking on our in vivo studies, we took into consideration of the results
of our bioinformatics analyses and chose to evaluate AV220 in the HCT116 p53
+/+ colon adenocarcinoma cell line. The HCT116 cell line carries a point
mutation in the MLH1 gene and as a result has a high level of microsatellite
instability. Based on our colony formation assays we found HCT116 cells to be
more sensitive to AV220 than other cell lines tested. Once tumors were
established (100 mm3) treatment was initiated (Figure 7.17). Over the course of
38-45 days, mice received increasing doses of AV220 (n=3 , 10mg/kg initially,
arrows indicate escalation to 20, 40 mg/kg respectively). Little to no toxicity was
191
observed in the mice throughout the entire course of treatment. As a single
agent, AV220 significantly reduced HCT116 xenograft growth compared to
vehicle treated mice (p=0.019). Tumor growth was reduced by 72-86 %, when
comparing endpoints in AV220 versus vehicle treated groups. Mechanistic
evaluation of tissue and tumor samples is currently underway.

Figure 7.17 Single agent AV220 inhibits HCT116 p53 +/+ colon cancer growth in a mouse
xenograft model.

AV220 enhances the effect of erlotinib in some resistant cell lines

Erlotinib is a tyrosine kinase inhibitor used in the treatment of pancreatic cancer
(Conradt et al., 2011). It is now recognized as a multi-target tyrosine kinase
inhibitor but was originally marketed as an epidermal growth factor receptor
(EGFR) inhibitor. EGFR is a negative regulator of PKM2, promoting dimerization
and nuclear localization. We evaluated the combined effects of AV220 and
192
erlotinib on cell proliferation in a panel of breast and pancreatic cancer cell lines
(Table 7.11) In the PDAC cell lines MIA PaCa-2 and Panc-1, and both breast
cancer cell lines combination treatment appeared to potentiate the effects of
AV220. The BxPC3 cell line was more sensitive to erlotinib than the other cell
lines but still showed a marked decrease in IC50 value when combined with
AV220.
Table 7.11 IC 50 values of AV220, erlotinib in a panel of cancer cell lines
IC 50, µM
MIA PaCa-2 BxPC3 Panc-1 MDA MB 231 Bt 549
AV220 19, 22 18,12 22 18 22
erlotinib
>10(S=67.2%),
>10 (S=80.0%)
>10 (S=67.3
%), 6.5 µM
>10 (S=60%)
> 10
(S=73.5%)
>10 (S=
62.2%)
AV220 + 0.1
um erlotinib
19, 20 15, 9 22 18 22
AV220 + 1 uM
erlotinib
18, 20 10, 8.5 22 16 20
AV220 + 10 uM
erlotinib
12,15 1.8, <0.5 10 8.5 12
S is percent survival at 10 µM
Discussion

AV220 was first identified during routine screening of our small molecule library
as having moderate anti-proliferative activity in a panel of cancer cell lines. Due
to its favorable drug-like properties, we performed an extensive search of our
chemical database to identify potential structural analogs. Our efforts retrieved 6
unique scaffolds comprising over 150 compounds. High-throughput screening for
growth inhibition identified 35 active hits across 5 out of 6 scaffolds with IC50
values similar to or better than AV220.

Routine library screening is a common approach to drug discovery and has been
very successful in identifying small molecules with anti-cancer activity. In the
193
course of developing these leads, target identification has proven to be a more
significant challenge. As this was the case with AV220, we attempted target
identification using a simple, cost –effective method, the drug affinity responsive
target stability (DARTS). The principle behind DARTS is that small molecule-
binding will shield the target protein from protease digestion. When used with
A220, characterization of the resulting protein fragments retrieved numerous
putative targets.

Mechanistic studies performed in conjunction with the DARTS target
identification strategy provided information that proved insightful and allowed us
to focus in on several putative targets including Hsc70, Grp78, PKM2 and Hsp90.
Under normal growth conditions, AV220 was a potent activator of APMK. When
nutrients were limiting, AV220 caused profound changes in cell morphology and
stimulated an exaggerated response in signaling pathways commonly activated
in response to metabolic stress. These observations indicate that AV220 likely
interacts with key molecules regulating tumor cell metabolism, energy production
and protein stability.

Substructure and similarity searches have confirmed that AV220 and its related
scaffolds are novel small molecules yet due to the immense variability capable
within ‘chemical space’, they share common chemical features with known
activators of PKM2. Guided by co-crystal structures of PKM2 with these
compounds it is possible to infer which chemical features contribute to AV220
194
ant-proliferative activity. In this crystal structure a sulfonamide oxygen interacts
with Tyr390 of the PKM2 backbone and Lys311 can form hydrogen bonds with
an oxygen molecule on an amide (Kung et al.). Replacement of the terminal
cycloalkane group of AV220 with a piperidine or a pyrrolidine rings improved the
anti-proliferative properties. In the PKM2 co-crystal structure a quinoline group at
this position inserted within a pocket formed by Phe26, Leu27 and Met30
residues of the A and B chains. In another co-crystal structure involving a
different class of PKM2 activating small molecule, a piperazine ring formed
hydrophobic stacking interactions between the Phe26 residues of PKM2 (Boxer
et al., 2010).

Proteomics analysis of normal and tumor tissue samples demonstrated
significant upregulation of PKM2 in patient tumor samples derived from similar
cancers as those in which AV220 and its close analogs were screened. In the
process we uncovered an association between MIN status and PKM2 in colon
cancer.

As a single agent in vivo, AV220 showed remarkable anti-tumor activity when
tested in a mouse xenograft model of human colon cancer. Although the
treatment groups were small (n=3), the fact that there was little variability
between the tumor sizes within the treatment group suggests these effects were
real and not due to inconsistency in tumor growth among the mice in the group.
Dose escalation to 20 mg/kg and then later 40 mg/kg was well tolerated and at
195
no time did mice exhibit any physical indication of treatment-related toxicities.
Overall these promising in vivo results support further preclinical development of
AV220 for the treatment of colon cancer.

In drug combination studies, AV220 enhanced erlotinib potency in BxPC3 and
Panc-1 PDAC cell lines and two breast cancer cell lines, MDA MB 231 and
BT549. Erlotinib is used clinically in the treatment of PDAC and CRC there the
translational potential of these findings warrant further study.

Experimental Procedures

Cell culture
MDA-MB-435, MCF7, MDA MB 231 and MDA MB 468 breast cancer and MIA
PaCa-2, BxPC3 pancreatic cancer cell lines were purchased from the American
Type Cell Culture (Manassas, VA). HCT116 p53+/+ and HCT116 p53-/- cells
were kindly provided by Dr. Bert Vogelstein (Johns Hopkins Medical Institutions,
Baltimore, MD). Cell lines were maintained in the appropriate growth media
(DMEM for MDA-MB-435, MCF7, MDA MB 231 and MDA MB 468, RPMI for the
MIA PaCa-2 BxPC3 and HCT116 cell lines) containing 10% heat-inactivated fetal
bovine serum and supplemented with 2 mM L-gutamine at 37º C in a humidified
atmosphere of 5% CO2. For subculture cells were washed with 1x PBS,
detached using 0.025% Trypsin-EDTA, collected in growth media and
centrifuged. Experiments under nutrient replete conditions were performed in
196
growth media using sub-confluent cells in the exponential growth phase. Nutrient
limited conditions were recapitulated using BME supplemented with 10 % FBS.
For use in tissue culture experiments, compounds were prepared at 10 mM
concentration in sterile dimethylsulfoxide (DMSO) and stored at -20ºC when not
in use.

Cytotoxicity assay
Cytotoxicity was assessed by a 3-(4,5-dimethylthiazol-2-yl)-2,5-
diphenyltetrazolium bromide (MTT) assay as previously described (Carmichael et
al., 1987a). Cells were seeded in 96-well tissue culture treated dishes and
allowed to adhere overnight. Cells were subsequently treated with a continuous
exposure to drugs for 72 hours. An MTT solution was added to each well to give
a final concentration of 0.3 mg/mL MTT. Cells were incubated with MTT for 3-4
hours at 37°C. After removal of the supernatant, DMSO was added and the
absorbance was read at 570 nm. All assays were done in triplicate. The IC
50
was
then determined for each drug from a plot of log drug concentration versus
percentage of cell kill.

Colony formation assay
Colony formation assay was performed as previously described to further assess
drug toxicity (Munshi et al., 2005). To this end, cells were seeded in 96 well
tissue culture dishes at a density of 200 cells per well in growth media and
allowed to adhere overnight. Cells were subsequently treated with varying
197
concentrations of compound for 24 h. Following treatment, monolayers were
washed with 1x PBS and incubated in growth media for a period of 7-10 days,
allowing sufficient time for colonies to form in control wells. To visualize the
extent of colony formation, cells were fixed and stained in a 2% solution of crystal
violet containing 1% glutaraldehyde. Excess stain was removed through multiple
washes in distilled water and allowed to air dry. Stained plates were imaged
using Quantity One software running on the VersaDoc imaging platform.
(BioRAD). Quantification by measurement of optical density at 570 nm was
performed after solublization in a 2% solution of sodium dodecyl sulfate
accompanied by 2 h shaking on a platform rocker.

Cell cycle analysis
Cells were seeded in 100 mm tissue culture dishes at a density of 1x10
6

cells/plate in BME media containing 10 % FBS and allowed to adhere overnight.
The following day cells were treated with 1 ,5 and 10 µM AV220, rapamycin or
DMSO alone as vehicle control for 24 h. Upon completion of treatment, cells
were detached with trypsin and both media and cells were collected by
centrifugation. Cells were washed and resuspended in 1x PBS prior to fixation in
ethanol overnight at -20 ºC. Fixed cells were treated with 10 µg/mL RNase A and
stained in a 50 µg/mL solution of propidium iodide. DNA content was determined
by flow cytometry using the BD LSR II equipped with a 488 nM Sapphire™
argon-ion laser and PE emission detector.

198
SDS-PAGE and Western blotting
For molecular studies, whole cell lysates were prepared from cells treated for
varying durations. To this end, cell monolayers were washed with 1x PBS and
scraped into ice-cold RIPA buffer [50 mM Tris HCl pH 8.0, 150 mM sodium
chloride, 1 % Nonidet P-40, 0.5 % sodium deoxycholate, 0.1 % sodium dodecyl
sulfate] containing 1x sigmafast protease inhibitor cocktail. Lysates were cleared
by centrifugation at 14,000 rpm for 10 minutes at 4 °C. For SDS-PAGE, 25 µg
protein prepared in SDS loading buffer containing 10 mM DTT was loaded onto a
15 % polyacrylamide gel and resolved by electrophoresis for 2 hours at 100 volts.
Proteins were transferred to PVDF membrane. Following electrophoretic transfer,
membranes were blocked 1 h with 5 % non-fat dry milk prepared in Tris-buffered
saline / 0.1 % Tween-20 (NFDM/ TBST). Primary antibodies were prepared in
either 5 % bovine serum albumin (BSA/ T[0BST) or 5 % NFDM / TBST according
to the manufacturer’s guidelines. Secondary antibodies were prepared in
NFDM/TBST. Blots were incubated with primary antibodies (1 :1000 dilution)
overnight at 4 °C and with horseradish peroxidase-conjugated (HRP) secondary
antibodies (1:5000) for 2-4 h at RT. Membranes were washed 3 x 5 minutes with
TBST following both primary and secondary antibody incubations. HRP-
catalyzed chemiluminescence was generated using Durawest supersignal
enhanced chemiluminescence substrate. Signals corresponding to levels of
protein expression were visualized using Biorad Chemidoc XRS + CCD camera
and QuantityOne detection software

199
Fluorescence microscopy
For fluorescence microscopy MDA MB 435 cells were seeded on chambered
glass coverslips at a density of 50,000 cells per well. Next day, cells were
treated with either rapamycin, thapsagargin, AV220 or analog. At the end of
treatment, media was removed and cells were washed with 500 µL 1x PBS prior
to fixation with 3.7% formaldehyde for 15 m at RT. Fixed cells were washed with
500 µL 1x PBS prior and subsequent to permeablization with ice-cold acetone for
5 minutes at -20C. Coverslips (were blocked for 30 m with 1 % bovine serum
albumin (BSA) in PBS to inhibit non-specific antibody binding prior to incubation
overnight at 4C with antibody raised against LC3B, calreticulin or CHOP, each
diluted 1:1000 in 1% BSA/PBS. Antibodies were removed and coverslips were
washed 500 µL 1x PBS with gentle agitation. Goat-anti-rabbit –Cy3 conjugated or
Goat-anti-mouse Cy3-conjugated antibodies were diluted 1:200 in 1% BSA/PBS
and incubated with coverslips for 2 h. Coverslips were again washed with 1x PBS
under gentle agitation, air-dried and mounted on pre-cleaned glass slides using
Prolong Gold anti-fade mounting media. Images were obtained using a Nikon
fluorescence microscope.

200
DARTS protease foot-printing assay
The DARTS assay was carried out as previously described (Lomenick et al.,
2009). In brief, BxPC3 cells were grown to near confluence in at T125 flask. Cell
monolayers were washed twice with 1x PBS and scraped into ice-cold DARTS
assay buffer (50 mM Tris-HCl pH 7.5, 200 mM NaCl, 0.5% Triton X-100, 10%
glycerol, 1 mM DTT) containing protease and phosphatase inhibitor cocktails. For
the DARTS assay, 50 µg of whole cell lysates were incubated with 500 µM
compound for 1 hour at RT. Pronase was added to the lysates at a dilution of
1:3000 and cells were incubated in a 37C water bath for 30 m. The action of the
pronase was terminated by the addition of SDS-PAGE buffer and subsequent
heating at 100 C for 5 m in a heat block. Samples were resolved by SDS-PAGE
and stained with Coomasie Brilliant Blue following by destaining in graded
solutions of methanol/acetic acid. Bands were visualized using a CCD camera.
Excised bands were washed twice in 50% acetonitrile, ddH2O and stored at -
80C until Mass Spectrometry could be performed.

DARTS-Western Blotting
For DARTS-Western blotting experiments whole cell lysates were prepared and
processed as previously described for DARTS assay. Following resolution by
SDS-PAGE, proteins were transferred to nitrocellulose and western blotting was
performed as described.

201
Oncomine bioinformatics analysis
Bioinformatics analysis was performed on the Oncomine bioinformatics platform
using publicly available datasets published in peer-reviewed journals.

Mouse Xenograft Studies
Tumor xenografts were established by subcutaneous injection of 0.5 x 10
6

HCT116 p53 +/+ cells into 6 week old female athymic nude mice. When tumors
reached a volume of 100 mm3, mice were randomized to vehicle control and
treatment groups. Mice received AV220 prepared in 100 µL of 5% DMSO/95%
sesame oil via intraperitoneal injection. Dosing was initiated at 10 mg/kg and was
eventually escalated to 20 mg/kg and 40 mg/kg when tumor volumes reached
500 mm
3
. Tumors were measured three times weekly using Vernier calipers.
Tumor volume was calculated using the following equation: V= d
2
x D/2 , where d
= the width or smaller measure and D= the length or larger measure. Data
collected was plotted and analyzed to determine average tumor volumes and
weights, SEM values, and p-values using Microsoft Excel. Health checks were
performed daily. Mice exhibiting toxicities or excessive tumor burden (> 2.0 cm
3
)
were sacrificed using CO
2
gas, necropsies were performed and tumor samples
and organs were harvested and fixed in 10% neutral buffered formalin prior to
processing for histological analysis. The animal studies were approved by the
USC Animal Care and Use Committee under protocol number 11458. Animal
care and manipulation were in agreement with the USC institutional guidelines,
202
which were in accordance with the Guidelines for the Care and Use of Laboratory
Animals.

Statistical analysis
Statistical analysis was performed for tumor volumes in Microsoft Excel using the
Student’s t-test, assuming unequal variances. P-values less than 0.05, obtained
by this method were considered to be significant.
203
References

Agapova, L.S., Chernyak, B.V., Domnina, L.V., Dugina, V.B., Efimenko, A.Y., Fetisova, E.K.,
Ivanova, O.Y., Kalinina, N.I., Khromova, N.V., Kopnin, B.P., et al. (2008). Mitochondria-targeted
plastoquinone derivatives as tools to interrupt execution of the aging program. 3. Inhibitory effect
of SkQ1 on tumor development from p53-deficient cells. Biochemistry (Mosc) 73, 1300-1316.
Agarwala, S.S., and Kirkwood, J.M. (2000). Temozolomide, a novel alkylating agent with activity
in the central nervous system, may improve the treatment of advanced metastatic melanoma.
Oncologist 5, 144-151.
Alphonse Taghian, M., PhD, Moataz N El-Ghamry, M., and Sofia D Merajver, M., PhD (2013).
Overview of the treatment of newly diagnosed, non-metastatic breast cancer. In UpToDate, M.
Daniel F Hayes, ed. (Waltham, MA, UpToDate).
Amrein, L., Loignon, M., Goulet, A.C., Dunn, M., Jean-Claude, B., Aloyz, R., and Panasci, L.
(2007). Chlorambucil cytotoxicity in malignant B lymphocytes is synergistically increased by 2-
(morpholin-4-yl)-benzo[h]chomen-4-one (NU7026)-mediated inhibition of DNA double-strand
break repair via inhibition of DNA-dependent protein kinase. J Pharmacol Exp Ther 321, 848-855.
Anastasiou, D., Poulogiannis, G., Asara, J.M., Boxer, M.B., Jiang, J.K., Shen, M., Bellinger, G.,
Sasaki, A.T., Locasale, J.W., Auld, D.S., et al. Inhibition of pyruvate kinase M2 by reactive
oxygen species contributes to cellular antioxidant responses. Science 334, 1278-1283.
Animati, F., Arcamone, F., Bigioni, M., Capranico, G., Caserini, C., De Cesare, M., Lombardi, P.,
Pratesi, G., Salvatore, C., Supino, R., et al. (1996). Biochemical and pharmacological activity of
novel 8-fluoroanthracyclines: influence of stereochemistry and conformation. Mol Pharmacol 50,
603-609.
Ashizawa, K., McPhie, P., Lin, K.H., and Cheng, S.Y. (1991a). An in vitro novel mechanism of
regulating the activity of pyruvate kinase M2 by thyroid hormone and fructose 1, 6-bisphosphate.
Biochemistry 30, 7105-7111.
Ashizawa, K., Willingham, M.C., Liang, C.M., and Cheng, S.Y. (1991b). In vivo regulation of
monomer-tetramer conversion of pyruvate kinase subtype M2 by glucose is mediated via fructose
1,6-bisphosphate. J Biol Chem 266, 16842-16846.
Ashley, N., and Poulton, J. (2009). Mitochondrial DNA is a direct target of anti-cancer
anthracycline drugs. Biochem Biophys Res Commun 378, 450-455.
Asin-Cayuela, J., Manas, A.R., James, A.M., Smith, R.A., and Murphy, M.P. (2004). Fine-tuning
the hydrophobicity of a mitochondria-targeted antioxidant. FEBS Lett 571, 9-16.
Babcock, D.F., Herrington, J., Goodwin, P.C., Park, Y.B., and Hille, B. (1997). Mitochondrial
participation in the intracellular Ca2+ network. J Cell Biol 136, 833-844.
204
Badea, L., Herlea, V., Dima, S.O., Dumitrascu, T., and Popescu, I. (2008). Combined gene
expression analysis of whole-tissue and microdissected pancreatic ductal adenocarcinoma
identifies genes specifically overexpressed in tumor epithelia. Hepatogastroenterology 55, 2016-
2027.
Belostotsky, I., da Silva, S.M., Paez, M.G., and Indig, G.L. (2011). Mitochondrial targeting for
photochemotherapy. Can selective tumor cell killing be predicted based on n-octanol/water
distribution coefficients? Biotech Histochem 86, 302-314.
Bergeron, K.L., Murphy, E.L., Majofodun, O., Munoz, L.D., Williams, J.C., and Almeida, K.H.
(2009). Arylphosphonium salts interact with DNA to modulate cytotoxicity. Mutation Research-
Genetic Toxicology and Environmental Mutagenesis 673, 141-148.
Beyer, U., Roth, T., Schumacher, P., Maier, G., Unold, A., Frahm, A.W., Fiebig, H.H., Unger, C.,
and Kratz, F. (1998). Synthesis and in vitro efficacy of transferrin conjugates of the anticancer
drug chlorambucil. J Med Chem 41, 2701-2708.
Bielawski, K., and Bielawska, A. (2008). Small-molecule based delivery systems for alkylating
antineoplastic compounds. ChemMedChem 3, 536-542.
Bielawski, K., Bielawska, A., Muszynska, A., Poplawska, B., and Czarnomysy, R. Cytotoxic
activity of G3 PAMAM-NH(2) dendrimer-chlorambucil conjugate in human breast cancer cells.
Environ Toxicol Pharmacol 32, 364-372.
Blank, B., DiTullio, N., Deviney, L., Roberts, J., and Saunders, H. (1975). Synthesis and
hypoglycemic activity of phenacyltriphenylphosphoranes and phosphonium salts. J Med Chem
18, 952-954.
Bos, J.L. (1989). ras oncogenes in human cancer: a review. Cancer Res 49, 4682-4689.
Boxer, M.B., Jiang, J.K., Vander Heiden, M.G., Shen, M., Skoumbourdis, A.P., Southall, N., Veith,
H., Leister, W., Austin, C.P., Park, H.W., et al. (2010). Evaluation of substituted N,N'-
diarylsulfonamides as activators of the tumor cell specific M2 isoform of pyruvate kinase. J Med
Chem 53, 1048-1055.
Carmichael, J., DeGraff, W., Gazdar, A., Minna, J., and Mitchell, J. (1987a). Evaluation of a
tetrazolium-based semiautomated colorimetric assay: assessment of chemosensitivity testing.
Cancer Res 47, 936-942.
Carmichael, J., DeGraff, W.G., Gazdar, A.F., Minna, J.D., and Mitchell, J.B. (1987b). Evaluation
of a Tetrazolium-based Semiautomated Colorimetric Assay: Assessment of Chemosensitivity
Testing. Cancer Research 47, 936-942.
Catovsky, D., Else, M., and Richards, S. Chlorambucil--still not bad: a reappraisal. Clin
Lymphoma Myeloma Leuk 11 Suppl 1, S2-6.
Chabner BA, B.J., Cleary J, Ortiz T, Lane A, Supko JG, Ryan D. (2011). Goodman & Gilman's
Pharmacologic Basis of Therapeutics. In Chapter 61 Cytotoxic Agents In:, C.B. Brunton LL,
Knollmann BC, , ed. (New York, McGraw-Hill).
205
Chan, D.C. (2006). Mitochondria: dynamic organelles in disease, aging, and development. Cell
125, 1241-1252.
CHANCE, B., and WILLIAMS, G. (1956). The respiratory chain and oxidative phosphorylation.
Adv Enzymol Relat Subj Biochem 17, 65-134.
Chen, L.B. (1988a). Mitochondrial Membrane Potential in Living Cells. Annual Review of Cell
Biology 4, 155-181.
Chen, L.B. (1988b). Mitochondrial membrane potential in living cells. Annu Rev Cell Biol 4, 155-
181.
Christofk, H.R., Vander Heiden, M.G., Wu, N., Asara, J.M., and Cantley, L.C. (2008). Pyruvate
kinase M2 is a phosphotyrosine-binding protein. Nature 452, 181-186.
Chun, S.Y., Johnson, C., Washburn, J.G., Cruz-Correa, M.R., Dang, D.T., and Dang, L.H. (2010).
Oncogenic KRAS modulates mitochondrial metabolism in human colon cancer cells by inducing
HIF-1alpha and HIF-2alpha target genes. Mol Cancer 9, 293.
Clavel, C.M., Zava, O., Schmitt, F., Kenzaoui, B.H., Nazarov, A.A., Juillerat-Jeanneret, L., and
Dyson, P.J. Thermoresponsive chlorambucil derivatives for tumour targeting. Angew Chem Int Ed
Engl 50, 7124-7127.
Conradt, L., Godl, K., Schaab, C., Tebbe, A., Eser, S., Diersch, S., Michalski, C.W., Kleeff, J.,
Schnieke, A., Schmid, R.M., et al. (2011). Disclosure of erlotinib as a multikinase inhibitor in
pancreatic ductal adenocarcinoma. Neoplasia 13, 1026-1034.
Cooper, W.A., Bartier, W.A., Rideout, D.C., and Delikatny, E.J. (2001). 1H NMR visible lipids are
induced by phosphonium salts and 5-fluorouracil in human breast cancer cells. Magn Reson Med
45, 1001-1010.
Cromlish, J., and Roeder, R. (1989). Human transcription factor IIIC (TFIIIC). Purification,
polypeptide structure, and the involvement of thiol groups in specific DNA binding. J Biol Chem
264, 18100-18109.
Cullinane, C., and Bohr, V.A. (1998). DNA Interstrand Cross-Links Induced by Psoralen Are Not
Repaired in Mammalian Mitochondria. Cancer Research 58, 1400-1404.
Curtis, C., Shah, S.P., Chin, S.F., Turashvili, G., Rueda, O.M., Dunning, M.J., Speed, D., Lynch,
A.G., Samarajiwa, S., Yuan, Y., et al. (2012). The genomic and transcriptomic architecture of
2,000 breast tumours reveals novel subgroups. Nature 486, 346-352.
D.Gutterman, D. (2005). Mitochondria and reactive oxygen species: an evolution in function. Circ
Res 97, 302-304.
D'Errico, M., de Rinaldis, E., Blasi, M.F., Viti, V., Falchetti, M., Calcagnile, A., Sera, F., Saieva,
C., Ottini, L., Palli, D., et al. (2009). Genome-wide expression profile of sporadic gastric cancers
with microsatellite instability. Eur J Cancer 45, 461-469.
206
De Vita, S., De Matteis, S., Laurenti, L., Chiusolo, P., Reddiconto, G., Fiorini, A., Leone, G., and
Sica, S. (2005). Secondary Ph+ acute lymphoblastic leukemia after temozolomide. Ann Hematol
84, 760-762.
DeBerardinis, R.J., Sayed, N., Ditsworth, D., and Thompson, C.B. (2008a). Brick by brick:
metabolism and tumor cell growth. Curr Opin Genet Dev 18, 54-61.
DeBerardinis, R.J., Sayed, N., Ditsworth, D., and Thompson, C.B. (2008b). Brick by brick:
metabolism and tumor cell growth. Current Opinion in Genetics and Development 18, 54-61.
Deby, C., and Goutier, R. (1990). New perspectives on the biochemistry of superoxide anion and
the efficiency of superoxide dismutases. Biochem Pharmacol 39, 399-405.
Delsite, R.L., Rasmussen, L.J., Rasmussen, A.K., Kalen, A., Goswami, P.C., and Singh, K.K.
(2003). Mitochondrial impairment is accompanied by impaired oxidative DNA repair in the
nucleus. Mutagenesis 18, 497-503.
Descoteaux, C., Brasseur, K., Leblanc, V., Parent, S., Asselin, E., and Berube, G. Design of
novel tyrosine-nitrogen mustard hybrid molecules active against uterine, ovarian and breast
cancer cell lines. Steroids 77, 403-412.
Descoteaux, C., Leblanc, V., Brasseur, K., Gupta, A., Asselin, E., and Berube, G. Synthesis of D-
and L-tyrosine-chlorambucil analogs active against breast cancer cell lines. Bioorg Med Chem
Lett 20, 7388-7392.
Desgrosellier, J.S., Barnes, L.A., Shields, D.J., Huang, M., Lau, S.K., Prevost, N., Tarin, D.,
Shattil, S.J., and Cheresh, D.A. (2009). An integrin alpha(v)beta(3)-c-Src oncogenic unit
promotes anchorage-independence and tumor progression. Nat Med 15, 1163-1169.
Dubois, R., Lin, C., and Beisler, J. (1978a). Synthesis and antitumor properties of some
isoindolylalkylphosphonium salts. J Med Chem 21, 303-306.
Dubois, R.J., Lin, C.C., and Beisler, J.A. (1978b). Synthesis and antitumor properties of some
isoindolylalkylphosphonium salts. J Med Chem 21, 303-306.
Echave, P., Machado-da-Silva, G., Arkell, R.S., Duchen, M.R., Jacobson, J., Mitter, R., and
Lloyd, A.C. (2009). Extracellular growth factors and mitogens cooperate to drive mitochondrial
biogenesis. J Cell Sci 122, 4516-4525.
Evers, B., Schut, E., van der Burg, E., Braumuller, T.M., Egan, D.A., Holstege, H., Edser, P.,
Adams, D.J., Wade-Martins, R., Bouwman, P., et al. (2010). A High-Throughput Pharmaceutical
Screen Identifies Compounds with Specific Toxicity against BRCA2-Deficient Tumors. Clinical
Cancer Research 16, 99-108.
Faubert, B., Boily, G., Izreig, S., Griss, T., Samborska, B., Dong, Z., Dupuy, F., Chambers, C.,
Fuerth, B.J., Viollet, B., et al. (2013). AMPK is a negative regulator of the Warburg effect and
suppresses tumor growth in vivo. Cell Metab 17, 113-124.
207
Filomeni, G., Rotilio, G., and Ciriolo, M. (2005). Disulfide relays and phosphorylative cascades:
partners in redox-mediated signaling pathways. Cell Death Differ 12, 1555-1563.
Fonseca, S.B., Pereira, M.P., Mourtada, R., Gronda, M., Horton, K.L., Hurren, R., Minden, M.D.,
Schimmer, A.D., and Kelley, S.O. Rerouting chlorambucil to mitochondria combats drug
deactivation and resistance in cancer cells. Chem Biol 18, 445-453.
Fridovich, I. (1978). The biology of oxygen radicals. Science 201, 875-880.
Gane, E.J., Weilert, F., Orr, D.W., Keogh, G.F., Gibson, M., Lockhart, M.M., Frampton, C.M.,
Taylor, K.M., Smith, R.A., and Murphy, M.P. (2010). The mitochondria-targeted anti-oxidant
mitoquinone decreases liver damage in a phase II study of hepatitis C patients. Liver Int 30,
1019-1026.
Gao, X., Wang, H., Yang, J.J., Liu, X., and Liu, Z.R. (2012). Pyruvate kinase M2 regulates gene
transcription by acting as a protein kinase. Mol Cell 45, 598-609.
Geiger, H., Schleimer, D., Nattamai, K.J., Dannenmann, S.R., Davies, S.M., and Weiss, B.D.
(2006). Mutagenic potential of temozolomide in bone marrow cells in vivo. Blood 107, 3010-3011.
Gewirtz, D.A. (1999). A critical evaluation of the mechanisms of action proposed for the antitumor
effects of the anthracycline antibiotics adriamycin and daunorubicin. Biochem Pharmacol 57, 727-
741.
Goff, R.D., and Thorson, J.S. Assessment of chemoselective neoglycosylation methods using
chlorambucil as a model. J Med Chem 53, 8129-8139.
Grinius, L., Jasaitis, A., Kadziauskas, Y., Liberman, E., Skulachev, V., Topali, V., Tsofina, L., and
Vladimirova, M. (1970). Conversion of biomembrane-produced energy into electric form. I.
Submitochondrial particles. Biochim Biophys Acta 216, 1-12.
Grzesiak, J.J., Ho, J.C., Moossa, A.R., and Bouvet, M. (2007). The integrin-extracellular matrix
axis in pancreatic cancer. Pancreas 35, 293-301.
Guano, F., Pourquier, P., Tinelli, S., Binaschi, M., Bigioni, M., Animati, F., Manzini, S., Zunino, F.,
Kohlhagen, G., Pommier, Y., et al. (1999). Topoisomerase poisoning activity of novel
disaccharide anthracyclines. Mol Pharmacol 56, 77-84.
Guaragna, A., Chiaviello, A., Paolella, C., D'Alonzo, D., and Palumbo, G. Synthesis and
evaluation of folate-based chlorambucil delivery systems for tumor-targeted chemotherapy.
Bioconjug Chem 23, 84-96.
Gupta, A., Saha, P., Descoteaux, C., Leblanc, V., Asselin, E., and Berube, G. Design, synthesis
and biological evaluation of estradiol-chlorambucil hybrids as anticancer agents. Bioorg Med
Chem Lett 20, 1614-1618.
Gysin, S., Salt, M., Young, A., and McCormick, F. (2011). Therapeutic strategies for targeting ras
proteins. Genes Cancer 2, 359-372.
208
Haga, N., Saito, S., Tsukumo, Y., Sakurai, J., Furuno, A., Tsuruo, T., and Tomida, A. (2010).
Mitochondria regulate the unfolded protein response leading to cancer cell survival under glucose
deprivation conditions. Cancer Sci.
Hansson, J., Lewensohn, R., Ringborg, U., and Nilsson, B. (1987). Formation and Removal of
DNA Cross-Links Induced by Melphalan and Nitrogen Mustard in Relation to Drug-induced
Cytotoxicity in Human Melanoma Cells. Cancer Research 47, 2631-2637.
Harris, A.L. (2002). Hypoxia--a key regulatory factor in tumour growth. Nat Rev Cancer 2, 38-47.
Hashiguchi, K., and Zhang-Akiyama, Q.-M. (2009). Establishment of Human Cell Lines Lacking
Mitochondrial DNA
Mitochondrial DNA. In, J.A. Stuart, ed. (Humana Press), pp. 383-391.
Hatzivassiliou, G., Andreadis, C., and Thompson, C.B. (2005). Akt-directed metabolic alterations
in cancer. Drug Discov Today Dis Mech 2, 255-262.
National Institutes of Health, (2013). www.clinicaltrials.gov.
Heerdt, B.G., Houston, M.A., and Augenlicht, L.H. (2005a). The Intrinsic Mitochondrial Membrane
Potential of Colonic Carcinoma Cells Is Linked to the Probability of Tumor Progression. Cancer
Research 65, 9861-9867.
Heerdt, B.G., Houston, M.A., and Augenlicht, L.H. (2005b). The intrinsic mitochondrial membrane
potential of colonic carcinoma cells is linked to the probability of tumor progression. Cancer Res
65, 9861-9867.
Heerdt, B.G., Houston, M.A., and Augenlicht, L.H. (2006a). Growth Properties of Colonic Tumor
Cells Are a Function of the Intrinsic Mitochondrial Membrane Potential. Cancer Research 66,
1591-1596.
Heerdt, B.G., Houston, M.A., and Augenlicht, L.H. (2006b). Growth properties of colonic tumor
cells are a function of the intrinsic mitochondrial membrane potential. Cancer Res 66, 1591-1596.
Heerdt, B.G., Houston, M.A., Wilson, A.J., and Augenlicht, L.H. (2003). The Intrinsic
Mitochondrial Membrane Potential (Œîœàm) Is Associated with Steady-State Mitochondrial
Activity and the Extent to Which Colonic Epithelial Cells Undergo Butyrate-mediated Growth
Arrest and Apoptosis. Cancer Research 63, 6311-6319.
Hidalgo, M. (2010). Pancreatic cancer. N Engl J Med 362, 1605-1617.
Hlavata, L., and Nystrom, T. (2003). Ras proteins control mitochondrial biogenesis and function in
Saccharomyces cerevisiae. Folia Microbiol (Praha) 48, 725-730.
Holmgren, A. (1985). Thioredoxin. Annu Rev Biochem 54, 237-271.
Howlader N, N.A., Krapcho M, Neyman N, Aminou R, Altekruse SF, Kosary CL, Ruhl J,
Tatalovich Z, Cho H, Mariotto A, Eisner MP, Lewis DR, Chen HS, Feuer EJ, Cronin KA (eds).
209
(2013). SEER Cancer Statistics Review, 1975-2009 (Vintage 2009 Populations) (Bethesda,
National Cancer Institute).
Hruban, R.H., Goggins, M., Parsons, J., and Kern, S.E. (2000). Progression model for pancreatic
cancer. Clin Cancer Res 6, 2969-2972.
Hu, Y., Lu, W., Chen, G., Wang, P., Chen, Z., Zhou, Y., Ogasawara, M., Trachootham, D., Feng,
L., Pelicano, H., et al. (2012). K-ras(G12V) transformation leads to mitochondrial dysfunction and
a metabolic switch from oxidative phosphorylation to glycolysis. Cell Res 22, 399-412.
Isola, J., Helin, H., Helle, M., and Kallioniemi, O. (1990). Evaluation of cell proliferation in breast
carcinoma. Comparison of Ki-67 immunohistochemical study, DNA flow cytometric analysis, and
mitotic count. Cancer 65, 1180-1184.
Ji, H., Zhao, X., Yuza, Y., Shimamura, T., Li, D., Protopopov, A., Jung, B.L., McNamara, K., Xia,
H., Glatt, K.A., et al. (2006). Epidermal growth factor receptor variant III mutations in lung
tumorigenesis and sensitivity to tyrosine kinase inhibitors. Proc Natl Acad Sci U S A 103, 7817-
7822.
Jones, S., Zhang, X., Parsons, D.W., Lin, J.C., Leary, R.J., Angenendt, P., Mankoo, P., Carter,
H., Kamiyama, H., Jimeno, A., et al. (2008). Core signaling pathways in human pancreatic
cancers revealed by global genomic analyses. Science 321, 1801-1806.
Jung-whan Kim, and Dang, C.V. (2005). Multifaceted roles of glycolytic enzymes. TRENDS in
Biochemical Sciences 30, 142-150.
Kemp, B., Stapleton, D., Campbell, D., Chen, Z., Murthy, S., Walter, M., Gupta, A., Adams, J.,
Katsis, F., van Denderen, B., et al. (2003). AMP-activated protein kinase, super metabolic
regulator. Biochem Soc Trans 31, 162-168.
Kim, J.W., Gardner, L.B., and Dang, C.V. (2005). Oncogenic alterations of metabolism and the
Warburg effect. Drug Discov Today Dis Mech 2, 233-238.
Kim, Y., Yang, C., Wang, J., Wang, L., Li, Z., Chen, X., and Liu, S. (2008a). Effects of targeting
moiety, linker, bifunctional chelator, and molecular charge on biological properties of 64Cu-
labeled triphenylphosphonium cations. J Med Chem 51, 2971-2984.
Kim, Y.S., Yang, C.T., Wang, J., Wang, L., Li, Z.B., Chen, X., and Liu, S. (2008b). Effects of
targeting moiety, linker, bifunctional chelator, and molecular charge on biological properties of
64Cu-labeled triphenylphosphonium cations. J Med Chem 51, 2971-2984.
Kinnamon, K., Steck, E., Hanson, W., and Chapman, W.J. (1977). In search of anti-Trypanosoma
cruzi drugs: new leads from a mouse model. J Med Chem 20, 741-744.
Knauf, W.U., Lissitchkov, T., Aldaoud, A., Liberati, A.M., Loscertales, J., Herbrecht, R., Juliusson,
G., Postner, G., Gercheva, L., Goranov, S., et al. Bendamustine compared with chlorambucil in
previously untreated patients with chronic lymphocytic leukaemia: updated results of a
randomized phase III trial. Br J Haematol 159, 67-77.
210
Koczor, C.A., Shokolenko, I.N., Boyd, A.K., Balk, S.P., Wilson, G.L., and LeDoux, S.P. (2009).
Mitochondrial DNA Damage Initiates a Cell Cycle Arrest by a Chk2-associated Mechanism in
Mammalian Cells. Journal of Biological Chemistry 284, 36191-36201.
Kondo, N., Takahashi, A., Ono, K., and Ohnishi, T. (2010). DNA Damage Induced by Alkylating
Agents and Repair Pathways. Journal of Nucleic Acids 2010.
Kung, C., Hixon, J., Choe, S., Marks, K., Gross, S., Murphy, E., DeLaBarre, B., Cianchetta, G.,
Sethumadhavan, S., Wang, X., et al. Small molecule activation of PKM2 in cancer cells induces
serine auxotrophy. Chem Biol 19, 1187-1198.
KUO, L.J., and YANG, L.-X. (2008). Œ≥-H2AX - A Novel Biomarker for DNA Double-strand
Breaks. In Vivo 22, 305-309.
Kurt Højlund, Martin Mogensen, Kent Sahlin , and Beck-Nielsen, H. (2008). Mitochondrial
dysfunction in type 2 diabetes and obesity. Endocrinol Metab Clin North Am 37, 713-731.
LeDoux, S.P., Wilson, G.L., Beecham, E.J., Stevnsner, T., Wassermann, K., and Bohr, V.A.
(1992). Repair of mitochondrial DNA after various types of DNA damage in Chinese hamster
ovary cells. Carcinogenesis 13, 1967-1973.
Lee, F.Y., Coe, P., and Workman, P. (1986). Pharmacokinetic basis for the comparative
antitumour activity and toxicity of chlorambucil, phenylacetic acid mustard and beta, beta-
difluorochlorambucil (CB 7103) in mice. Cancer Chemother Pharmacol 17, 21-29.
Li, J., and Yuan, J. (2008). Caspases in apoptosis and beyond. Oncogene 27, 6194-6206.
Li, Y.-H., Wang, X., Pan, Y., Lee, D.-H., Chowdhury, D., and Kimmelman, A.C. (2012). Inhibition
of Non-Homologous End Joining Repair Impairs Pancreatic Cancer Growth and Enhances
Radiation Response. PLoS ONE 7, e39588.
Li, Z., Lopez, M., Hardy, M., McAllister, D.M., Kalyanaraman, B., and Zhao, M. (2009). A
(99m)Tc-labeled triphenylphosphonium derivative for the early detection of breast tumors. Cancer
Biother Radiopharm 24, 579-587.
Lock, R., and Debnath, J. (2008). Extracellular matrix regulation of autophagy. Curr Opin Cell Biol
20, 583-588.
Logsdon, C.D., Simeone, D.M., Binkley, C., Arumugam, T., Greenson, J.K., Giordano, T.J.,
Misek, D.E., Kuick, R., and Hanash, S. (2003). Molecular profiling of pancreatic adenocarcinoma
and chronic pancreatitis identifies multiple genes differentially regulated in pancreatic cancer.
Cancer Res 63, 2649-2657.
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.
Lu, Z. PKM2 functions as a histone kinase. Cell Cycle 11, 4101-4102.
211
Luo, W., and Semenza, G.L. Emerging roles of PKM2 in cell metabolism and cancer progression.
Trends Endocrinol Metab 23, 560-566.
Madar, I., Weiss, L., and Izbicki, G. (2002). Preferential accumulation of (3)H-
tetraphenylphosphonium in non-small cell lung carcinoma in mice: comparison with (99m)Tc-
MIBI. J Nucl Med 43, 234-238.
Mahadevan, D., and Von Hoff, D.D. (2007). Tumor-stroma interactions in pancreatic ductal
adenocarcinoma. Mol Cancer Ther 6, 1186-1197.
Manetta, A., Gamboa, G., Nasseri, A., Podnos, Y., Emma, D., Dorion, G., Rawlings, L.,
Carpenter, P., Bustamante, A., Patel, J., et al. (1996a). Novel phosphonium salts display in vitro
and in vivo cytotoxic activity against human ovarian cancer cell lines. Gynecol Oncol 60, 203-212.
Manetta, A., Gamboa, G., Nasseri, A., Podnos, Y.D., Emma, D., Dorion, G., Rawlings, L.,
Carpenter, P.M., Bustamante, A., Patel, J., et al. (1996b). Novel phosphonium salts display in
vitro and in vivo cytotoxic activity against human ovarian cancer cell lines. Gynecol Oncol 60,
203-212.
Mao, Z., Jiang, Y., Liu, X., Seluanov, A., and Gorbunova, V. (2009). DNA repair by homologous
recombination, but not by nonhomologous end joining, is elevated in breast cancer cells.
Neoplasia 11, 683-691.
Marroquin, L.D., Hynes, J., Dykens, J.A., Jamieson, J.D., and Will, Y. (2007). Circumventing the
Crabtree effect: replacing media glucose with galactose increases susceptibility of HepG2 cells to
mitochondrial toxicants. Toxicol Sci 97, 539-547.
Mart√≠nez-Diez, M., Santamar√≠a, G., Ortega, Å.D., and Cuezva, J.M. (2006). Biogenesis and
Dynamics of Mitochondria during the Cell Cycle: Significance of 3‚Ä≤UTRs. PLoS ONE 1, e107.
Mathews, L.A., Cabarcas, S.M., Hurt, E.M., Zhang, X., Jaffee, E.M., and Farrar, W.L. (2011).
Increased expression of DNA repair genes in invasive human pancreatic cancer cells. Pancreas
40, 730-739.
McLean, A., Newell, D., Baker, G., and Connors, T. (1980). The metabolism of chlorambucil.
Biochemical Pharmacology 29, 2039-2047.
Miguel A Rodriguez-Bigas, M., Axel Grothey, M., Editors, S., Kenneth K Tanabe, M., and Richard
M Goldberg, M. (2013). Overview of the management of primary colon cancer. In UpToDate, M.
Kenneth K Tanabe, and M. Richard M Goldberg, eds. (Waltham, MA, UpToDate).
Millard, M., Pathania, D., Shabaik, Y., Taheri, L., Deng, J., and Neamati, N. (2010a). Preclinical
evaluation of novel triphenylphosphonium salts with broad-spectrum activity. PLoS One 5.
Millard, M., Pathania, D., Shabaik, Y., Taheri, L., Deng, J., and Neamati, N. (2010b). Preclinical
Evaluation of Novel Triphenylphosphonium Salts with Broad-Spectrum Activity. PLoS ONE 5,
e13131.
212
Min, J.J., Biswal, S., Deroose, C., and Gambhir, S.S. (2004). Tetraphenylphosphonium as a novel
molecular probe for imaging tumors. J Nucl Med 45, 636-643.
Minotti, G., Menna, P., Salvatorelli, E., Cairo, G., and Gianni, L. (2004). Anthracyclines: molecular
advances and pharmacologic developments in antitumor activity and cardiotoxicity. Pharmacol
Rev 56, 185-229.
Mirkes, P.E., and Greenaway, J.C. (1982). Teratogenicity of chlorambucil in rat embryos in vitro.
Teratology 26, 135-143.
Modica-Napolitano, J.S., and Aprille, J.R. (2001). Delocalized lipophilic cations selectively target
the mitochondria of carcinoma cells. Adv Drug Deliv Rev 49, 63-70.
Moreno-Sánchez, R., Rodríguez-Enríquez, S., Marín-Hernández, A., and Saavedra, E. (2007).
Energy metabolism in tumor cells. The FEBS Journal 274, 1393-1418.
Mossner, J. (2010). What's new in therapy of pancreatic cancer? Dig Dis 28, 679-683.
Munshi, A., Hobbs, M., and Meyn, R. (2005). Clonogenic cell survival assay. Methods Mol Med
Murphy, M.P. (2008). Targeting lipophilic cations to mitochondria. Biochimica et Biophysica Acta
(BBA) - Bioenergetics 1777, 1028-1031.
Myrberg, H., Zhang, L., Mae, M., and Langel, U. (2008). Design of a tumor-homing cell-
penetrating peptide. Bioconjug Chem 19, 70-75.
Nagy, A., Schally, A.V., Armatis, P., Szepeshazi, K., Halmos, G., Kovacs, M., Zarandi, M., Groot,
K., Miyazaki, M., Jungwirth, A., et al. (1996). Cytotoxic analogs of luteinizing hormone-releasing
hormone containing doxorubicin or 2-pyrrolinodoxorubicin, a derivative 500-1000 times more
potent. Proc Natl Acad Sci U S A 93, 7269-7273.
Niraula, S., Seruga, B., Ocana, A., Shao, T., Goldstein, R., Tannock, I.F., and Amir, E. (2012).
The price we pay for progress: a meta-analysis of harms of newly approved anticancer drugs. J
Clin Oncol 30, 3012-3019.
Owusu-Ansah, E., Yavari, A., Mandal, S., and Banerjee, U. (2008). Distinct mitochondrial
retrograde signals control the G1-S cell cycle checkpoint. Nat Genet 40, 356-361.
Panasci, L., Paiement, J.-P., Christodoulopoulos, G., Belenkov, A., Malapetsa, A., and Aloyz, R.
(2001). Chlorambucil Drug Resistance in Chronic Lymphocytic Leukemia. Clinical Cancer
Research 7, 454-461.
Pathania, D., Millard, M., and Neamati, N. (2009a). Opportunities in discovery and delivery of
anticancer drugs targeting mitochondria and cancer cell metabolism. Adv Drug Deliv Rev 61,
1250-1275.
213
Pathania, D., Millard, M., and Neamati, N. (2009b). Opportunities in discovery and delivery of
anticancer drugs targeting mitochondria and cancer cell metabolism. Adv Drug Deliv Rev 61,
1250-1275.
Pedersen, P.J., Christensen, M.S., Ruysschaert, T., Linderoth, L., Andresen, T.L., Melander, F.,
Mouritsen, O.G., Madsen, R., and Clausen, M.H. (2009). Synthesis and biophysical
characterization of chlorambucil anticancer ether lipid prodrugs. J Med Chem 52, 3408-3415.
Porteous, C.M., Logan, A., Evans, C., Ledgerwood, E.C., Menon, D.K., Aigbirhio, F., Smith, R.A.,
and Murphy, M.P. (2010). Rapid uptake of lipophilic triphenylphosphonium cations by
mitochondria in vivo following intravenous injection: implications for mitochondria-specific
therapies and probes. Biochim Biophys Acta 1800, 1009-1017.
Pratesi, G., De Cesare, M., Caserini, C., Perego, P., Dal Bo, L., Polizzi, D., Supino, R., Bigioni,
M., Manzini, S., Iafrate, E., et al. (1998). Improved efficacy and enlarged spectrum of activity of a
novel anthracycline disaccharide analogue of doxorubicin against human tumor xenografts. Clin
Cancer Res 4, 2833-2839.
Propper, D.J., Braybrooke, J.P., Taylor, D.J., Lodi, R., Styles, P., Cramer, J.A., Collins, W.C.,
Levitt, N.C., Talbot, D.C., Ganesan, T.S., et al. (1999). Phase I trial of the selective mitochondrial
toxin MKT077 in chemo-resistant solid tumours. Ann Oncol 10, 923-927.
RA Gatenby, ET Gawlinski, AF Gmitro, B Kaylor, and Gillies, R. (2006). Acid-mediated tumor
invasion: a multidisciplinary study. Cancer Cell Research 66, 5216-5223.
Ralph J DeBerardinis, Julian J Lum, Georgia Hatzivassiliou, and Thompson, C. (2008). The
Biology of Cancer: Metabolic Reprogramming Fuels Cell Growth and Proliferation. Cell
Metabolism 7, 11-20.
Rebollo, A., Perez-Sala, D., and Martinez, A.C. (1999). Bcl-2 differentially targets K-, N-, and H-
Ras to mitochondria in IL-2 supplemented or deprived cells: implications in prevention of
apoptosis. Oncogene 18, 4930-4939.
Rehman, J., Zhang, H.J., Toth, P.T., Zhang, Y., Marsboom, G., Hong, Z., Salgia, R., Husain,
A.N., Wietholt, C., and Archer, S.L. (2012). Inhibition of mitochondrial fission prevents cell cycle
progression in lung cancer. The FASEB Journal 26, 2175-2186.
Rideout, D., Bustamante, A., and Patel, J. (1994). Mechanism of inhibition of FaDu
hypopharyngeal carcinoma cell growth by tetraphenylphosphonium chloride. Int J Cancer 57,
247-253.
Rideout, D., Calogeropoulou, T., Jaworski, J., Dagnino, R.J., and McCarthy, M. (1989).
Phosphonium salts exhibiting selective anti-carcinoma activity in vitro. Anticancer Drug Des 4,
265-280.
Ross, M., Kelso, G., Blaikie, F., James, A., CochemÈ, H., Filipovska, A., Da Ros, T., Hurd, T.,
Smith, R., and Murphy, M. (2005a). Lipophilic triphenylphosphonium cations as tools in
mitochondrial bioenergetics and free radical biology. Biochemistry (Mosc) 70, 222-230.
214
Ross, M.F., Da Ros, T., Blaikie, F.H., Prime, T.A., Porteous, C.M., Severina, II, Skulachev, V.P.,
Kjaergaard, H.G., Smith, R.A., and Murphy, M.P. (2006). Accumulation of lipophilic dications by
mitochondria and cells. Biochem J 400, 199-208.
Ross, M.F., Kelso, G.F., Blaikie, F.H., James, A.M., Cocheme, H.M., Filipovska, A., Da Ros, T.,
Hurd, T.R., Smith, R.A., and Murphy, M.P. (2005b). Lipophilic triphenylphosphonium cations as
tools in mitochondrial bioenergetics and free radical biology. Biochemistry (Mosc) 70, 222-230.
Ross, M.F., Prime, T.A., Abakumova, I., James, A.M., Porteous, C.M., Smith, R.A., and Murphy,
M.P. (2008). Rapid and extensive uptake and activation of hydrophobic triphenylphosphonium
cations within cells. Biochem J 411, 633-645.
Sarpietro, M.G., Giuffrida, M.C., Ottimo, S., Micieli, D., and Castelli, F. (2011). Evaluation of the
interaction of coumarins with biomembrane models studied by differential scanning calorimetry
and Langmuir-Blodgett techniques. J Nat Prod 74, 790-795.
Sawyer, D.B., Peng, X., Chen, B., Pentassuglia, L., and Lim, C.C. (2010). Mechanisms of
anthracycline cardiac injury: can we identify strategies for cardioprotection? Prog Cardiovasc Dis
53, 105-113.
Shabaik, Y., Millard, M., and Neamati, N. (2013). Mechanistic evaluation of a novel small
molecule targeting mitochondria in pancreatic cancer cells. PLoS ONE.
Shaheen, M., Allen, C., Nickoloff, J.A., and Hromas, R. (2011). Synthetic lethality: exploiting the
addiction of cancer to DNA repair. Blood 117, 6074-6082.
Smith, R.A., Porteous, C.M., Gane, A.M., and Murphy, M.P. (2003). Delivery of bioactive
molecules to mitochondria in vivo. Proc Natl Acad Sci U S A 100, 5407-5412.
Smith, R.A.J., and Murphy, M.P. (2010). Animal and human studies with the mitochondria-
targeted antioxidant MitoQ. Annals of the New York Academy of Sciences 1201, 96-103.
Smolková, K., Bellance, N., Scandurra, F., Génot, E., Gnaiger, E., Plecitá-Hlavatá, L., Jezek, P.,
and Rossignol, R. (2010). Mitochondrial bioenergetic adaptations of breast cancer cells to
aglycemia and hypoxia. J Bioenerg Biomembr 42, 55-67.
Snow, B.J., Rolfe, F.L., Lockhart, M.M., Frampton, C.M., O'Sullivan, J.D., Fung, V., Smith, R.A.,
Murphy, M.P., and Taylor, K.M. (2010). A double-blind, placebo-controlled study to assess the
mitochondria-targeted antioxidant MitoQ as a disease-modifying therapy in Parkinson's disease.
Mov Disord 25, 1670-1674.
Spiro, T.P., Liu, L., Majka, S., Haaga, J., Willson, J.K., and Gerson, S.L. (2001). Temozolomide:
the effect of once- and twice-a-day dosing on tumor tissue levels of the DNA repair protein O(6)-
alkylguanine-DNA-alkyltransferase. Clin Cancer Res 7, 2309-2317.
Stark, P.A., Thrall, B.D., Meadows, G.G., and Abdel-Monem, M.M. (1992). Synthesis and
evaluation of novel spermidine derivatives as targeted cancer chemotherapeutic agents. J Med
Chem 35, 4264-4269.
215
Steinbeck, M., Khan, A., and Karnovsky, M. (1993). Extracellular production of singlet oxygen by
stimulated macrophages quantified using 9,10-diphenylanthracene and perylene in a polystyrene
film. J Biol Chem 268, 15649-15654.
Stevens, M.F.G. (2008). Cancer Drug Design and Discovery, 1st Edition edn (New York, NY,
Elsevier).
Su, Y.W., Chang, M.C., Chiang, M.F., and Hsieh, R.K. (2005). Treatment-related myelodysplastic
syndrome after temozolomide for recurrent high-grade glioma. J Neurooncol 71, 315-318.
Takagi, Y., Kobayashi, N., Chang, M.S., Lim, G.J., and Tsuchiya, T. (1998). Synthesis and
antitumor activity of 7-O-(2,6-dideoxy-2-fluoro-alpha-L-talopyranosyl) daunomycinone derivatives
modified at C-3' or C-4'. Carbohydr Res 307, 217-232.
Tamada, M., Suematsu, M., and Saya, H. Pyruvate kinase M2: multiple faces for conferring
benefits on cancer cells. Clin Cancer Res 18, 5554-5561.
Tann, A.W., Boldogh, I., Meiss, G., Qian, W., Van Houten, B., Mitra, S., and Szczesny, B. (2011).
Apoptosis induced by persistent single-strand breaks in the mitochondrial genome: Critical role of
EXOG (5‚Ä≤ EXO/Endonuclease) in their repair. Journal of Biological Chemistry.
Tim Brenner, P., BCOP, Shrina Duggal, P., BCOP, Jim Natale, P., BCOP, and Scott M Wirth, P.,
BCOP (2013a). Treatment protocols for pancreatic cancer. In UpToDate, P. John W Mucenski,
ed. (Waltham, MA, UpToDate).
Tim Brenner, P., BCOP, Shrina Duggal, P., BCOP, Jim Natale, P., BCOP, Scott M Wirth, P.,
BCOP, Editor, S., John W Mucenski, P., Editor, D., and Don S Dizon, M., FACP (2013b).
Treatment protocols for breast cancer. In UpToDate, P. John W Mucenski, ed. (Waltham, MA,
UpToDate).
Tong, X., Zhao, F., and Thompson, C.B. (2009). The molecular determinants of de novo
nucleotide biosynthesis in cancer cells. Curr Opin Genet Dev 19, 32-37.
Trinei, M., Berniakovich, I., Pelicci, P.G., and Giorgio, M. (2006). Mitochondrial DNA copy number
is regulated by cellular proliferation: a role for Ras and p66(Shc). Biochim Biophys Acta 1757,
624-630.
United Nations Department of Economics and Social Affairs, Population Estimates and
Projections Sections (2010). World Population Prospects, The 2010 Revision, online database,
population.
van der Heijden, M.S., Brody, J.R., Dezentje, D.A., Gallmeier, E., Cunningham, S.C., Swartz,
M.J., DeMarzo, A.M., Offerhaus, G.J., Isacoff, W.H., Hruban, R.H., et al. (2005). In vivo
therapeutic responses contingent on Fanconi anemia/BRCA2 status of the tumor. Clin Cancer
Res 11, 7508-7515.
Veronese, S., Gambacorta, M., Gottardi, O., Scanzi, F., Ferrari, M., and Lampertico, P. (1993).
Proliferation index as a prognostic marker in breast cancer. Cancer 71, 3926-3931.
216
Wallace, D.C. (2005). A mitochondrial paradigm of metabolic and degenerative diseases, aging,
and cancer: a dawn for evolutionary medicine. Annu Rev Genet 39, 359-407.
Walter, P., and Ron, D. (2011). The unfolded protein response: from stress pathway to
homeostatic regulation. Science 334, 1081-1086.
Wang, J., Yang, C., Kim, Y., Sreerama, S., Cao, Q., Li, Z., He, Z., Chen, X., and Liu, S. (2007a).
64Cu-Labeled triphenylphosphonium and triphenylarsonium cations as highly tumor-selective
imaging agents. J Med Chem 50, 5057-5069.
Wang, J., Yang, C.T., Kim, Y.S., Sreerama, S.G., Cao, Q., Li, Z.B., He, Z., Chen, X., and Liu, S.
(2007b). 64Cu-Labeled triphenylphosphonium and triphenylarsonium cations as highly tumor-
selective imaging agents. J Med Chem 50, 5057-5069.
Wang, X. (2001). The expanding role of mitochondria in apoptosis. Genes Dev 15, 2922-2933.
Warburg, O. (1956a). On the origin of cancer. Science 123, 310-314.
Warburg, O. (1956b). On the origin of cancer cells. Science 123, 309-314.
Wasowska, M., Oszczapowicz, I., Wietrzyk, J., Opolski, A., Madej, J., Dzimira, S., and
Oszczapowicz, J. (2005). Influence of the structure of new anthracycline antibiotics on their
biological properties. Anticancer Res 25, 2043-2048.
Weinberg, F., Hamanaka, R., Wheaton, W.W., Weinberg, S., Joseph, J., Lopez, M.,
Kalyanaraman, B., Mutlu, G.M., Budinger, G.R., and Chandel, N.S. (2010). Mitochondrial
metabolism and ROS generation are essential for Kras-mediated tumorigenicity. Proc Natl Acad
Sci U S A 107, 8788-8793.
Wittig, G. (1980). From Diyls to Ylides to My Idyll. Science 210, 600-604.
Yang, C.T., Kim, Y.S., Wang, J., Wang, L., Shi, J., Li, Z.B., Chen, X., Fan, M., Li, J.J., and Liu, S.
(2008). 64Cu-labeled 2-(diphenylphosphoryl)ethyldiphenylphosphonium cations as highly
selective tumor imaging agents: effects of linkers and chelates on radiotracer biodistribution
characteristics. Bioconjug Chem 19, 2008-2022.
Yang, W., Xia, Y., Hawke, D., Li, X., Liang, J., Xing, D., Aldape, K., Hunter, T., Alfred Yung, W.K.,
and Lu, Z. (2012a). PKM2 phosphorylates histone H3 and promotes gene transcription and
tumorigenesis. Cell 150, 685-696.
Yang, W., Xia, Y., Ji, H., Zheng, Y., Liang, J., Huang, W., Gao, X., Aldape, K., and Lu, Z. Nuclear
PKM2 regulates beta-catenin transactivation upon EGFR activation. Nature 480, 118-122.
Yang, W., Zheng, Y., Xia, Y., Ji, H., Chen, X., Guo, F., Lyssiotis, C.A., Aldape, K., Cantley, L.C.,
and Lu, Z. ERK1/2-dependent phosphorylation and nuclear translocation of PKM2 promotes the
Warburg effect. Nat Cell Biol 14, 1295-1304.
217
Yang, W., Zheng, Y., Xia, Y., Ji, H., Chen, X., Guo, F., Lyssiotis, C.A., Aldape, K., Cantley, L.C.,
and Lu, Z. (2012b). ERK1/2-dependent phosphorylation and nuclear translocation of PKM2
promotes the Warburg effect. Nat Cell Biol 14, 1295-1304.
Yang, Z., and Klionsky, D.J. (2010). Mammalian autophagy: core molecular machinery and
signaling regulation. Curr Opin Cell Biol 22, 124-131.
Zhou, Y., and Liu, S. (2011). 64Cu-labeled phosphonium cations as PET radiotracers for tumor
imaging. Bioconjug Chem 22, 1459-1472.
Zunino, F., Pratesi, G., and Perego, P. (2001). Role of the sugar moiety in the pharmacological
activity of anthracyclines: development of a novel series of disaccharide analogs. Biochem
Pharmacol 61, 933-938.

Abstract (if available)

Abstract Advanced age is a risk-factor common to most cancers. In coming years, a marked increase in the population aged 65 and over will compound the already dire need for improved anti-cancer drugs. Metabolic plasticity is an adaptive mechanism tumors rely upon for growth advantage. Without sufficient energy and biomass, cell replication cannot proceed. Therefore, the use of small molecules to interrupt cancer-specific metabolic processes presents a novel means to inhibit tumor growth. The clinical benefit of this approach has not yet been fully realized due to a lack of suitable agents targeting cancer cell metabolism. The research presented in this dissertation represents efforts directed at addressing this unmet need. TP compounds, a series of novel triphenylphosphonium salts were found to inhibit cell proliferation via accumulation in mitochondria, inhibition of oxygen consumption and increases in superoxide production. As single agents TP compounds slowed tumor growth in vivo. Mito-drugs prepared by conjugation of chlorambucil, doxorubicin or temozolomide to the triphenylphosphonium moiety redirected the these drugs to mitochondria. Redirecting drug uptake in this manner maintained or enhanced the antiproliferative activity of these agents in cell-based assays and could potentially improve therapeutic profiles in vivo. In another approach to metabolic targeting, the discovery of a novel class of sulfonamides targeting the glycolytic enzyme pyruvate kinase M2 is reported. The prototype molecule, AV220 and its close analogs inhibited cell proliferation and activated AMPK, the unfolded protein response, and increased autophagic flux. As a single agent, AV220 inhibited tumor growth in a mouse xenograft model of human colon cancer.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

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

PDF

Discovery of novel small molecules for ovarian cancer treatment

PDF

Targeting cellular redox modulations for pancreatic cancer treatment

PDF

Design of novel anticancer agents targeting cellular stress response pathways

PDF

Discovery of small molecules for brain cancer treatment

PDF

Mechanistic studies of novel small molecule anti-cancer agents using next generation sequencing

PDF

Modulation of transcription and receptor function with synthetic small molecules and multi-finctional integrin ligands

PDF

Discovery of novel small-molecule compounds with anticancer properties

PDF

Identification of small-molecules targeting CXCR2 function and signaling

PDF

Using cyclotides as a bioscaffold to target intracellular protein-protein interactions

PDF

Mass spectrometry-based proteomic analysis of inhibitor of kappa b kinase beta and its role in cytokine-induced drug resistance in cancer

PDF

Blockade of CXCR2 as a novel approach for cancer chemotherapy

PDF

Engineering immunotoxin and viral vectors for cancer therapy

PDF

Treatment of HPV driven tumors with SH5-07, a small molecule inhibitor targeting STAT3

PDF

Identification of CBP/FOXM1 as a molecular target in triple negative breast cancer

PDF

Regulation of mitochondrial bioenergetics via PTEN (phosphatase and tensin homolog deleted on chromosome 10)/estrogen-related receptor alpha (ERRα) signaling

PDF

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

PDF

Steatosis induces Wnt/β-catenin pathway to stimulate proliferation of hepatic tumor initiating cells and promote liver cancer development

PDF

Small molecule modulators of HIF1α signaling

PDF

Inhibition of tumor cell growth by mefloquine via multimechanistic effects involving increased cellular stress, inhibition of autophagy, and impairment of cellular energy metabolism

Asset Metadata

Creator Millard, Melissa (author)

Core Title Discovery of novel small molecules targeting cancer cell metabolism

School School of Pharmacy

Degree Doctor of Philosophy

Degree Program Pharmaceutical Sciences

Publication Date 05/10/2013

Defense Date 03/12/2013

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

Tag breast cancer,cancer cell metabolism,cancer therapy,mitochondria,OAI-PMH Harvest,oncology,pancreatic cancer,PKM2,pyruvate kinase M2,small molecule anti-cancer agents

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Neamati, Nouri (committee chair), Camarero, Julio A. (committee member), Olenyuk, Bogdan (committee member), Stiles, Bangyan L. (committee member), Zandi, Ebrahim (committee member)

Creator Email melzmail1@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-254502

Unique identifier UC11293519

Identifier etd-MillardMel-1678.pdf (filename),usctheses-c3-254502 (legacy record id)

Legacy Identifier etd-MillardMel-1678.pdf

Dmrecord 254502

Document Type Dissertation

Format application/pdf (imt)

Rights Millard, Melissa

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