University of Southern California Dissertations and Theses

Genomic stability of transcriptionally targeted replication competent retroviral vectors

(USC Thesis Other)

Genomic stability of transcriptionally targeted replication competent retroviral vectors

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

GENOMIC STABILITY OF TRANSCRIPTIONALLY TARGETED REPLICATION
COMPETENT RETROVIRAL VECTORS

by

Laurent Yoon

________________________________________________________________________

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirement for the Degree
DOCTOR OF PHILOSOPHY
(BIOCHEMISTRY AND MOLECULAR BIOLOGY)

May 2008

Copyright 2008 Laurent Yoon

DEDICATION

I would like to dedicate this work to my family – my father who gave me strength, my
mother who gave me unconditional love, my sister who gave me confidence, and my
brother who gave me his opportunities – and my love Minhee.
ii

ACKNOWLEDGEMENTS

I would like to take this opportunity to express my greatest gratitude towards my mentor
Dr. Noriyuki Kasahara, for his guidance, support, encouragements, and words of wisdom.
I would also like to express my appreciation to past and present members of my
dissertation committee – Dr. Donald Kohn, Dr. Paula Cannon, Dr. Laurence Kedes, Dr.
W. French Anderson, and Dr. Baruch Frenkel - and the Kasahara lab at USC and UCLA
– Dr. Toshiaki Shichinohe, Dr. Ataru Sazawa, Dr. Kei Hiraoka, Dr. Takahiro Kimura, Dr.
Shuji Kubo, Dr. Kazuo Mizutani, Dr. Emmanuelle Faure, Dr. Samira Kaissi, Dr. Renata,
Stripecke, Dr. Richard Koya, Dr. Chris Logg, Dr. Chien-kuo Tai, Dr. Harris Soifer, Dr.
Brian Baranick, Aki Logg, Collin Higo, Jill Nagashima, and David Cohen. Thank you
for your invaluable advice and the privilege of working with you all.
iii

TABLE OF CONTENTS

Dedication ii

Acknowledgements iii

List of Tables vii

List of Figures viii

Abstract xii

Introduction 1

Chapter 1: Genomic stability of murine leukemia virus long terminal 6
repeat in long-term culture
Introduction 6
Materials and Methods 7
Nomenclature
Tissue Culture
Construction of Plasmids
Vector Production
Retroviral Vector Titer Determination
Luciferase and β-Galactosidase Assay
Polymerase Chain Reaction and Sequencing
Hirt Prep
Results 11
Wild type LTR of MLV is stable in U87 glioma cells in
culture without gross recombination events as analyzed by
PCR
Wild type LTR of MLV is stable in various cell lines in
culture without gross recombination events as analyzed by
PCR and restriction digests.
MLV LTR promoter strength is stable in extended
propagation in various cell lines but its enhancer sequences
are often deleted
Discussion 12

Chapter 2: Genomic stability of glucose regulated protein 78 promoter 28
hybrid long terminal repeat in replication competent murine leukemia
virus in long-term culture
iv

Introduction 28
Materials and Methods 30
Nomenclature
Tissue Culture
Construction of Plasmids
Vector Production
Retroviral Vector Titer Determination and Grp Promoter
Induction
Luciferase and β-Galactosidase Assay
Polymerase Chain Reaction and Sequencing
Hirt Prep
Results 34
Various Grp78 hybrid-LTR constructs show varying levels
of basal level activity and inducibility
RCR.Grp.GFP is able to transduce various cell lines and
induce expression of GFP transgene upon hypoglycemic
stress
The ability of RCR.Grp.GFP to replicate selectively in
under hypoglycemic condition could not be validated in
vitro assay
RCR.Grp.GFP is able to propagate in U87 cells, transduced
at varying MOI, under normal conditions without
hypoglycemic stress
RCR.Grp.GFP is able to propagate in various cell lines
under normal condition without hypoglycemic stress
Grp78 hybrid LTR is stable in various cell lines in culture
without gross recombination events analyzed by PCR and
restriction digests
Grp78 hybrid LTRs isolated from long term propagation
consistently shows similar basal level activity
Discussion 61

Chapter 3: Genomic stability of probasin-LTR hybrid promoter in 65
long-term culture
Introduction 65
Materials and Methods 66
Nomenclature
Tissue Culture
Construction of Plasmids
Vector Production
Retroviral Vector Titer Determination
Luciferase and β-Galactosidase Assay
Polymerase Chain Reaction and Sequencing
Hirt Prep
Results 70
RCR.Pro.GFP selectively replicates in androgen responsive
v

prostate cancer cells and not in other cell type
Probasin hybrid LTR is not stable in various cell lines in
long-term culture as analyzed by PCR and restriction
digests
Probasin hybrid LTR is stably selective in extended
propagation in various cell lines but its enhancer sequences
are often deleted
Discussion 79

Chapter 4: Long-term effects of promoter stability and selectivity in 87
suicide gene therapy
Introduction 87
Materials and Methods 89
Nomenclature
Tissue Culture
5-FC and Ganciclovir (GCV) Cytotoxicity Assay
Construction of Plasmids
Vector Production
Genomic DNA Isolation
Polymerase Chain Reaction
Results 92
At least up to 1mM 5-FC and 1ug/ml of GCV in MDA-MB
435, U87, and LNCaP cells are non-cytotoxic
RCR.U3.CD is able to transduce various cancer cell lines
and efficiently eradicates them with its pro-drug 5-FC
RCR.Grp.CD is able to transduce various cancer cell lines
and efficiently eradicates them under normal uninduced
condition with its pro-drug 5-FC
RCR.Grp.CD expresses detectable levels of its transgene
cytosine deaminase under normal uninduced conditions as
detected by RT-PCR
RCR.Pro.CD is able to transduce various cancer cell lines
and shows partial cell killing even in non-androgen-
responsive cancer cells with its pro-drug 5-FC
RCR.Pro.CD expresses detectable levels of its transgene
cytosine deaminase in long-term culture even in non-
androgen responsive cancer cells as detected by RT-PCR
Discussion 117

Chapter 5: Discussion & Future Directions 121
Benefits and potential safety considerations 123
Additional methods under development 128

Bibliography 133
vi

LIST OF TABLES

Table 1: Mutations found in wild-type LTR isolated from long-term 24
culture in various cells lines

Table 2: Mutations found in Grp hybrid-LTR isolated from long-term 60
cultures in various cells lines

Table 3: Mutations found in Probasin hybrid-LTR isolated from long- 81
term cultures in LNCaP human prostate cancer cells and U87
human glioma cells
vii

LIST OF FIGURES

Figure 1.1: Schematic diagram of pRCR.U3.GFP construct. 12

Figure 1.2: Fluorescence microscopy of U87 cells transduced with 13
RCR.U3.GFP shows expression of transgene – green
fluorescent protein (GFP).

Figure 1.3: RCR.U3.GFP shows stable expression of GFP transgene 14
throughout long-term cultures

Figure 1.4: Long terminal repeat (LTR) of MLV seems stable and does 15
not show any gross recombination event throughout long-
term cultures

Figure 1.5: RCR.U3.GFP shows stable expression of GFP transgene 17
throughout the a long term culture in various cell lines.

Figure 1.6: MLV LTR seems stable without any gross recombination 19
event in various cell lines in long-term cultures.

Figure 1.7: MLV LTRs seem stable and show consistent strength in long 22
-term cultures

Figure 1.8: Schematic diagram of mutations found in MLV LTRs in 23
extended culture

Figure 2.1: Schematic diagram of glucose regulated protein 78 (Grp78) 29
promoter with three endoplasmic reticulum stress response
elements (ERSE).

Figure 2.2: Schematic diagram of luciferase reporter constructs used in 35
this study

Figure 2.3: Full length Grp78 promoter is most inducible with lowest 37
background level

Figure 2.4: Schematic diagram of pRCR.Grp.GFP construct 38

viii

Figure 2.5: RCR.Grp.GFP is able to transduce MDA-MB 435 human 39
breast cancer cells and selectively induce transgene
expression (green fluorescent protein) under hypoglycemic
stress

Figure 2.6: RCR.Grp.GFP is able to transduce various cell lines and 40
selectively induce transcription of transgene under
hypoglycemic stress.

Figure 2.7: Fluorescence Microscopy of U87 cells transduced with 42
RCR.Grp.GFP shows minimal level of expression under
normal conditions without stress

Figure 2.8: Schematic diagram of RCR.Grp.GFP Replication Assay 43

Figure 2.9: Schematic diagram of pRCAdEGrpTK 44

Figure 2.10: Grp78 promoter in adenovirus RCAdGrpTK selectively 46
expresses E1a when induced by hypoglycemia

Figure 2.11: Flow diagram of RCAdGrpTK selective replication 47
competency assay by hypoglycemic stress

Figure 2.12: FACS analysis shows enhanced replication and titer of 48
RCAdGrpTK when subjected to hypoglycemic stress

Figure 2.13: RCR.Grp.GFP is able to propagate in U87 culture under 50
normal conditions without induction

Figure 2.14: Hybrid LTR with Grp78 promoter seems stable U87 and 51
does not show any gross recombination event over an
extended period of time.

Figure 2.15: RCR.Grp.GFP is able to propagate in culture under normal 53
conditions without induction

Figure 2.16: Grp78 hybrid LTR seems stable without any gross 56
recombination event in various cell lines over an extended
period of time

Figure 2.17: Grp78 hybrid LTR seems less stable than wild type LTR and 58
shows leaky expression

Figure 2.18: Schematic diagram of mutations found in glucose regulated 59
protein 78n (Grp 78) promoter in long-term cultures.

ix

Figure 2.19: Schematic diagram of insertion found in pGL2-Grp9 62

Figure 3.1: RCR.Prob.GFP is able to selectively replicate in prostate 71
cancer cells

Figure3.2: Fluorescence Microscopy of U87 cells transduced with 72
RCR.Prob.GFP shows undetectable level of expression in non-
prostate cells

Figure3.3: RCR.Prob.GFP is unable to propagate efficiently in cell types 73
other than prostate cell line

Figure 3.4: Probasin hybrid LTR seems less stable than wild type LTR 76
with gross recombination events over an extended propagation
period

Figure 3.5: Probasin hybrid LTR seems less stable than wild type LTR 78

Figure 3.6: Schematic diagram of mutations found in hybrid probasin 80
LTR

Figure 3.7: Schematic diagram of pGL2-Pro9 mutant 82

Figure 4.1: Schematic diagram of pRCR.U3.CD 88

Figure 4.2: At least up to 1mM 5-FC or 1ug/ml of GCV in MDA-MB 435 93
is non-cytotoxic

Figure 4.3: At least up to 1mM 5-FC or 1ug/ml of GCV in U87 is non- 94
cytotoxic

Figure 4.4: At least up to 1mM 5-FC or 1ug/ml of GCV in LNCaP is non 95
cytotoxic

Figure 4.5: RCR.U3.CD infects various cell lines as confirmed by 96
genomic PCR

Figure 4.6: RCR.U3.CD with its pro-drug 5-FC efficiently kills various 98
human cancer cells (A) MDA-MB 435, (B) LNCaP, and (C)
U87 without selection

Figure 4.7: RCR.U3.TK with its pro-drug GCV efficiently kills various 100
human cancer cells (A) MDA-MB 435, (B) LNCaP, and (C)
U87 without selection

x

Figure 4.8: Schematic diagram of pRCR.Grp.CD and pRCR.Grp.TK 102
constructs

Figure 4.9: RCR.Grp.CD infects various cell lines as confirmed by 103
genomic PCR.

Figure 4.10: RCR.Grp.CD with its pro-drug 5-FC partially kills various 105
human cancer cells (A) MDA-MB 435, (B) LNCaP, and (C)
U87 under non-hypoglycemic condition

Figure 4.11: RCR.Grp.TK with its pro-drug GCV partially kills various 107
human cancer cells (A) MDA-MB 435, (B) LNCaP, and (C)
U87 under normal non-hypoglycemic condition.

Figure 4.12: RT-PCR nearly 4 weeks after the initial RCR.Grp.CD 109
infection confirms expression of cytosine deaminase
transgene in RCR.Grp.CD infected cells in long-term culture
under normal non-hypoglycemic conditions in various cancer
cell lines

Figure 4.13: Schematic diagram of pRCR.Pro.CD and pRCR.Pro.TK 111
constructs

Figure 4.14: RCR.Pro.CD infects various cell lines as confirmed by 112
genomic PCR

Figure 4.15: RCR.Pro.CD with its pro-drug 5-FC partially kills human 113
cancer cells (A) MDA-MB 435 and (C) U87 and efficiently
eradicates (B) LNCaP
Figure 4.16: RCR.Pro.TK with its pro-drug GCV partially kills human 115
cancer cells (A) MDA-MB 435 and (C) U87 and efficiently
eradicates (B) LNCaP

Figure 4.17: RT-PCR nearly 4 weeks after the initial RCR.Pro.CD 118
infection confirms expression of cytosine deaminase
transgene in RCR.Pro.CD infected cells in non-androgen
responsive cancer cells in long-term culture.
xi

ABSTRACT

Advances in techniques for gene transfer and expression have made feasible the
treatment of malignancies at the genetic level by introduction of exogenous genes into
tumor cells. Numerous clinical trials of cancer gene therapy have been initiated utilizing
replication defective vectors only to face limited success due to low, often undetectable,
level of transduction efficiency. Replication competent vectors have been developed
more recently demonstrating efficient transduction as well as tissue specific expression.
These studies demonstrate the potential benefits of utilizing replication competent
retrovirus as efficient gene delivery vehicles but undermine the potential risks in their
innate tendency to mutate potentially resulting in an uncontrolled spread of replication-
competent mutant virus with adverse effects on normal cells. To understand the potential
risks in persistent long-term infection of replication competent retrovirus vectors (RCR),
our study focused on the genomic stability of the long terminal repeat (LTR) of murine
leukemia virus (MLV) responsible for driving the expression of viral genes for
replication since our vector designs to date manipulate this region to drive selective
replication of vectors. After long term propagation of various RCR vectors in various
tumor cell lines, we demonstrate here that there are regions of instability in the LTR but
very low risk in having a mutant virus with adverse effects as the very reason for utilizing
RCR vector is for efficient spread and the vectors are normally equipped with a self-
destruction mechanism.
xii

INTRODUCTION

Advances in techniques for gene transfer and expression have made feasible the
treatment of malignancies at the genetic level by introduction of exogenous genes into
tumor cells. Numerous clinical trials of cancer gene therapy have been initiated utilizing
a wide variety of therapeutic genes, including (1) cytotoxic or cytotoxin-activating
“suicide” genes, (2) pro-apoptotic genes, (3) immunogenic genes or cytokine genes, (4)
anti-oncogenes, and (5) anti-angiogenic genes. However, for any of these strategies to be
effective, efficient gene transfer to tumor tissue is required but conventional virus-based
vector systems have resulted in inadequate gene transfer efficiency (for reviews see refs.
[Robbins and Ghivizzani, 1998; Vile and Russell, 1994; Vile et al., 2000]).
Conventional virus-based vectors used for human gene therapy have generally
been rendered replication-defective by removal of key viral genes required for efficient
replication and replaced with therapeutic genes, and produced in cells that trans-
complement the defective viral elements. After infection of a target cell, the vectors are
incapable of secondary horizontal infections of adjacent cells due to the deletion of one or
multiple essential viral genes. With intra-tumoral injection of replication-defective virus
vectors, penetration into the tissue appears to be diffusion-limited, resulting in effective
gene delivery within a range of only a few cell diameters surrounding the injection site.
More efficient transduction could be achieved if a replication-competent virus
were used, as the virus would replicate and multiply after the initial infection event and
each infected tumor cell would in effect become a virus-producing cell. In the case of
1

viruses that intrinsically infect replicating cells only, such as the murine leukemia virus
(MLV) and other retroviruses in general, the pathogenetic mechanisms evolved by the
virus for replicating itself by using the host’s replication machinery can be further
utilized as a tumor targeting mechanism [Temin, 1989]. In the case of viruses that have
an intrinsically lytic life cycle, such as adenoviruses, the pathogenetic mechanisms
evolved by the virus for inducing cytolysis during viral replication can be further utilized
as a tumor cell-killing strategy (for reviews, see refs. [Kirn and McCormick, 1996;
Nemunaitis, 1999; Russell, 1994]). For those viruses with intrinsically non-lytic
lifecycle, such as the MLV, their reduced immunogenicity can be further explored to
engineer more efficient gene delivery vehicle. Vectors combining beneficial
characteristics from various types of viruses have been engineered as well in order to
make them more selective and efficient to deliver target genes [Kubo and Mitani, 2003;
Kubo et al., 2003; Mitani and Kubo, 2002; Murphy et al., 2002] (more detailed
discussion in Chapter 5).
In order to control the replication of the virus and to control the expression of
exogenous gene in select tissues, various tissue-specific and inducible promoters have
been engineered into the virus controlling the expression of key gene(s) responsible for
replication, such as gag, env, and pol in retroviruses, or the expression of exogenous
gene(s) such as cytosine deaminase (CD) and herpes simplex virus thymidine kinase
(HSV-tk) suicide genes for suicide gene therapy and have shown promising results
[Anderson et al., 1999; Brand et al., 1998; Dong et al., 2004; Logg et al., 2002; Varda-
Bloom et al., 2001; Yu et al., 2004].
These previous studies demonstrate the potential benefits of utilizing replication
2

competent retrovirus as efficient gene delivery vehicles but undermine the potential risks
in their innate tendency to mutate potentially resulting in an uncontrolled spread of
replication-competent mutant virus with adverse effects on normal cells.
To understand the potential risks in persistent long-term infection of replication
competent retrovirus vectors (RCR), our study focused on the genomic stability of the
long terminal repeat (LTR) of murine leukemia virus (MLV) responsible for driving the
expression of viral genes for replication since our vector designs to date manipulate this
region to drive selective replication of vectors. We utilized three different promoters of
varying strength and selectivity to drive the replication of murine leukemia virus (MLV)
– the wild-type U3 of MLV long terminal repeat (LTR), Glucose Regulated Protein 78
kilo-Dalton (Grp78) promoter, and the probasin promoter. The wild type MLV LTR is
strong and unselective widely utilized in current vector designs; Grp78 promoter is weak
and selective with high basal level activity in general; and the probasin promoter is highly
selective [Claessens et al., 2001; Dai and Burnstein, 1996; Zhang et al., 2000] and strong
showing similar replicative kinetics as the wild type LTR (to be discussed in more detail
in chapter 2).
MLV is an RNA virus with several stages in its life cycle where genetic mutations
can take place. First, MLV must convert its RNA genome to DNA during its replication
cycle before integrating the DNA form of the viral genome into the host genome where
its genes are transcribed by host’s RNA polymerase II. The process of converting from
RNA to DNA is called reverse transcription and it is carried out by a viral enzyme
reverse transcriptase (for review see ref [Hu and Temin, 1990b]). This enzyme is
naturally error-prone and does not have error-correcting function, probably due to lack of
3

accessory proteins in the cytoplasm and nuclease activity, enabling retroviruses to mutate
at a relatively high rate. These mutations include base-pair substitutions, frameshifts,
deletions, and insertions, as well as combinations of these alterations. At normal
physiological nucleotide concentrations, mismatch extension can occur during normal
retroviral replication and it has been demonstrated that reverse transcriptase has no
problem extending mismatches of as many as three bases under normal in vivo conditions
[Pulsinelli and Temin, 1994].
Secondly, converting RNA to DNA requires reverse transcriptase to facilitate two
template switches and these steps, first one during the plus strand primer transfer and the
other during negative strand primer transfer, are also error prone potentially leading to
mutant progeny. It has been suggested that the requirement for the reverse transcriptase
to facilitate these primer jumps may lead to the overall high error rate [Pulsinelli and
Temin, 1991; Temin, 1981].
Thirdly, it has been shown that there is about 2% per kilo-base rate of
recombination between the two retrovirus RNA genome in a virion since retroviruses are
pseudodiploids [Hu and Temin, 1990a] – that is, there are two copies of RNA genome
per virion. Recombination between two retroviruses has not been observed after co-
infection of two differently marked retroviruses, though theoretically possible, but
recombination has been observed from progeny produced from cells co-infected with the
two differently marked retroviruses [Hu and Temin, 1990a; Weiss et al., 1973]
suggesting that recombination occurs at the DNA level after reverse transcription or
between the co-packaged viral RNA’s.
Lastly, it has been suggested that viruses could pick up host sequences by
4

recombination events leading to viruses with proto-oncogene sequences such as c-myc
[Colby et al., 1983; Feuerman et al., 1988; Vennstrom et al., 1982; Watson et al., 1983].
Overall, these mutations lead to high variability and ability to meet the needs of the
challenging environments for viruses to successfully spread their progeny
Furthermore, the current method of producing retrovirus vectors – transfection- is
inherent with errors as well. Not only homologous recombination between plasmids a
potential threat but transcription beyond the poly-A signal in 3’ LTR may lead to
transcription of the U3 region of the 5’ LTR giving rise to transcripts with CMV
sequences.
Due to the error prone reverse transcription process, engineered viruses with
hybrid promoters of varying strength and selectivity can potentially mutate giving rise to
uncontrolled replication-competent viruses. Our study addresses the risks from
potentially undesirable consequences of mutations accrued from these error prone process
of retrovirus vector production and replication for transcriptionally targeted cancer gene
therapy.

5

CHAPTER 1
GENOMIC STABILITY OF MURINE LEUKEMIA VIRUS LONG TERMINAL
REPEAT IN LONG-TERM CULTURE

INTRODUCTION
Replication defective MLV vectors have been widely used to date as gene
delivery vehicles in the lab and for experimental therapies due to its simple and well-
characterized biology (for review see ref. [Hu and Temin, 1990b]), and replication
competent retroviral vectors (RCR) have been developed more recently due to its strong
and unselective promoter allowing MLV to replicate efficiently in many cell types
[Bachrach et al., 2003; Logg and Kasahara, 2004].
Previous works have shown that MLV based vectors containing an internal
ribosome entry site (IRES) and a transgene cassette inserted just between the envelope
gene and the 3’ untranslated region (UTR) exhibit greatly improved functional and
genomic stability compared with all other reported MLV-based RCR vector designs,
which used transgene insertion positions within 3’ LTR U3 region [Logg et al., 2001a;
Stuhlmann et al., 1989]. We monitored the genomic stability of RCR vectors containing
IRES-GFP marker gene cassette over multiple passages by Southern blot analysis as well
as by genomic PCR of proviral DNA to detect any recombination or deletion events
[Logg et al., 2001a]. This demonstrated the genomic stability of the transgene and
validated its expression but failed to address the possible changes in the transcriptional
regulatory elements of MLV, located in the LTR, which regulate the expression of all
6

viral genes in the MLV including the inserted transgene [Hu and Temin, 1990b; Logg et
al., 2001a].
The U3 region of the LTR is responsible for driving the expression of viral genes
and it mainly consists of two tandem repeats of 75-bp enhancers that are required for
efficient replication of the virus. However, it has been previously reported that large
direct repeats, tandem or spaced up to 1-kilobases, are not stable in replicating retrovirus
[Rhode et al., 1987] and recombination events leading to deletion of these repeats would
render the promoter less effective. Also, as described in the Introduction, the reverse
transcriptase itself is error prone as well leading to point mutations. There are
transcription enhancer binding sites, such as NF-κB and AP-1, that would decrease the
strength of the promoter if mutated and there is Oct-1 binding site which trans-activate in
T-cells making MLV replication stronger and more selective [Bandres and Ratner, 1994;
Bassuk et al., 1997; Bohnlein et al., 1989; Bramblett et al., 1995; Gruters et al., 1991;
Hannibal et al., 1994; Liu and Latchman, 1997; Parrott et al., 1991]. In long-term
propagation of the virus, it’s almost certain that there will be mutations throughout the
vector genome and those accrued in the LTR could potentially lead to overall adverse
effects. This chapter addresses the stability of the U3 region in long-term culture in
various cancer cell lines and its effects on strength and selectivity of the wild type LTR.

MATERIALS AND METHODS

Nomenclature
A “p” at the beginning of construct names designates the plasmid vectors,
7

whereas names without a “p” denote viruses derived from the corresponding plasmid
vector. All sequence coordinates are given relative to the 5’ end of the 5’ long terminal
repeat (LTR) in the plasmid unless indicated.

Tissue Culture
MDA-MB 435 human breast cancer cells, U87 human glioma cells, and 293A and
293T human embryonic kidney cells were cultivated in DMEM with 4.5g glucose/ml and
10% fetal bovine serum (FBS). LNCaP human prostate cancer cells were cultivated in
RPMI 1640 with 10% FBS.

Construction of Plasmids
The designation pACE-GFP from Logg et al. [Logg et al., 2002; Logg et al.,
2001a] is referred to as pRCR.U3.GFP in our study for simplification and consistency
with other constructs used. It contains full-length amphotropic replication competent
retrovirus (RCR) vector encoding GFP, and cytomegalovirus (CMV) promoter in place of
the 5’ U3 region to drive the initial expression after transfection.

Vector Production
Retrovirus vector stocks were produced by transfection of 293T cells with vector
plasmids using Fugene 6 (Roche). Supernatant containing vectors was collected at 48,
60, and 72 hours post transfection and filtered using 0.45 μm filters. Filtered supernatant
was stored in –80 degrees Celsius until used.

8

Retroviral Vector Titer Determination
Serial dilutions of viral supernatant in a total volume of 1 mL culture medium
were added to 293A cells at 1 x 10
5
cells in six-well plates. Twenty-four hours post-
infection, the cells were incubated in media with 50 uM 3’-azido-3’deoxythymidine
(AZT) for 24 hours and subject to FACS analysis to quantitate GFP fluorescence. The
viral titer was calculated according to the formula: transducing units (TU)/mL = (number
of cells counted immediately before infection x percentage of transduced cells reported
from FACS analysis)/dilution factor of viral supernatant.

Luciferase and β-Galactosidase Assay
293T cells at 70 to 90% confluency in 6-well dish were transfected with 2 ug of
pGL2 constructs with mutant promoters driving luciferase expression using Fugene 6
(Roche). PGL2-LTR, which has long terminal repeat (LTR) of murine leukemia virus
(MLV), was used for positive control and pGL2-Basic which lacks a promoter driving
luciferase expression was used for negative control. Each culture was co-transfected with
pCH-110, which expresses β-galactosidase (Promega) for transfection control. At 48
hours post-transfection, reporter lysis buffer (Promega) was used to prepare cell extracts.
Luciferase and β-galactosidase activities were measured with luciferase assay system and
β-galactosidase enzyme assay system (Promega), respectively.

Polymerase Chain Reaction and Sequencing
For amplification of hybrid LTR to observe gross recombination, Platinum Taq
DNA Polymerase (Invitrogen) was used. 300ng of sample DNA in 50 ul reactions
9

containing 20 mmol Tris-HCl (pH 8.4), 50 mmol KCl, 0.2mM dNTP, 1.5mM MgCl
2
, 15
pmol of primers (forward primer 5’ – AATGAAAGACCCCACCTG- TAG – 3’ &
reverse primer 5’ – ATACACCAAGACCATCCTCTGC), and 1 unit of Platinum Taq.
The first step was to heat the polymerase chain reaction (PCR) samples to 94 degrees
Celsius for 1 minute and 30 seconds to activate the enzyme. This was followed by 30
cycles of incubation at 94 degrees Celsius for 30 seconds, 56 degrees Celsius for 45
seconds, and then 72 degrees for 1 minute. For amplification of hybrid LTR to clone into
pGL2 luciferase expression plasmid, Pfu Ultra (Stratagene) was used. 200 ng of Hirt
prep DNA in 50 uL reactions containing 1X Pfu Ultra buffer, 0.2 mM dNTP, 15 pmol of
each primer, and 2.5 units of Pfu Ultra. The first step was to heat the PCR samples to 95
degrees Celsius for 2 minutes to activate the enzyme. This was followed by 30 cycles of
95 degrees Celsius for 30 seconds to denature the DNA, 56 degrees Celsius for 45
seconds for primer annealing to template, and then 72 degrees for 1 minute 15 seconds
for extension. 18-mer oligonucleotide, 5’- GGCGTCTTCCATTTTACCAACAGTA
CCG – 3’, downstream of the multiple cloning site of pGL2 was used for sequencing in
the reverse direction.

Hirt Prep
Cells were washed twice with 1X PBS and lysed in 1 mL of Hirt lysis buffer
(0.6% SDS, 10mM EDTA pH 7.5). After 15 minutes in room temperature, 250 uL of 5M
NaCl was added to the lysate and kept in 4 degrees Celsius overnight. Precipitated lysate
was spun at greater than 20,000 G and RNase was added to the supernatant and incubated
at 37 degrees Celsius for an hour. Resulting solution was phenol-extracted and DNA was
10

recovered by isopropanol precipitation

RESULTS

Wild type LTR of MLV is stable in U87 glioma cells in culture without gross
recombination events as analyzed by PCR
In order to investigate the stability of the genomic structure of the LTR, RCR
vector with wild type LTR and IRES-GFP cassette, RCR.U3.GFP (figure 1.1), was
propagated in U87 human glioma cells in extended culture for over 35 days. U87 cells
were transduced at multiplicity of infection (MOI) of 0.05, 0.5, and 5 and partial cultures
were collected on days 3, 7 12, 18, 25, and 35 for analysis. We have previously
demonstrated that stability of GFP expression by FACS is a reliable surrogate marker for
the structural stability of the RCR vector genome. We therefore confirmed the spread of
the vector throughout the U87 culture by monitoring the spread of GFP (figure 1.2 &
1.3). When GFP positive cells reached over 90%, the cells were mixed with fresh
uninfected U87 cells at about 1:1 ratio. Extrachromosomal DNA from the transduced
cells was isolated using Hirt prep protocol and PCR was performed on this isolated
extrachromosomal DNA using primers specific for the 5’ end of U3 region of the LTR
and the end of the LTR’s R region. Gel electrophoresis show consistent size of the
amplified LTR region for over a month at various initial MOIs but also show faint
smaller band that is between 50 and 100 base pairs less than the major band (figure 1.4).
11

CMV R U5 gag pol env U3 R U5 IRES GFP
ψ
A
R U5 gag pol env U3 R U5 IRES GFP
ψ
U3 B
Insert
CMV R U5 gag pol env U3 R U5 IRES GFP
ψ
A
R U5 gag pol env U3 R U5 IRES GFP
ψ
U3 B
Insert
CMV R U5 gag pol env U3 R U5 IRES GFP
ψ
A
R U5 gag pol env U3 R U5 IRES GFP
ψ
U3 B
Insert
Figure 1.1: Schematic diagram of pRCR.U3.GFP construct. (A)
pRCR.U3.GFP construct with packaging signal and CMV promoter
driving the initial transcription of viral genes gag, pol, and env and
GFP transgene. (B) viral genome in its DNA form, called provirus,
after reverse transcription. 3’ U3 is duplicated and replaces CMV
promoter at the 5’ LTR during reverse transcription.

12
B A
Figure 1.2: Fluorescence microscopy of U87 cells transduced
with RCR.U3.GFP shows expression of transgene – green
fluorescent protein (GFP). (A) untransduced U87 control and (B)
RCR.U3.GFP transduced U87 cells 10 days after transduction

13
0
20
40
60
80
100
120
0 5 10 15 20 25 30
Day
% GFP Positive
Figure 1.3: RCR.U3.GFP shows stable expression of GFP transgene
throughout long-term cultures. U87 cells were transduced with
RCR.U3.GFP at varying multiplicity of infection (MOI) and cultured
under normal conditions. Square blocks represent MOI of 5, diamonds
represent MOI of 0.5 and triangles represent MOI of 0.05.

14
MOI 5
MOI 0.5
MOI 0.05
C D3 D7 D12 D18 D25 D35
C D3 D7 D12 D18 D25 D35
C D3 D7 D12 D18 D25 D35
minor band
minor band
minor band
MOI 5
MOI 0.5
MOI 0.05
C D3 D7 D12 D18 D25 D35
C D3 D7 D12 D18 D25 D35
C D3 D7 D12 D18 D25 D35
MOI 5
MOI 0.5
MOI 0.05
C D3 D7 D12 D18 D25 D35
C D3 D7 D12 D18 D25 D35
C D3 D7 D12 D18 D25 D35
C D3 D7 D12 D18 D25 D35 C D3 D7 D12 D18 D25 D35
C D3 D7 D12 D18 D25 D35 C D3 D7 D12 D18 D25 D35
C D3 D7 D12 D18 D25 D35 C D3 D7 D12 D18 D25 D35
minor band
minor band
minor band
Figure 1.4: Long terminal repeat (LTR) of MLV seems stable and does
not show major gross recombination event throughout long-term
cultures. U87 human glioma cells were transduced at MOI of 0.05, 0.5,
or 5 with RCR.U3.GFP. Data show PCR amplification of LTRs from
Hirt prep DNA collected on days indicated with “D.” “C” is control
PCR from pRCR.U3.GFP plasmid and blank lane indicates
unsuccessful attempts to amplify the fragment.

15
Band intensity of the smaller bands suggests presence of minor species throughout the
culture but they fail to dominate the entire culture as the consistently high intensity of
major bands suggest. Also, since cultures transduced at lower MOI must replicate more
to spread throughout the culture than cultures transduced at higher MOI, our results
demonstrate that the initial transduction efficiency does not affect the mutation
frequency. These results demonstrate lack of gross recombination event leading to
dominance of a recombinant species but not lack of smaller point mutations or
recombination events resulting in similar size PCR amplified fragments.

Wild type LTR of MLV is stable in various cell lines in culture without gross
recombination events as analyzed by PCR and restriction digests.
For further analysis of the MLV LTR in extended propagation, MDA-MB 435
human breast cancer cells, U87 human glioma cells, and LNCaP human prostate cancer
cells were transduced with RCR.U3.GFP at an MOI of at least 0.5 and propagated in
extended cultures for 45 days. Using GFP as a surrogate marker, vector spread was
detected by FACS and fresh untransduced cells were added at ratio of about 1:1 when
GFP positive cells reached over 90% of culture. For all three cell lines, MLV LTR stably
express consistent GFP transgene expression in culture for 45 days (figure 1.5).
Partial cultures were collected on day 35 from all three cell lines and
extrachromosomal DNA was prepared using the Hirt prep protocol. As previously, MLV
LTR specific primers were used to PCR amplify the U3 region of the LTR (figure 1.6A).
16

A
0
20
40
60
80
100
120
0 1020304050
Day
% GFP Positive
Figure 1.5: RCR.U3.GFP shows stable expression of GFP transgene
throughout long-term culture in various cell lines. (A), (B), and (C)
show spread of RCR.U3.GFP in MDA-MB 435, U87, and LNCaP
cells respectively. Cells were re-plated with fresh untransduced cells
every three days once GFP levels were over 90%. Approximate ratio
of GFP positive to fresh cells was 1:1. This experiment was done in
triplicates and indicated by squares, diamonds, and triangles.

17
Figure 1.5, Continued
Figure 1.5 B
0
20
40
60
80
100
120
0 1020 30405
Day
% G FP Po sitiv e
0
0
20
40
60
80
100
120
0 1020 30405
Day
% G FP Po sitiv e
Figure 1.5 C
0

18
+ -UML
A
U87 RCR.GFP
(+) Control Digested
MDA-MB435 RCR.GFP
LnCap RCR.GFP
(+) Control
B
C
minor
band
*
*
*
* 650 bp
+ -UML
A
U87 RCR.GFP
(+) Control Digested
MDA-MB435 RCR.GFP
LnCap RCR.GFP
(+) Control
U87 RCR.GFP
(+) Control Digested
MDA-MB435 RCR.GFP
LnCap RCR.GFP
(+) Control
(+) Control Digested
MDA-MB435 RCR.GFP
LnCap RCR.GFP
(+) Control
B
C
minor
band
*
*
*
* 650 bp
Figure 1.6: MLV LTR seems stable by length without gross
recombination event in various cell lines in long-term cultures.
Cells were harvested on day 35 of culture post RCR.U3.GFP
transduction and extrachromosomal DNA was collected using Hirt
prep protocol. MLV LTR was amplified by MLV LTR specific
promoters. (A) “-” is negative control, “+” is positive control with
pRCR.U3.GFP, “U” is U87 human glioma cell line, “M” is MDA-
MB 435 human breast cancer cell line, and “L” is LNCaP human
prostate cancer cell line. Amplified PCR products were digested
with (B) Ava II or (C) Hae III, which recognizes very common
sequences, to detect small recombination events.

19
We demonstrate here that the amplified fragments from each cell line are consistent with
the PCR control from vector plasmid suggesting no major gross recombination events
during propagation of the vector. These PCR fragments from Hirt prep DNA were
further analyzed by restriction digests to look for smaller recombination events (figure
1.6B & C). Restriction enzymes Hae III and Ava II were used since they recognize
frequently occurring sequences of four or five nucleotides, respectively –5’-GGCC-3’ for
Hae III and GG(A/T)CC for Ava II. Also, by using restriction enzymes that recognize
sequences with G’s, we wanted to see whether or not APOBEC3G, which is a host
enzyme that have been implicated in host defense by mutating retroviral RNA genome,
affects the strength of the promoter by mutating G to A [Turelli et al., 2004]. Consistent
restriction digest patterns suggest again lack of gross recombination events but do not
rule out the possibility of point mutations that may affect promoter selectivity and
strength.

MLV LTR promoter strength is stable in extended propagation in various cell lines but its
enhancer sequences are often deleted
MLV LTR is a constitutively strong promoter and we demonstrate here that the
promoter is stably strong after 35 days of propagation in various cell lines. PCR
amplified LTRs from extended propagation of MLV in various cell lines were inserted
into pGL2 luciferase reporter plasmid upstream of the luciferase gene. Positive clones
with LTR inserts were identified by PCR with primers flanking the insert and these
clones were transfected into 293T cells. pCH-110 β-galactosidase reporter plasmid was
transfected in parallel with pGL2 constructs to normalize for transfection efficiency. 36
20

hours post-transfection, cells were harvested to measure luciferase activity, which would
indicate the strength of the LTR promoter with accrued mutations after 35 days of
propagation in various cells lines (figure 1.7). Nine clones tested show consistent level of
luciferase activity compared with the positive control in which the LTR was PCR
amplified from pRCR.U3.GFP plasmid. Results indicate that the LTR promoter is stably
strong in extended culture but it is possible that acquired mutations do not affect the
strength of the promoter in 293T cells.
PCR amplified fragments of LTRs from extended cultures were sequenced to
identify regions(s) of instability. Sequence results show several point mutations that
seem random and do not affect the promoter strength significantly (figures 1.7, 1.8, and
table 1). More interestingly, the first enhancer sequence of two in MLV LTR was often
deleted but this did not affect the strength of the promoter in 293T cells.

DISCUSSION
The results of our study demonstrate that the MLV LTR promoter is stably strong
even with molecular lesions acquired during extended propagation in various cell lines.
Furthermore, as demonstrated in previous studies, MLV LTR is able to consistently drive
the expression of an inserted transgene in multiple passages in extended culture in
various cell lines [Logg et al., 2001a; Logg et al., 2001b].
However, since reverse transcriptase, the enzyme that converts RNA to DNA in
retroviruses, is naturally error-prone enabling retroviruses to mutate at a relatively high
rate, mutations such as base-pair substitutions, frameshifts, deletions, and insertions were
21

1
10
100
1000
10000
100000
1000000
pGL2-Basic
pGL2-LTR C
pGL2-LTR 1
pGL2-LTR 2
pGL2-LTR 3
pGL2-LTR 4
pGL2-LTR 5
pGL2-LTR 6
pGL2-LTR 7
pGL2-LTR 8
pGL2-LTR 9
Relative Luciferase Units
Figure 1.7: MLV LTRs seem stable and show consistent strength in
long-term cultures. LTRs of RCR.U3.GFP were amplified from Hirt
prep DNA samples prepared on day 35 of vector propagation in
various cell lines and these were cloned into pGL-2 luciferase
reporter construct. These constructs were transfected into 293T cells
with -gal plasmids to normalize for transfection efficiency. pGL2-
Basic negative control without a promoter and pGL2-LTR C is
positive control with wild type LTR driving luciferase expression.

22

Enhancer 1
1449
Enhancer 2
109 192 267
“R” region
deletion
Enhancer 1
1449
Enhancer 2
109 192 267
“R” region
deletion
Enhancer 1
1449
Enhancer 2
109 192 267
“R” region “R” region
deletion
Figure 1.8: Schematic diagram of mutations found in MLV LTRs in
extended culture. Sequence analysis shows various mutations
throughout the U3 region of the MLV LTR. Samples pGL2-LTR 5
thru 9 had deletions in enhancer 1. All except pGL2-LTR2 had
random point mutations. pGL2-LTR2 did not have any mutation in
the region sequenced (see table 1). Solid lines within the promoter
schematic represent mutations. See table for more detail information.
23

Table 1: Mutations found in wild-type LTR isolated from long-term culture in various
cells lines

Clone Cell Type Mutations
pGL2-LTR 1 LNCaP • 362 (G to A)
• 449 (G to C)
• 526 (G to A) R region
pGL2-LTR 2 MDA-MB 435 • None
pGL2-LTR 3 MDA-MB 435 • Enhancer 1 deletion
pGL2-LTR 4 LNCaP • Enhancer 1 deletion
• 526 (G to A) R region
pGL2-LTR 5 MDA-MB 435 • 172 C deletion
• 193 (A to G)
pGL2-LTR 6 LNCaP • Enhancer 1 deletion
• 358 (A to G)
pGL2-LTR 7 LNCaP • Enhancer 1 deletion
• 526 (G to A) R region
pGL2-LTR 8 LNCaP • Enhancer 1 deletion
• 358 (A to G)
• 449 (G to C)
• 535 (C to T) R region
pGL2-LTR 9 LNCaP • 358 (A to G)
• 362 (G to A)

24

observed in this study as well as previous studies [Hu and Temin, 1990b]. Dougherty et
al have shown using an SNV-based vector that the error frequency of reverse
transcriptase is not very high – the base pair substitution rate is 2 x 10
-5
per base pair per
replication cycle and the insertion rate is 10-7 per base pair per replication cycle
[Dougherty and Temin, 1988]. Also, reverse transcriptase can extend mismatches
ranging in size from 1 to 3 bases [Creighton et al., 1992; Mendelman et al., 1990; Perrino
et al., 1989; Pulsinelli and Temin, 1994; Yu and Goodman, 1992]. Accrual of these
mutations that do not lower the strength of the promoter or that result in more efficient
replication of the virus are more likely to become permanent by positively contributing to
the molecular evolution of these viruses in enabling them to meet the needs of
challenging environments and successfully reproduce.
Retroviruses are prone to mutations not only due to the misincorporation by
retrotranscriptase but also due to the actual process of reverse transcription. Reverse
transcription requires two primer transfer steps (plus-strand primer transfer and minus-
strand primer transfer; for further details of the reverse transcription process, see ref. [Hu
and Temin, 1990b]) and these steps can generate mutant progeny. In fact, it has been
suggested that this requirement for reverse transcriptase to facilitate primer jumps may
lead to the overall high error rate, not the innate mutation rate of the reverse transcriptase
enzyme itself [Temin, 1993]. Also, in HIV and SNV, the high HIV-1 misincorporation
rate is not due to a higher frequency of incorporating incorrect nucleotides into the
growing strand but rather to a higher frequency of extending these incorrect nucleotides
after they are incorporated [Perrino et al., 1989; Pulsinelli and Temin, 1994] .
In our current study, individual clones differed in sequence from the original
25

vector by various point mutations that seemed random and these mutations did not affect
the strength of the promoter significantly. However, species with mutations that weaken
the promoter would be dominated by those with wild type promoter or dominated by
those with mutations that further enhance the strength of the promoter.
Rhode et al have previously reported that direct repeats, tandem or spaced up to 1-
kilobase, are not stable in replicating retroviruses [Rhode et al., 1987]. Consistent with
this report, five of nine clones in our current study had deletions in one of two 75-bp
enhancer sequences in the U3 region. Structural genes were not analyzed for stability
leading to aberrant characteristics of the vector in our study but it has been shown that
small tandem direct repeats in structural genes are also a source of variations in
retroviruses [Alizon et al., 1986]. Although not observed in our study, it has been
reported previously that duplication of enhancer sequences resulting in more than two
copies and deletion of both enhancer sequences do occur [Ito et al., 1998]. Our transient
transfection studies in 293T cells with luciferase reporter plasmid show that deletions in
one enhancer while maintaining the other do not significantly affect the strength of the
promoter in 293T, which is a human cell line. However, such mutants in murine cell
lines have shown to result in less efficient production of progeny from infected cells
[LoSardo et al., 1989].
Experiments with hybrid LTRs between Moloney MLV and Friend MLV (which
induces erythroleukemia) have implicated the tandem repeats as the primary determinants
of disease specificity in the M-MLV LTR [DesGroseillers et al., 1983; Ishimoto et al.,
1987; Li et al., 1987; Thiesen et al., 1988]. Also, 23 bases of GC-rich sequences
following the tandem repeats have been implicated in the T-lymphoid specificity of the
26

wild-type M-MLV LTR. Without that GC-rich region, the virus’s disease specificity
broadened causing T and B lymphoid, acute myeloid, and erythroid leukemias in mice
while the wild-type M-MLV with all of its sequences intact has more specific disease
specificity causing T lymphoid tumors [Hanecak et al., 1991].
As demonstrated in this chapter, there are various species with mutations,
including deletion of an enhancer element, accrued over time that coexist in long-term
propagation. The species with the most efficient mode of replication with a strong
promoter would become dominant but our luciferase data suggest that the mutations
accrued during propagation, even the deletion of an enhancer, did not affect the strength
of the promoter significantly in the cell lines tested. As the very reason for using the wild
type LTR in RCR vector is to ensure effective spread, our data suggest that LTR in long-
term propagation is stably effective. It’s possible to have minor species as shown here
that may potentially be less effective in replication but these will be dominated by the
more efficiently replicating vectors and the suicide gene will serve as a self-destruction
mechanism anyway (further discussion on this topic in Chapter 5).
27

CHAPTER 2
GENOMIC STABILITY OF GLUCOSE REGUATED PROTEIN 78 PROMOTER
HYBRID LTR IN REPLICATION COMPETENT MURINE LEUKEMIA VIRUS IN
LONG-TERM CULTURE

INTRODUCTION
Glucose regulated proteins were first identified as proteins that are expressed in
eukaryotic cells when depleted of glucose [Shiu et al., 1977]. Since then, it has been
found that these proteins are up regulated by variety of reagents that inhibit protein
glycosylation or disrupt intracellular calcium storage [Chang et al., 1987; Drummond et
al., 1987; Kim and Lee, 1987]. The promoter regions of these proteins have been
analyzed in human, rat, murine and rabbit cells and shown to consist conserved
sequences called endoplasmic reticulum stress elements (ERSE) [Parker et al., 2001] that
are responsible for the selective expression upon stress (figure 2.1).
Glucose Regulated Protein-78 KDa (Grp78) promoter is a weak but inducible
promoter activated by hypoglycemia, hypoxia, heat-shock, and various chemical stress
[Chen et al., 2000a; Hou et al., 1993; Koong et al., 1994; Song et al., 2001]. Also,
Grp78 protein has been shown to be overexpressed in some tumors making its promoter a
suitable candidate for cancer gene therapy [Chen et al., 2000b; Dong et al., 2004;
Fernandez et al., 2000; Gazit et al., 1999; Shuda et al., 2003]. Grp78 consists of 3 ERSEs
that drives selective expression under stressful environments [Li et al., 2000]. It has been
28

ERSE ERSE ERSE
-524
+174
-60
TATA
-131 -163 -98
ERSE ERSE ERSE
-524
+174
-60
TATA
-131 -163 -98
Figure 2.1: Schematic diagram of glucose regulated protein 78
(Grp78) promoter with three endoplasmic reticulum stress
response elements (ERSE).

29
shown that the loss of each ERSE results in decreased inducibility of the
promoter under stress [Li et al., 2000], hence the sequence as well as the structural
integrity of ERSEs is crucial to its inducibility.
In this chapter, we engineered an RCR vector with Grp78 promoter driving the
expression viral and exogenous genes and replication of the vector to explore the
genomic stability of hybrid-LTR in long-term propagation in various cancer cell lines.
As the recombinant RCR with Grp78 promoter would be more selective in replication,
we can quickly identify mutant species that are non-selective in replication due to
mutations accrued from long-term propagation that would be indicative of potential risks
in cancer gene therapy with tissue specific or inducible RCR vectors.

MATERIALS AND METHODS

Nomenclature
A “p” at the beginning of construct names designates the plasmid vectors,
whereas names without a “p” denote viruses derived from the corresponding plasmid
vector. All sequence coordinates are given relative to the 5’ end of the 5’ long terminal
repeat (LTR) in the plasmid, unless indicated.

Tissue Culture
MDA-MB 435 human breast cancer cells, MDA-MB 231 human breast cancer
cells, U87 human glioma cells, A172 human glioma cells, HeLa, and 293A and 293T
human embryonic kidney cells were cultivated in DMEM with 4.5g glucose/ml and 10%
30

fetal bovine serum (FBS). J774.1A murine macrophage cells and LNCaP human prostate
cancer cells were cultivated in RPMI 1640 with 10% FBS.
Construction of Plasmids
All luciferase reporter constructs for this study were derived from pGL2
luciferase reporter construct (Promega). pGL2-Grp.TATA contains Grp78 promoter
from –457 to –50 with all three endoplasmic reticulum stress elements and its own TATA
box. pGL2-Grp.MLV.TATA contains Grp78 promoter from –457 to –60 fused at MLV
TATA box. pGL2-LTR has complete MLV LTR with U3, R and U5. All retrovirus
vector constructs for this study were derived from pACE-GFP-dm [Logg et al., 2001a],
which contains full-length amphotropic RCR vector encoding GFP, and cytomegalovirus
(CMV) promoter in place of the 5’ U3 region. pRCR.Grp.GFP was constructed by taking
out U3 region of the MLV LTR from Nhe I and Sac I, and Grp78 promoter from –457 to
–60 was inserted in place of MLV U3.

Vector Production
Retrovirus vector stocks were produced by transfection of 293T cells with vector
plasmids using Fugene 6 (Roche). Supernatant containing vectors was collected at 48,
60, and 72 hours post transfection and filtered using 0.45 μm filters. Filtered supernatant
was stored in –80 degrees Celsius until used.

Retroviral Vector Titer Determination and Grp Promoter Induction
For RCR.U3.GFP, serial dilutions of viral supernatant in a total volume of 1mL
culture medium were added to 293A cells at 1 x 10
5
cells in six-well plates. Twelve
31

hours post-transduction, cells were incubated in medium with 50uM 3’-azido-
3’deoxythymidine (AZT) for 24 hours to prevent horizontal spread and subject to FACS
analysis to quantitate GFP fluorescence. The viral titer was calculated according to the
formula: transducing units (TU)/mL = (number of cells counted immediately before
infection x percentage of transduced cells reported from FACS analysis)/dilution factor
of viral supernatant.
For RCR.Grp.GFP, serial dilutions of viral supernatant in a total volume of 1 mL
culture medium were added to 293A cells at 1 x 10
5
cells in six-well plates. Twelve
hours post-transduction, media was changed to glucose and pyruvate free media
(Invitrogen) with 10% dialyzed FBS containing minimal level of glucose. This
hypoglycemic media also contained 50uM AZT to prevent possible horizontal spread.
Twenty-four hours later, cells were collected and subjected to FACS analysis to
quantitate GFP fluorescence.

Luciferase and β-Galactosidase Assay
293T cells at 70 to 90% confluency in 6-well dish were transfected with 2ug of
pGL2 constructs with mutant promoters driving luciferase expression using Fugene 6
(Roche). pGL2-LTR, which has long terminal repeat (LTR) of murine leukemia virus
(MLV), was transfected in parallel for positive control and pGL2-Basic which lacks a
promoter driving luciferase expression was transfected in parallel for negative control.
Each culture was co-transfected with pCH-110, which expresses β-galactosidase
(Promega) for transfection control. At 48 hours post-transfection, reporter lysis buffer
(Promega) was used to prepare cell extracts. Luciferase and β-galactosidase activities
32

were measured with luciferase assay system and β-galactosidase enzyme assay system
(Promega), respectively.

Polymerase Chain Reaction and Sequencing
For amplification of hybrid LTR to observe gross recombination, Platinum Taq DNA
Polymerase (Invitrogen) was used. 300ng of sample DNA in 50 ul reactions containing
20mmol Tris-HCl (pH 8.4), 50mmol KCl, 0.2mM dNTP, 1.5mM MgCl2, 15pmol of
primers, and 1 unit of Platinum Taq. The first step was to heat the polymerase chain
reaction (PCR) samples to 94 degrees Celsius for 1 minute and 30 seconds to activate the
enzyme. This was followed by 30 cycles of incubation at 94 degrees Celsius for 30
seconds, 56 degrees Celsius for 45 seconds, and then 72 degrees for 1 minute. For
amplification of hybrid LTR to clone into pGL2 luciferase expression plasmid, Pfu Ultra
(Stratagene) was used. 200ng of Hirt prep DNA in 50uL reactions containing 1X Pfu
Ultra buffer, 0.2mM dNTP, 15pmol of each primer, and 2.5 units of Pfu Ultra. The first
step was to heat the PCR samples to 95 degrees Celsius for 2 minutes to activate the
enzyme. This was followed by 30 cycles of 95 degrees Celsius for 30 seconds to
denature the DNA, 56 degrees Celsius for 45 seconds for primer annealing to template,
and then 72 degrees for 1 minute 15 seconds for extension. For sequencing, 18-mer oligo
downstream of the multiple cloning site of pGL2 was used.

Hirt Prep
Cells were washed twice with 1X PBS and lysed in 1mL of Hirt lysis buffer
(0.6% SDS, 10mM EDTA pH 7.5). After 15 minutes in room temperature, 250uL of 5M
33

NaCl was added to the lysate and kept in 4 degrees Celsius overnight. Precipitated lysate
was spun at greater than 20,000 G and RNase was added to the supernatant and incubated
at 37 degrees Celsius for an hour. Resulting solution was phenol-extracted and DNA was
recovered by isopropanol precipitation.

RESULTS

Various Grp78 hybrid-LTR constructs show varying levels of basal level activity and
inducibility
Selectivity and strength of hybrid promoters differ depending on the precise
location of fusion of promoters and the best version for hybrid MLV LTR is at the TATA
maintaining the MLV TATA box [Logg et al., 2002]. In order to evaluate the selectivity
and inducibility of Grp78 hybrid LTR, three versions of Grp78 hybrid LTR were cloned
into pGL2 luciferase expression plasmid upstream of the luciferase gene (Figure 2.2).
pGL2-Grp.MLV.TATA contains Grp78 promoter fused at the MLV TATA, thereby
maintaining MLV TATA, pGL2-Grp.TATA contains Grp78 promoter sequence
including Grp78 TATA box, and pGL2-GrpL.TATA contains the entire Grp78 promoter
used in various studies down to +174 [Chen et al., 1997; Dong et al., 2004; Gazit et al.,
1999; Gazit et al., 1995] showing inducibility of ~7 fold under hypoglycemic and ER
stress. These constructs were transfected into 293T cells along with pCH-110 β-gal
reporter plasmid to normalize for transfection efficiency. 36 hours post-transfection,
cells were harvested to measure luciferase activity. Consistent with previous studies, the
full length Grp78 promoter is induced by about 7-fold upon equivalent strength upon
34

C
A
B
luciferase pA
Grp78 TATA
R U5 Grp78
-524 +174
luciferase pA
Grp78 TATA
R U5 Grp78
-524 -50
ML V TATA
luciferase pA
R U5 Grp78
-524 -60
C
A
B
luciferase pA
Grp78 TATA
R U5 Grp78
-524 +174
luciferase pA
Grp78 TATA
R U5 Grp78
-524 +174
luciferase pA
Grp78 TATA
R U5 Grp78
-524 -50
luciferase pA
Grp78 TATA
R U5 Grp78
-524 -50
ML V TATA
luciferase pA
R U5 Grp78
-524 -60
ML V TATA
luciferase pA
R U5 Grp78
-524 -60
Figure 2.2: Schematic diagram of luciferase reporter constructs
used in this study. (A) pGL2-Grp.MLV.TATA - Hybrid LTR
containing Grp78 promoter from –524 to –60 fused at the MLV
TATA in U3 of LTR. (B) pGL2-Grp.TATA - Hybrid LTR
containing Grp78 promoter from –524 to –50, which includes its
wild type Grp78 TATA box, fused at the R of wild type MLV LTR
(C) pGL2-GrpL.TATA - Hybrid LTR containing Grp78 promoter
from –524 to +174 used in previous studies.

35
induction but due to higher basal level activity, it is inducible only by about 2-fold.

RCR.Grp.GFP is able to transduce various cell lines and induce weak expression of GFP
transgene upon hypoglycemic stress
pRCR.Grp.GFP with Grp78 promoter fused at the MLV TATA was constructed
(figure 2.4) and the viral vector RCR.Grp.GFP was produced by transient transfection in
293T. Viral supernatant collected was used to transduce MDA-MB 435 human breast
cancer cells at varying MOI to measure inducibility and to compare against RCR with
wild-type LTR (figure 2.5). Data show that the hybrid promoter is inducible by about 2-
fold as detected by FACS under hypoglycemic stress while wild type LTR had no
significant effect when subjected to hypoglycemic stress.
RCR.Grp.GFP was used to transduce various cell lines to test whether selectivity
and inducibility is consistent throughout various cell types - LNCaP human prostate
cancer cells, MDA-MB 435 human breast cancer cells, MDA-MB 231 human breast
cancer cells, A172 human glioma cells, U87 human glioma cells, HeLa, and J774.1A
murine macrophage cells were used. Results show that the inducibility depends on the
cell type with MDA-MB 435 being the most inducible cell line (figure 2.6). LNCaP cells
have lower GFP expressing cells when induced because most LNCaP cells died when
subjected to hypoglycemic stress.
36

01 23 45 67 8 9
Negative Control
pGL2-Grp.TATA
pGL2-GrpL.TATA
pGL2-
Grp.ML V.TATA
Relative Luciferase Units
Figure 2.3: Full length Grp78 promoter is most inducible with
lowest background level. Luciferase reporter constructs with hybrid
LTRs were transfected into 293T cells and assayed for inducibility
of transcription. Solid black bars represent induced values and
striped bars represent basal level activity without induction.
Arbitrary luciferase units were normalized for transfection
efficiency. Results are the means of three independent experiments
and thin extending lines show standard deviations.

37
CMV R U5 gag pol env Grp 78 R U5 IRES GFP
ψ
A
R U5 gag pol env Grp 78 R U5 IRES GFP
ψ
Grp 78 B
Insert
CMV R U5 gag pol env Grp 78 R U5 IRES GFP
ψ
A
R U5 gag pol env Grp 78 R U5 IRES GFP
ψ
Grp 78 B
Insert
Figure 2.4: Schematic diagram of pRCR.Grp.GFP construct.
(A) shows pRCR.Grp.GFP construct with CMV driving the
initial transcription of viral genes gag, pol, and env and GFP
transgene. (B) shows the viral genome in its DNA form, called
provirus, after reverse transcription. 3’ Grp78 promoter has
been duplicated and replaced CMV promoter at the 5’ LTR.

38
0
20
40
60
80
100
10 1 0.1
Multiplicity of Infection (MOI)
% GFP Positive
Figure 2.5: RCR.Grp.GFP is able to transduce MDA-MB 435 human
breast cancer cells and selectively induce transgene expression (green
fluorescent protein) under hypoglycemic stress. MDA-MB 435 cells
were transduced at three different multiplicity of infection (MOI) and
either kept in normal conditions or subjected to hypoglycemic stress
for twenty four hours. Value shows GFP positive cells detected by
flow cytometry. Solid black shows RCR.U3.GFP transduced cells
under normal conditions and gray bar shows RCR.U3.GFP
transduced cells under hypoglycemic stress. Striped bar shows
RCR.Grp.GFP transduced cells under normal conditions and white
bar shows RCR.Grp.GFP transduced cells under hypoglycemic stress.
Results are the means of three independent experiments and
extending lines are standard deviations.

39
0
50
100
150
200
250
LnCaP MDA-
MB 435
MDA-
MB 231
A172 U87 HeLa J774.1A
Cell Line
Relative Units
Figure 2.6: RCR.Grp.GFP is able to transduce various cell lines and
selectively induce transcription of transgene under hypoglycemic
stress. Cells were transduced with RCR.Grp.GFP at MOI of at least
1. Black bar indicates normal conditions and striped bar indicates
hypoglycemia for twenty-four hours. Values were normalized by
assigning a value of “100” to cells kept in normal conditions after
transduction. Results are means of three independent experiments
and thin extending bars represent standard deviations.

40
Due to very weak expression of the GFP marker (figure 2.7) western blot was
performed to see whether endogenous Grp78 protein levels correlate with promoter
inducibility but the band intensity was not strong enough to quantitate or differentiate one
to two-fold induction (data not shown).

The ability of RCR.Grp.GFP to replicate selectively under hypoglycemic condition could
not be validated in vitro assay
In order to demonstrate selective replication of RCR.Grp.GFP under
hypoglycemic conditions in vitro, MDA-MB 435 cells were transduced with
RCR.Grp.GFP at MOI of 1. Twelve hours post transduction media was changed to
glucose free media to induce the Grp78 promoter driving the expression of viral genes
(figure 2.8). However, supernatant collected from this experiment had undetectable level
of RCR.Grp.GFP (data not shown).
It is likely that the reason for the absence of RCR.Grp.GFP is the hypoglycemic
stress itself as hypoglycemia would slow, or completely shutdown, cellular metabolism
inhibiting vector production in the host cell and inhibit vector spread as cells would no
longer divide – a must-have prerequisite for MLV transduction. In order to test whether
the lack of propagation of RCR.Grp.GFP is due to reduced cellular metabolism and
inhibition of cellular division, adenovirus was used as they are able to transduce non-
dividing cells. Recombinant adenovirus was engineered with Grp78 promoter driving the
E1a early transcription gene and CMV promoter driving the GFP marker (figure 2.9). To
41

Figure 2.7: Fluorescence Microscopy of U87 cells transduced
with RCR.Grp.GFP shows minimal level of expression under
normal conditions without stress. Pictures were taken 48 hours
after transduction. (A) is untransduced U87 negative control
and (B) is RCR.U3.GFP transduced U87 positive control, and
(C) is RCR.Grp.GFP transduced U87 cells.

42
RCR.Grp.GFP
Transduction
Media with
Thapsigargon f or
Induction
Fresh Normal Media
Fresh Normal Media
Normal Media
1mL Supernatant
1mL Supernatant
10
5
Cells
10
5
Cells
AB C D
Induction and FACS Analysis
E
RCR.Grp.GFP
Transduction
Media with
Thapsigargon f or
Induction
Fresh Normal Media
Fresh Normal Media
Normal Media
1mL Supernatant
1mL Supernatant
10
5
Cells
10
5
Cells
AB C D
Induction and FACS Analysis
E
Figure 2.8: Schematic diagram of RCR.Grp.GFP Replication Assay
(A) 10
6
Cells were transduced with RCR.Grp.GFP at MOI of 1. (B)
12 Hours post-transduction, one set of RCR.Grp.GFP transduced cells
were induced with thapsigargon. (C) 24 Hours post-induction, media
was changed to fresh media to allow virus production. (D) After 24
hours in normal media, 1mL of supernatant was used to transduce fresh
cells. (E) 12 Hours post-transduction, cells were induced for 24 hours
and harvested for flow cytometry.

43
ITR ITR
delta E1 delta E3
Homologous Recombination
Grp78 CMV HSVtk IRES E1a GFP
pRC AdTGrpTK
pRC AdEGrpTK
Figure not drawn to scale
ITR ITR
delta E1 delta E3
Homologous Recombination
Grp78 CMV HSVtk IRES E1a GFP
pRC AdTGrpTK
Grp78 CMV HSVtk IRES E1a GFP Grp78 CMV HSVtk IRES E1a GFP
pRC AdTGrpTK
pRC AdEGrpTK
Figure not drawn to scale
Figure 2.9: Schematic diagram of pRCAdEGrpTK. Herpes simplex
virus thymidine kinase and adenovirus early gene 1a (E1a) driven by
Grp78 Promoter was cloned into pAdTrack opposite of GFP expression
cassette. Positive clones were cloned into pAdEasy vector via
homologous recombination as shown.

44
ensure selective expression of E1a, western blot was performed after hypoglycemic
induction of MDA-MB 435 cells transduced with the recombinant Grp78 adenovirus at
MOI of 0.1 and 1. Results demonstrated selective induction of E1a expression under
hypoglycemic stress (figure 2.10).
Protocol to collect adenovirus after hypoglycemic induction was optimized where
cells under hypoglycemic stress after viral transduction was replenished with regular
media for 48 hoursto help produce viral progeny (figure 2.11) (similar protocol tried for
RCR but did not work). 48-hours after replenishing with fresh media, adenovirus vectors
were collected by freeze-thaw method and these vectors were used to transduce fresh
MDA-MB 435 for FACS analysis. Results demonstrate that the Grp78 promoter is
inducible by 10-fold in adenovirus system suggesting that the failure to produce RCR
under hypoglycemic stress is due to inability of RCR to produce progeny under cellular
stress and inability of RCR to transduce non-dividing cells (figure 2.12).

45

12 3 4 56
E1a
Actin
12 3 4 56
E1a
Actin
12 3 4 56 12 3 4 56
E1a
Actin
Figure 2.10: Grp78 promoter in adenovirus RCAdGrpTK selectively
expresses E1a when induced by hypoglycemia. MDA-MB 435 cells were
transduced at MOI of 0.1 (lanes 3 and 4) or MOI of 1 (lanes 5 and 6).
Lanes 3 and 5 show samples kept in normal uninduced condition and lanes
4 and 6 show samples subjected to hypoglycemic stress. Results show
increase in E1a expression at higher MOI and upon hypoglycemic stress .
Lane 1 shows sample kept in normal condition without infection and lane 2
shows sample subjected to hypoglycemic stress without transduction.

46
Freez e &
Thaw 3X
Flow
Cytometry
RCAdGrpTK
Transduction
24 Hours 24 Hours
48 Hours
24 Hours
Glucose
Deprivation
Fresh Normal
Media
RCAdGrpTK
Transduction
Freez e &
Thaw 3X
Flow
Cytometry
RCAdGrpTK
Transduction
24 Hours 24 Hours
48 Hours
24 Hours
Glucose
Deprivation
Fresh Normal
Media
RCAdGrpTK
Transduction
Freez e &
Thaw 3X
Flow
Cytometry
RCAdGrpTK
Transduction
24 Hours 24 Hours
48 Hours
24 Hours
Glucose
Deprivation
Fresh Normal
Media
RCAdGrpTK
Transduction
Figure 2.11: Flow diagram of RCAdGrpTK selective replication
competency assay by hypoglycemic stress. Two 10 centimeter plates with 5
x 10
5
MDA-MB 435 human breast cancer cells were transduced with
RCAdGrpTK at MOI of 0.1 or 1. These samples were subjected to
hypoglycemic stress for twenty four hours to induce expression of E1a and
cultured in normal media for 48 hours after induction to allow viral gene
expression and assembly. Cells were harvested and subjected to three
freeze and thaw cycles to release virus and the supernatant was used to
transduce a fresh plate of 10
6
293A cells. These cells were collected 24
hours post transduction to quantitate GFP positive cells by flow cytometry.
Two 10 centimeter plates of MDA-MB 435 cells were transduced at MOI
of 0.1 or 1 and kept in normal media without induction for comparison.

47
A
BC
DE
A
BC
DE
Figure 2.12: FACS analysis shows enhanced replication and titer of
RCAdGrpTK when subjected to hypoglycemic stress. RCAdGrpTK
transduced MDA-MB 435 at MOI of 0.1 or 1 and subjected to normal
conditions or hypoglycemic stress. Supernatant from freeze and thaw cycles
were used to transduced a fresh plate of 293A cells. (A) 293A control, (B) is
from MOI of 0.1 and uninduced (0.03%) and (C) is from induced (0.15%),
and (D) is from MOI of 1 and uninduced (4.1%) and (E) is from induced
(40.1%). Titer of the vector was enhanced by about ten fold when
transduced cells were subjected to hypoglycemic stress.

48
RCR.Grp.GFP is able to propagate in U87 cells, transduced at varying MOI, under
normal conditions without hypoglycemic stress
In order to observe propagation of RCR.Grp.GFP under normal condition, U87
human glioma cells were initially chosen as they are easily transduced by MLV and show
little inducibility (figure 2.6). We were interested in identifying mutant species that are
able to spread under normal conditions. U87 cells were transduced with RCR.Grp.GFP
at MOI of 5, 0.5, and 0.05 and propagated in long term culture without hypoglycemic
stress. We observed the spread of the vector under normal, non-hypoglycemic,
conditions and we did not identify mutants with strong replicative kinetics similar to the
wild type LTR but the vector was able to propagate throughout the culture as determined
by FACS analysis (figure 2.13). Cells were harvested every few days up to day 35 post-
transduction to isolate extrachromosomal DNA by Hirt prep in order to look for major
recombination events in the viral genome, specifically in the LTR, potentially allowing
selective advantage to propagate (figure 2.14). Results indicated that there were no major
recombination events detectable by gel electrophoresis that gave rise to selective
advantage to propagate. There were minor bands present (e.g., U87 cells transduced at
MOI of 0.5 collected on days 18 and 25) but they disappeared by day 35 suggesting
selective disadvantage to propagate due to additional length of the viral genome that did
not give additional benefit to replicate.
Extrachromosomal DNA could not be isolated on day 3 and 7 from cells
transduced at MOI of 0.05. This is most likely due to low copy number of the viral
genome due to low initial transduction.
49

0
20
40
60
80
100
120
0 5 10 15 20 25 30
Day
% GFP Positive
Figure 2.13: RCR.Grp.GFP is able to propagate in U87 culture under
normal conditions without induction. U87 cells were transduced with
RCR.Grp.GFP at varying multiplicity of infection (MOI) and cultured
under normal conditions over an extended period as indicated. MOI
of 5 is indicated by squares, MOI of 0.5 is indicated by diamonds, and
MOI of 0.05 is indicated by triangles.

50
MOI 5
MOI 0.5
MOI 0.05
C D3 D7 D12 D18 D25 D35 C D3 D7 D12 D18 D25 D35
C D3 D7 D12 D18 D25 D35 C D3 D7 D12 D18 D25 D35
C D3 D7 D12 D18 D25 D35 C D3 D7 D12 D18 D25 D35
Figure 2.14: Hybrid LTR with Grp78 promoter seems stable U87
and does not show any gross recombination event over an extended
period of time. U87 human glioma cells were transduced at MOI of
0.05, 0.5, or 5 with RCR.Grp.GFP as indicated. Data show PCR
amplification of hybrid LTRs from Hirt prep DNA collected on
days indicated with “D.” “C” is control PCR from pRCR.U3.GFP
plasmid, hence shorter by about 100 base pairs, and blank lane
indicates unsuccessful attempts to amplify the fragment. Double
bands on days 18 and 25 of culture transduced at MOI of 0.5 is
repeatable and upper band disappeared by day 35.

51
RCR.Grp.GFP is able to propagate in various cell lines under normal condition without
hypoglycemic stress
Further experiments were done to select for mutant species of RCR.Grp.GFP that
are able to propagate aggressively, similar to wild-type LTR, having lost their
transcriptional control. MDA-MB 435 human breast cancer cells, U87 human glioma
cells, and LNCaP human prostate cancer cells were transduced with RCR.Grp.GFP at
MOI of 0.1 and cultured under normal, non-hypoglycemic, media to minimize Grp78
promoter induced expression of GFP. We re confirmed the spread of the vector
throughout the culture by monitoring the spread of GFP (figure 2.15) by FACS analysis.
Consistent with basal level activity of the promoter shown by luciferase assays (figure
2.6), GFP expressing cells increased over time indicating spread of RCR.Grp.GFP in
normal, non-hypoglycemic, conditions.

Grp78 hybrid LTR is stable in various cell lines in culture without gross recombination
events analyzed by PCR and restriction digests
For further analysis of the Grp78 hybrid-LTR in extended propagation, MDA-MB
435 human breast cancer cells, U87 human glioma cells, and LNCaP human prostate
cancer cells were transduced with RCR.Grp.GFP at an MOI of at least 0.5 and
propagated for 45 days. Using GFP as a surrogate marker, vector spread was detected by
FACS and fresh untransduced cells were added at ratio of about 1:1 when GFP positive
cells reached over 90% of culture. Consistent with basal level activities of the hybrid
52

A
0
5
10
15
20
25
0 1020 3040 5
Day
% G FP Po sitiv e
0
Figure 2.15: RCR.Grp.GFP is able to propagate in culture
under normal conditions without induction. (A) MDA-MB
435, (B) U87, and (C) LnCaP cells were transduced in
triplicates with RCR.Grp.GFP and kept in normal condition
without stress. Squares are sample 1, diamonds are sample 2,
and triangles are sample 3.

53
Figure 2.15, Continued
Figure 2.15 B
0
20
40
60
80
100
120
0 1020 30 4050
Day
% G FP Po sitiv e
Figure 2.15C
0
20
40
60
80
100
120
0 1020 3040 5
Day
% G F P P o sitiv e
0

54
LTR (figure 2.6), RCR.Grp.GFP slowly but persistently propagated throughout the
culture (figure 2.15).
Partial cultures were collected on day 35 from all three cell lines and
extrachromosomal DNA was isolated by Hirt prep. As previously done, MLV LTR
specific primers flanking the Grp78 insert in LTR were used to PCR amplify the U3
region of the LTR (figure 2.16A). We demonstrate here that the amplified fragments
from each cell line are consistent with the PCR control from vector plasmid suggesting
no major gross recombination events during propagation of the vector. Similar to the
results in chapter one, a very faint minor band is present that is 50 to 100bp smaller in
size, which coincides with the length of an ERSE, suggesting that the deletion does not
allow selective advantage to grow. These PCR fragments from Hirt prep DNA were
further analyzed by restriction digests (figure 2.16B & C). Restriction enzymes Hae III
and Ava II were used since they recognize frequently occurring sequences of four and
five nucleotides, respectively. Consistent restriction digest patterns suggest again lack of
gross recombination events but do not rule out the possibility of point mutations that may
affect promoter selectivity and strength.
Minor band could not be isolated after multiple attempts due to low yield (data
not shown).
55

(+) Control Digested
(+) Control
U87 RCR.Grp.GFP
MDA-MB435 RCR.Grp.GFP
LnCap RCR.Grp.GFP
(+) Control Digested
(+) Control
U87 RCR.Grp.GFP
MDA-MB435 RCR.Grp.GFP
LnCap RCR.Grp.GFP
B
C
+ -UML
A
Figure 2.16: Grp78 hybrid LTR seems stable without any gross
recombination event in various cell lines over an extended period
of time. Cells were harvested on day 35 of culture post
RCR.Grp.GFP transduction and extrachromosomal DNA was
collected using Hirt prep protocol. Grp78 hybrid LTR was
amplified by LTR specific promoters. (A) “-” is negative control,
“+” is positive control with pRCR.Grp.GFP, “U” is U87 human
glioma cell line, “M” is MDA-MB 435 human breast cancer cell
line, and “L” is LNCaP human prostate cancer cell line.
Amplified PCR products were digested with (B) Ava II or (C)
Hae III, which recognizes very common sequences, to detect
small recombination events.

56
Grp78 hybrid LTRs isolated from long term propagation consistently shows similar basal
level activity
Grp78 hybrid LTR is constitutively induced at low levels strong after 35 days of
propagation in various cell lines. PCR amplified hybrid LTRs from extended
propagation of MLV in various cell lines were inserted into pGL2 luciferase reporter
plasmid upstream of the luciferase gene. Positive clones with LTR inserts were identified
by PCR with primers flanking the insert and these clones were transfected into 293T
cells. pCH-110 β-galactosidase reporter plasmid was transfected in parallel with pGL2
constructs to normalize for transfection efficiency. 36 hours post-transfection, cells were
harvested to measure luciferase activity, which would indicate the strength of the LTR
promoter after 35 days of propagation in various cells lines (figure 2.17). Eight of nine
clones, pGL2-Grp 1 thru pGL2-Grp 8, tested show consistent levels of luciferase activity
compared with the positive control in which the hybrid LTR was PCR amplified from
pRCR.Grp.GFP plasmid. However, isolate from LnCAP, pGL2-Grp 9, showed greater
basal level activity by about 30-fold. Results indicate that the Grp78 hybrid LTR
promoter is relatively stable in extended culture but it’s possible that accrued mutations
in the promoter region over time can drive constitutive expression of downstream genes
losing its intended function.
PCR amplified fragments of Grp78 hybrid LTRs from extended cultures were
sequenced to identify regions(s) of instability. Sequence results of the first 8 samples
show point mutations that seemed random and did not affect the promoter strength
significantly (figures 2.18, table 2). However, pGL2-Grp 9, the one with 30-fold higher
basal level activity than the control, had 21 base-pair insert in the beginning of the hybrid
57

1
10
100
1000
10000
100000
1000000
10000000
pGL2-Basic
pGL2-Grp C
pGL2-Grp 1
pGL2-Grp 2
pGL2-Grp 3
pGL2-Grp 4
pGL2-Grp 5
pGL2-Grp 6
pGL2-Grp 7
pGL2-Grp 8
pGL2-Grp 9
Relative Luciferase Units
Figure 2.17: Grp78 hybrid LTR seems less stable than wild type
LTR and shows leaky expression. LTRs of RCR.U3.GFP were
amplified from Hirt prep DNA samples prepared on day 35 of vector
propagation in various cell lines and these were cloned into pGL-2
luciferase reporter construct. These constructs were transfected into
293T cells with -gal plasmids to normalize for transfection
efficiency. There were no correlation between the number of mutants
and cell type or the type of mutants and cell type.

58
-524 +174 -60
ERSE
TATA
-131 -163 -98
ERSE ERSE
-200 -300 -400 -500
-524 +174 -60
ERSE
TATA
-131 -163 -98
ERSE ERSE
-200 -300 -400 -500
Figure 2.18: Schematic diagram of mutations found in glucose-
regulated protein 78 (Grp78) promoter in long-term cultures.
Solid lines within the promoter represent mutation(s). Shaded
area starting at –60 was not included in the hybrid vector.

59

Table 2: Mutations found in Grp hybrid-LTR isolated from long term cultures in various
cells lines

Clone Cell Type Mutation(s)
pGL2-Grp 1 MDA-MB 435 • deletion (139 & 140)
• 317 C insertion
• 481 T insertion
pGL2-Grp 2 MDA-MB 435 • 142 A insertion
pGL2-Grp 3 MDA-MB 435 • 196 T and C insertion
pGL2-Grp 4 MDA-MB 435 • 127 A deletion
• 347 (G to A)
• 484 T insertion
pGL2-Grp 5 U87 • 110 (A to G)
• 111 (A to G)
• 112 (G to A)
pGL2-Grp 6 LNCaP • 33 T deletion
• 183 A insertion
• 185 A insertion
• 186 A insertion
pGL2-Grp 7 LNCaP
pGL2-Grp 8 LNCaP • 115 A deletion
• 122 T deletion
pGL2-Grp 9 LNCaP • Insertion (See Diagram)

60

LTR followed by series of point mutations but not interrupting any ERSE sequences
(figure 2.19). This 21 base pair sequence was searched using PUBMED’s BLAST but
no match was found suggesting random mutation events during reverse transcription
process.

DISCUSSION
Mutations such as point mutations and recombination events resulting in
duplication or deletion of repeat sequences in transcription regulatory elements of a
promoter may lead to aberrant expression pattern from the wild-type promoter [Temin,
1981; Temin, 1993]. The Grp78 promoter also has several transcription regulatory
elements such as the three repeats of ERSE that affect the wild type expression pattern
when mutated [Chang et al., 1989; Chao and Lin-Chao, 1992]. Hence, it’s reasonable to
assume that mutations in Grp78 hybrid LTRs accrued from long-term culture, similar to
long term persistent infection, can lead to mutant species with aberrant expression pattern
losing its intended function.
A comparison of promoter sequences of Grp78 across several species has lead to
discovery of a conserved domain of 28-base pairs. Footprinting analysis has shown
several protein factors interacting with this domain and these interactions were critical for
high level expression of Grp78 protein upon induction [Chang et al., 1989; Chao and Lin-
Chao, 1992]. Error-prone reverse transcription may lead to mutation in this domain and
lead to aberrant and undesirable expression pattern.
As in MLV LTR, there are two direct repeats of palindromic sequences in Grp78
promoter. Having two repeats may be responsible for high basal expression of Grp78 as
61

R U5 Grp78 R U5 Grp78 pGL2-Grp 9
GAGGCCGG - -- --- -- -- --- - -- --- - - - CTACG - GTCGGGA - GC- GCGTAC
GAGGCCGGGCAGCTGAAGACATGATGAATCTAGGAGAAGAAAGGCAGCGTAC
| | | | | | | |
| | | | | | | | | | | | | | |
41 33 40
GAGGCCGG - -- --- -- -- --- - -- --- - - - CTACG - GTCGGGA - GC- GCGTAC
GAGGCCGGGCAGCTGAAGACATGATGAATCTAGGAGAAGAAAGGCAGCGTAC
| | | | | | | |
| | | | | | | | | | | | | | |
GAGGCCGG - -- --- -- -- --- - -- --- - - - CTACG - GTCGGGA - GC- GCGTAC
GAGGCCGGGCAGCTGAAGACATGATGAATCTAGGAGAAGAAAGGCAGCGTAC
| | | | | | | |
| | | | | | | | | | | | | | |
41 33 40
Figure 2.19: Schematic diagram of insertion found in pGL2-
Grp9. Inserted sequence in grey font.

62
there’s only one such sequence in Grp94 promoter, a related protein, and has lower basal
level activity [Reddy et al., 2002]. Such repeats make Grp78 promoter in retrovirus more
susceptible to mutations during reverse transcription and via homologous recombination
as it is in the MLV LTR.
Also, there are three endoplasmic reticulum stress elements (ERSE) in Grp78
promoter. ERSE sequence is CCATT(N9)CCACG where N9 is GC rich [Parker et al.,
2001]. There are two GGC motifs within N9 in ERSE and it is required for maximal
ERSE-mediated inducibility. Some common transcription factors like NF-Y, ATF6, and
TFII-I bind to ERSE [Chang et al., 1989; Li et al., 2000; Parker et al., 2001]. Deleting
one GGC is not enough to render the promoter ineffective while deleting both GGC
motifs within the same ERSE renders the ERSE ineffective. Specific mutation of the
GGC motif resulted in substantial loss of stress inducibility mediated by the ERSE.
However, transcription factor binding to the ERSE alone is not sufficient to activate the
ER stress response, since full stress inducibility of ERSE requires integrity of the
CCAAT, GGC, and CCACG sequence motifs, as well as precise spacing among these
sites [Chang et al., 1989; Li et al., 2000; Parker et al., 2001].
To demonstrate mutations in Grp78 promoter resulting in aberrant expression
pattern in long-term culture, we constructed a replication competent retrovirus with
Grp78 promoter, RCR.Grp.GFP, driving replication of the vector to test the long-term
genomic stability of the hybrid Grp78-LTR promoter. Consistent with our luciferase
assays demonstrating relatively high basal level activity of Grp78 promoter,
RCR.Grp.GFP was able to spread throughout the cultures of different cell types under
normal, non-hypoglycemic conditions. PCR analysis suggested that the structure of the
63

hybrid LTR is stable throughout the extended culture and sequencing results showed that
point mutations acquired during the extended culture did not increase the level of basal
level expression in normal conditions. However, we discovered one mutant hybrid LTR,
sequenced from pGL2-Grp 9, with a 21 base pair insert that showed abnormally high
level of basal level activity demonstrating the risks in transcriptionally regulated RCR
gene therapy as such mutations would abolish the function of the hybrid LTR that it was
originally intended to do. It is unclear as to where the insert came from as the sequence
did not match any sequence of backbone plasmids nor viral sequences. A search to
determine the origin of the insert was done on Pubmed’s BLAST but it did not match
anything significant. It is possible that the insert could have been yielded in a series of
random misincorporations and extensions during reverse transcription, as the retroviral
reverse transcriptase has been known to make errors. It is possible that the insertion of
21 bases in the beginning of the Grp-U3 hybrid LTR altered the spacing of an important
element in the promoter. It has been suggested that there may be a negative regulator of
Grp78 promoter in the 5’ end of the promoter (unpublished data/verbal communication
with Amy S. Lee). Since the insertion is at the beginning of the promoter where a
putative negative regulator resides, it is possible that this element of the promoter has
been disrupted causing constitutive expression of the promoter as our data suggests.
64

CHAPTER 3
GENOMIC STABILITYOF PROBASIN-LTR HYBRID PROMOTER IN LONG TERM
CULTURE

INTRODUCTION
Probasin promoter is a strong and selective promoter activated in androgen
responsive cells making it suitable for androgen responsive prostate cancer gene therapy.
The probasin promoter from –426 to +28 in the 5’ untranslated region contains androgen
responsive elements and has been shown to stringently direct prostate-specific gene
expression in vitro [Greenberg et al., 1994] and in transgenic mice [Rennie et al., 1993],
particularly for targeted over-expression of SV40 T antigen, resulting in establishment of
transgenic models of prostate cancer (TRAMP mice) [Greenberg et al., 1995]. In this
chapter, we explore the genomic stability of probasin promoter hybrid-LTR in long-term
culture in various cancer cell lines.
More recently, a synthetic probasin promoter, ARR2PB, with tandem duplication
of the androgen responsive regions, has been demonstrated to confer a high level of
transgene expression specifically in the prostatic luminal epithelium and is strongly
regulated by androgens [Chen et al., 2000b; Lyubchenko et al., 2001]. The ARR2PB
promoter has been successfully used to drive androgen-dependent, prostate cell-specific
expression of transgene in vitro and in vivo [Solly et al., 2003], particularly in transgenic
mice [Huang et al., 2002; Kuriyama et al., 1999; Lyubchenko et al., 2001; Wang et al.,
2003b] as well as from adenoviral vectors [Andriani et al., 2001; Lowe et al., 2001;
65

Rubinchik et al., 2001].
Our lab has already tested a number of RCR vectors driven by both the wild-type
rat probasin (wt PB) promoter fragment used to generate TRAMP mice, or the synthetic
probasin promoter construct ARR2PB, incorporated into the U3 region of the viral LTR.
Similar strategies have been employed previously to target transgene transcription in
conventional replication-defective retrovirus vectors to specific cell types [Diaz et al.,
1998; Emiliusen et al., 2001; Indraccolo et al., 2000; Klein et al., 2006]. Through
previous work, our lab demonstrated proof-of-concept for application of this strategy to
RCR vectors, the first reported example of redirecting the tropism of MLV replication at
the transcriptional level [Logg et al., 2002].
In this chapter, we tested the genomic stability of probasin hybrid-LTR in long-
term propagation in various cancer cell lines. As the recombinant RCR with probasin
hybrid-LTR highly selective in replication, much more so than Grp78 promoter, we can
quickly identify mutant species that are non-selective in replication due to mutations
accrued from long-term propagation that would be indicative of potential risks in cancer
gene therapy with tissue specific or inducible RCR vectors.

MATERIALS AND METHODS
Nomenclature
A “p” at the beginning of construct names designates the plasmid vectors,
whereas names without a “p” denote viruses derived from the corresponding plasmid
vector. All sequence coordinates are given relative to the 5’ end of the 5’ long terminal
repeat (LTR) in the plasmid.
66

Tissue Culture
MDA-MB 435 human breast cancer cells, U87 human glioma cells, and 293T
human embryonic kidney cells were cultivated in DMEM 4.5g glucose/ml with 10% fetal
bovine serum (FBS). LNCaP human prostate cancer cells were cultivated in RPMI 1640
with 10% FBS.

Construction of Plasmids
All luciferase reporter constructs for this study were derived from pGL2
luciferase reporter construct (Promega). All retrovirus vector constructs for this study
were derived from pACE-GFP-dm [Logg et al., 2001a], which contains full-length
amphotropic RCR vector encoding GFP, and cytomegalovirus (CMV) promoter in place
of the 5’ U3 region.

Vector Production
Retrovirus vector stocks were produced by transfection of 293T cells with vector
plasmids using Fugene 6 (Roche). Supernatant containing vectors was collected at 48,
60, and 72 hours post transfection and filtered using 0.45-micron filters. Filtered
supernatant was stored in –80 degrees Celsius until used.

Retroviral Vector Titer Determination
For RCR.Pro.GFP, serial dilutions of viral supernatant in a total volume of 1mL
culture medium were added to 293A cells at 1 x 10^5 cells in six-well plates. Twelve
67

hours post-transduction, cells were incubated in medium with 50uM 3’-
azido-3’deoxythymidine (AZT) for 24 hours to prevent horizontal spread and subject to
FACS analysis to quantitate GFP fluorescence. The viral titer was calculated according
to the formula: transducing units (TU)/mL = (number of cells counted immediately
before infection x percentage of transduced cells reported from FACS analysis)/dilution
factor of viral supernatant.

Luciferase and β-Galactosidase Assay
293T cells at 70 to 90% confluency in 6-well dish were transfected with 2ug of
pGL2 constructs with mutant promoters driving luciferase expression using Fugene 6
(Roche). pGL2-LTR, which has long terminal repeat (LTR) of murine leukemia virus
(MLV), was transfected in parallel for positive control and pGL2-Basic which lacks a
promoter driving luciferase expression was transfected in parallel for negative control.
Each culture was co-transfected with pCH-110, which expresses β-galactosidase
(Promega) for transfection control. At 48 hours post-transfection, reporter lysis buffer
(Promega) was used to prepare cell extracts. Luciferase and β-galactosidase activities
were measured with luciferase assay system and β-galactosidase enzyme assay system
(Promega), respectively.

Polymerase Chain Reaction and Sequencing
For amplification of hybrid LTR to observe gross recombination, Platinum Taq
DNA Polymerase (Invitrogen) was used. 300ng of sample DNA in 50 ul reactions
containing 20mmol Tris-HCl (pH 8.4), 50mmol KCl, 0.2mM dNTP, 1.5mM MgCl2,
68

15pmol of primers, and 1 unit of Platinum Taq. The first step was to heat the polymerase
chain reaction (PCR) samples to 94 degrees Celsius for 1 minute and 30 seconds to
activate the enzyme. This was followed by 30 cycles of incubation at 94 degrees Celsius
for 30 seconds, 56 degrees Celsius for 45 seconds, and then 72 degrees for 1 minute. For
amplification of hybrid LTR to clone into pGL2 luciferase expression plasmid, Pfu Ultra
(Stratagene) was used. 200ng of Hirt prep DNA in 50uL reactions containing 1X Pfu
Ultra buffer, 0.2mM dNTP, 15pmol of each primer, and 2.5 units of Pfu Ultra. The first
step was to heat the PCR samples to 95 degrees Celsius for 2 minutes to activate the
enzyme. This was followed by 30 cycles of 95 degrees Celsius for 30 seconds to
denature the DNA, 56 degrees Celsius for 45 seconds for primer annealing to template,
and then 72 degrees for 1 minute 15 seconds for extension. For sequencing, 18-mer oligo
downstream of the multiple cloning site of pGL2 was used.

Hirt Prep
Cells were washed twice with 1X PBS and lysed in 1mL of Hirt lysis buffer
(0.6% SDS, 10mM EDTA pH 7.5). After 15 minutes in room temperature, 250uL of 5M
NaCl was added to the lysate and kept in 4 degrees Celsius overnight. Precipitated lysate
was spun at greater than 20,000 G and RNase was added to the supernatant and incubated
at 37 degrees Celsius for an hour. Resulting solution was phenol-extracted and DNA was
recovered by isopropenol precipitation.

69

RESULTS
RCR.Pro.GFP selectively replicates in androgen responsive prostate cancer cells and not
in other cell type
MDA-MB 435 human breast cancer cells and LNCaP human prostate cancer cells
were transduced with RCR.Pro.GFP at MOI of 0.5. Transduced cells were collected
every few days and measured for GFP by FACS analysis. At MOI of 0.1, RCR.Pro.GFP
spread throughout the entire LNCaP culture within three weeks, similar replicative
kinetics as the wild-type LTR after reaching an initial threshold, while RCR.Pro.GFP was
unable to spread in MDA-MB 435 consistent with previous experiments (figure 3.1, 3.2)
[Logg et al., 2002].
U87 human glioma cell line was added to the previous two, MDA-MB 435 and
LNCaP, and these cells were transduced with RCR.Pro.GFP in triplets and propagated in
long-term culture. These cells were propagated for 45 days and collected every few days
to isolate extrachromosomal DNA to identify mutant species, specifically in the hybrid
LTR, that are able to replicate and spread in non-androgen responsive cells indicating
loss of function of probasin hybrid-LTR due to mutations. Results demonstrate that
RCR.Pro.GFP is unable to propagate in non-androgen responsive cells but we found one
mutant with replication kinetics similar to wild type RCR two weeks after initial
transduction in U87 cells (figure 3.3b). This mutant, designated as Pmt (probasin
mutant), spread throughout the U87 culture within two weeks after detection.
Identification of this mutant demonstrate that even the most stringent transcriptionally
regulated RCR vector, such as the one used in our experiment, has the potential to lose its
intended function with mutations accrued over time. Also, one of three RCR.Pro.GFP
70

B
A
0
20
40
60
80
100
0 5 10 15 20 25 30
Days
% G F P P o s itiv e
0
20
40
60
80
100
0 5 10 15 20 25 30
Days
% G F P P o s itiv e
0
20
40
60
80
100
0 5 10 15 20 25 30
Days
% G F P P o s itiv e
0
20
40
60
80
100
0 5 10 15 20 25 30
Days
% G F P P o s itiv e
Figure 3.1: RCR.Prob.GFP is able to selectively replicate in prostate cancer
cells. RCR.Prob.GFP (dashed lines) shows similar replication kinetics relative
to RCR.U3.GFP (solid lines). (A) is LNCaP human prostate cancer cell line and
(B) is MDA-MB 435 is human breast cancer cell line.

71
C A B
Figure3.2: Fluorescence Microscopy of U87 cells transduced with
RCR.Prob.GFP shows undetectable level of expression in non-prostate cells.
Pictures were taken 48 hours after transduction. (A) is untransduced U87
negative control and (B) is RCR.U3.GFP transduced U87 positive control,
and (C) is RCR.Prob.GFP transduced U87 cells.

72
A
0
5
10
15
20
25
0 10 2030 405
Day
% GFP Po sitiv e
0
Figure 3.3: RCR.Prob.GFP is unable to propagate efficiently in
cell types other than prostate cell line. Squares are sample 1,
diamonds are sample 2, and triangles are sample 3.
RCR.Prob.GFP is unable to propagate efficiently in (A) MDA-
MB 435 but a mutant, sample 3, is able to propagate efficiently in
(B) U87 (indicated with triangles). RCR.Prob.GFP propagates
efficiently in LNCaP prostate cancer cell line as expected but
sample 3, lost the expression of its transgene GFP.

73
Figure 3.3, Continued
Figure 3.3 B
0
20
40
60
80
100
120
0 102030 405
Day
% GFP Positive
0
0
20
40
60
80
100
120
0 1020 3040 50
Day
% GFP Positive
Figure 3.3 C

74
transduced LNCaP cultures showed decreasing population of GFP expressing cells
detected by FACS analysis indicating deletion of GFP expression cassette not rarely
observed in long-term cultures [Logg et al., 2001a]. This demonstrates that the risk of
losing the therapeutic transgene, such as cytosine deaminase, is real must be carefully
considered in cancer gene therapy with RCR vectors.

Probasin hybrid LTR is not stable in various cell lines in long-term culture as analyzed by
PCR and restriction digests
For further analysis of the probasin hybrid-LTR in extended propagation, partial
cultures of MDA-MB 435, U87, and LNCaP cells transduced with RCR.Pro.GFP were
harvested on day 35 post initial transduction and extrachromosomal DNA was isolated by
Hirt prep protocol. As previously done, MLV LTR specific primers flanking the
probasin insert in LTR were used to PCR amplify the U3 region of the LTR (figure 3.4
A). Results show that the amplified fragments from each cell line are not consistent with
the PCR control from vector plasmid suggesting deletions, additions, and/or
recombination events during the propagation of RCR.Pro.GFP in different cell lines.
These PCR fragments from Hirt prep DNA were further analyzed by restriction digests
(figure 3.4 B & C). Restriction enzymes Hae III and Ava II were used since they
recognize frequently occurring sequences of four and five nucleotides, respectively.
Consistent with the different sizes of the PCR amplified LTRs, restriction digest patterns
were also different suggesting mutation events. This demonstrates that the real risk of
losing intended transcriptional control in transcriptionally targeted RCR gene therapy
(discussed in further detail in Chapter 5).
75

U
Pmt
ML + -
A
(+) Control
(+) Control Digested
U87 RCR.Prob.GFP
MDA-MB435 RCR.Prob.GFP
LnCap RCR.Prob.GFP
U87 RCR.Prob.GFP Mt
(+) Control
(+) Control Digested
U87 RCR.Prob.GFP
MDA-MB435 RCR.Prob.GFP
LnCap RCR.Prob.GFP
U87 RCR.Prob.GFP Mt
(+) Control
(+) Control Digested
U87 RCR.Prob.GFP
MDA-MB435 RCR.Prob.GFP
LnCap RCR.Prob.GFP
U87 RCR.Prob.GFP Mt
B
C
Figure 3.4: Probasin hybrid LTR seems less stable than wild type LTR
with gross recombination events over an extended propagation period.
Cells were harvested on day 35 of culture post RCR.Prob.GFP
transduction and extrachromosomal DNA was collected using Hirt prep
protocol. Probasin hybrid LTR was amplified by LTR specific
promoters. (A) “-” is negative control, “+” is positive control with
pRCR.Pro.GFP, “U” is U87 human glioma cell line, “Pmt” is mutant
shown in Figure 3.3 (B), “M” is MDA-MB 435 human breast cancer cell
line, and “L” is LNCaP human prostate cancer cell line. Amplified PCR
products were digested with (B) Ava II or (C) Hae III, which recognizes
very common sequences, to detect small recombination events. Mutant
“Pmt” was harvested and analyzed separately. Ava II and Hae III shows
gross recombination events (“U87 RCR.Prob.GFP Mt”).

76
Probasin hybrid LTR is stably selective in extended propagation in various cell lines but
its enhancer sequences are often deleted
Probasin hybrid LTR is a constitutively strong promoter in androgen responsive
prostate cancer cells and inactive in other cell types, and we demonstrate here that the
promoter is stably selective after 35 days of propagation in various cell lines. PCR
amplified probasin hybrid LTRs from extended propagation of RCR.Pro.GFP in various
cell lines were inserted into pGL2 luciferase reporter plasmid upstream of the luciferase
gene. Positive clones with probasin hybrid LTR inserts were identified by PCR with
primers flanking the insert and these clones were transfected into 293T cells. pCH-110
β-galactosidase reporter plasmid was transfected in parallel with pGL2 constructs to
normalize for transfection efficiency. 36 hours post-transfection, cells were harvested to
measure luciferase activity, which would indicate the selectivity of the probasin hybrid
LTR in non-androgen responsive 293T cells after 35 days of propagation (figure 3.5).
Seven of nine clones tested, pGL2-Pro 1 thru pGL2-Pro 7 show basal level luciferase
activity that is comparable or below probasin positive control in which the probasin
hybrid LTR was PCR amplified from pRCR.Pro.GFP plasmid. These results indicate
that the probasin hybrid LTR is stably selective in extended culture but it is possible that
the acquired mutations do not affect the selectivity or strength of the promoter in 293T.
However, construct pGL2-Pro 8 shows greater than 10-fold increase in luciferase activity
and construct pGL2-Pro 9 show greater than 100-fold increase demonstrating, again, the
real risks in losing transcription control over time.
77

1
10
100
1000
10000
100000
1000000
10000000
pGL2-Basic
pGL2-Pro C
pGL2-Pro 1
pGL2-Pro 2
pGL2-Pro 3
pGL2-Pro 4
pGL2-Pro 5
pGL2-Pro 6
pGL2-Pro 7
pGL2-Pro 8
pGL2-Pro 9
R elative Luciferase Units
Figure 3.5: Probasin hybrid LTR seems less stable than wild type
LTR. Probasin hybrid LTRs of RCR.Prob.GFP were amplified
from Hirt prep DNA samples prepared on day 35 of vector
propagation in various cell lines and these were cloned into pGL-2
luciferase reporter construct. These constructs were transfected
into 293T cells with -gal plasmids to normalize for transfection
efficiency. There were no correlation between the number of
mutants and cell type or the type of mutants and cell type.

78
For further analysis, PCR amplified fragments of probasin hybrid LTRs from
extended cultures were sequenced to identify regions(s) of instability. Sequence results
show several point mutations that seemed random and partial deletions of ARBS 1 and
ARBS2 were common. (figure 3.6, table 3). Interestingly, pGL2-Pro 8, the clone with
10-fold increase in basal level activity, only had two point mutations - deletion of C on
position 54 (position 1 as the start sequence of ARR2PB) and G to C mutation on
position 490 which is in the “R” region of the hybrid-LTR. PGL2-Pro 9, had a complete
deletion of ARBS1 but it also had 80% of cytomegalovirus (CMV) sequence present in
the 5’ LTR of pRCR.Pro.GFP just missing the first ~100 base pairs from the 5’ end
(figure 3.7). As vector should not have CMV sequence present after the initial
transfection during the vector production, CMV sequence must have been picked up by
homologous recombination during vector production process.

DISCUSSION
Mutations such as point mutations and recombination events resulting in
duplication or deletion of repeat sequences in transcription regulatory elements of a
promoter may lead to aberrant expression pattern from the wild-type promoter [Temin,
1981; Temin, 1993]. The latter is especially of concern since the probasin promoter has
two regions that are homologous in sequence and spaced by 64 base pairs, making them
vulnerable to recombination events leading to duplication or deletion. Zhang et al. have
shown that the
79

37 186
AR 1 AR 2
242 391
393 422 454 587
de letion de letion
37 37 186 186
AR 1 AR 2
242 242 391 391
393 422 454 587
de letion de letion
Figure 3.6: Schematic diagram of mutations found in hybrid
probasin LTR. pGL2-Pro 3 thru 6 from Figure 3.5 had mutations in
AR 1 to AR 2; pGL2-7 had a deletion from 422 to 454; pGL2-8 had
a single point mutation at position 587; pgL2-Pro 9 had a
recombination event with CMV promoter as shown in Figure 3.7.

80

Table 3: Mutations found in Probasin hybrid-LTR isolated from long-term cultures in
LNCaP human prostate cancer cells and U87 human glioma cells

Clone Cell Type Mutation(s)
pGL2-Pro 1 LNCaP • 40 (G to C)
• 393 (C to C)
pGL2-Pro 2 LNCaP • 167 (G to C)
pGL2-Pro 3 LNCaP • No mutation
pGL2-Pro 4 LNCaP • Deletion of ARBS 1 (37 to 242) and
partially ARBS2
pGL2-Pro 5 LNCaP • Deletion of ARBS 1 (37 to 242) and
partially ARBS2
pGL2-Pro 6 LNCaP • deletion (188 to 393) ARBS 2
pGL2-Pro 7 LNCaP • deletion (308 to 339) ARBS 2
pGL2-Pro 8 LNCaP • 54 C deletion
• 490 (G to C) in “R” region
pGL2-Pro 9 U87 • CMV sequence insertion (figure 3.7)
81

R U5 Prob CMV
pGL2-Pro 9
CMV
Region of
Homology
ARBS 1
Deletion
R U5 Prob CMV
pGL2-Pro 9
CMV
Region of
Homology
ARBS 1
Deletion
R U5 Prob CMV R U5 Prob CMV
pGL2-Pro 9
CMV
Region of
Homology
ARBS 1
Deletion
ARBS 1 Deletion
Probasin Promoter Sequence
pGL2-Pro 9 Sequence
31 243
GCTAGCTAT TATGATAGC
GCTAGC - - - TATGATAGC
5’ 3’
ARBS 1 Deletion
Probasin Promoter Sequence
pGL2-Pro 9 Sequence
31 243
GCTAGCTAT TATGATAGC
GCTAGC - - - TATGATAGC
5’ 3’
ARBS 1 Deletion
Probasin Promoter Sequence
pGL2-Pro 9 Sequence
ARBS 1 Deletion
Probasin Promoter Sequence
pGL2-Pro 9 Sequence
31 243
GCTAGCTAT TATGATAGC
GCTAGC - - - TATGATAGC
5’ 3’
31 243
GCTAGCTAT TATGATAGC
GCTAGC - - - TATGATAGC
5’ 3’
GCTAGCTAT TATGATAGC
GCTAGC - - - TATGATAGC
5’ 3’
Probasin Promoter Sequence
pGL2-Pro 9 Promoter Sequence
CMV Promoter Sequence
Region of Homology
CCAATCA - TCC - - - TGAAAGAGCTCAATAA
CCAATCA - TCCCGCCCATTGACGTCAATAA
CCAACGACCCCCGCCCATTGACGTCAATAA
| | | | | | | | | | | | | | | | | | | |
| | | | | | |
| | | |
| | | | | | | | | | | | |
|
441 466
102 131
5’ 3’
Probasin Promoter Sequence
pGL2-Pro 9 Promoter Sequence
CMV Promoter Sequence
Region of Homology
CCAATCA - TCC - - - TGAAAGAGCTCAATAA
CCAATCA - TCCCGCCCATTGACGTCAATAA
CCAACGACCCCCGCCCATTGACGTCAATAA
| | | | | | | | | | | | | | | | | | | |
| | | | | | |
| | | |
| | | | | | | | | | | | |
|
441 466
102 131
5’ 3’
Probasin Promoter Sequence
pGL2-Pro 9 Promoter Sequence
CMV Promoter Sequence
Region of Homology
Probasin Promoter Sequence
pGL2-Pro 9 Promoter Sequence
CMV Promoter Sequence
Probasin Promoter Sequence
pGL2-Pro 9 Promoter Sequence
CMV Promoter Sequence
Region of Homology
CCAATCA - TCC - - - TGAAAGAGCTCAATAA
CCAATCA - TCCCGCCCATTGACGTCAATAA
CCAACGACCCCCGCCCATTGACGTCAATAA
| | | | | | | | | | | | | | | | | | | |
| | | | | | |
| | | |
| | | | | | | | | | | | |
|
441 466
102 131
5’ 3’
CCAATCA - TCC - - - TGAAAGAGCTCAATAA
CCAATCA - TCCCGCCCATTGACGTCAATAA
CCAACGACCCCCGCCCATTGACGTCAATAA
| | | | | | | | | | | | | | | | | | | |
| | | | | | |
| | | |
| | | | | | | | | | | | |
|
441 466
102 131
5’ 3’
B
A
C
Figure 3.7: Schematic diagram of pGL2-Pro9 mutant. (A) shows the regions of
analysis. (B) shows ARBS1 deletion sequences in pGL2-Pro9 mutant. (C)
shows region of homology between wild type probasin, pGL2-Pro9, and wild
type CMV promoter sequences that may have been responsible for homologous
recombination during initial transfection for vector production.

82
frequency of recombination depends on the length of sequence identity where too short or
too long sequence identity decreases the frequency [Zhang and Temin, 1994]. The
frequency increased when the sequence identity increased from 20 to 60 base pairs but
decreased when the sequence identity increased from 60 to 830 base pairs. It is also
interesting to note that the R region of the MLV LTR, which is used to transfer the
minus-strand primer during reverse transcription, is also about 60 base pairs.
In addition to the vulnerability due to homologous regions in the probasin
promoter, converting RNA to DNA during reverse transcription requires the virus
enzyme reverse transcriptase to facilitate two template switches and these steps are also
error prone leading to mutant progeny. It has been suggested that the requirement for the
reverse transcriptase to facilitate these primer jumps may lead to the overall high error
rate [Pulsinelli and Temin, 1991; Temin, 1993].
Androgen specific regulation of the probasin promoter requires two androgen-
responsive regions containing two androgen receptor binding sites ARBS-1 and ARBS-2.
ARBS-1 resembles glucocorticoid response elements while ARBS-2 is unique and both
sites are necessary for maximal androgen induction as shown by transient transfection
studies. Neither binding site functions independently as greater than 90% loss in activity
was observed when point mutations were introduced in either ARBS-1 or ARBS-2
[Kasper and Matusik, 2000; Kasper et al., 1994]. Also, a single point mutation in ARBS-
1 not only blocked Androgen Receptor-2, AR-2, binding to ARBS-1 but also ARBS-2
demonstrating cooperative binding in a mutually dependent matter. Similarly, a point
mutation in ARBS-2 prevented AR-2 binding to both sites [Kasper and Matusik, 2000;
Kasper et al., 1994]. Also, deletion of 64 bases between ARBS-1 and ARBS-2 resulted
83

in 60% loss of androgen inducibility suggesting the region’s involvement in AR2 binding
to both sites. Since no preference of androgen receptor binding to either site was
observed, it’s likely that the region mediates receptor binding to both site and the order of
AR-2 binding is unimportant [Kasper and Matusik, 2000; Kasper et al., 1994].
In this study, we constructed a replication competent retrovirus with the probasin
promoter, RCR.Pro.GFP, driving the replication of the vector to test the long-term
genomic stability of the hybrid probasin-LTR promoter. Initial experiments show that
the promoter selectively replicates in androgen-responsive LNCaP human prostate cancer
cells and not in MDA-MB 435 human breast cancer cells and U87 human glioma cells as
shown previously by Logg et al [Logg et al., 2002]. However, PCR analysis of
extrachromosomal DNA isolated by Hirt prep protocol demonstrates that probasin hybrid
LTR is unstable in extended propagation in various cell types (figure 3.4). Sequencing of
the mutated hybrid promoter show that there are random point mutations as observed in
RCR.U3.GFP and RCR.Grp.GFP. Also, consistent with deletion of the first enhancer in
long-term culture of RCR.U3.GFP, ARBS-2 deletions were also isolated (figure 3.4 A).
More interestingly, species isolated from one of three U87 cultures transduced
with RCR.Pro.GFP, had greater than 80% of CMV sequence just missing the first 100
bases from the 5’ end joined at around 447 of probasin promoter. This probasin promoter
with CMV promoter joined at around 447 also had complete deletion of ARBS-1 (figure
3.7).
CMV promoter is used to drive the initial transcription of the retrovirus vector
plasmid after transfection [Logg et al., 2002; Logg et al., 2001a; Logg et al., 2001b] and
CMV itself should not be present in the transcript as the 3’ LTR contains the hybrid LTR
84

replacing U3 [Temin, 1995]. The presence of CMV promoter from vectors produced
from pRCR.Pro.GFP suggest that there must have been a recombination event during
transfection, the initial production stage, since no CMV sequence must be present in the
vectors produced – RCR.Pro.GFP. This raises safety issues in clinical gene therapy as
retrovirus vectors are most commonly produced by transfection. In our study, it took
about twenty days for the mutant progeny with CMV to spread 10% of the culture but it
spread throughout the entire plate in a week similar to type MLV replication kinetics.
This is probably due to the low rate of recombination event but one single recombination
event leading to uncontrolled mutant promoters is potentially dangerous. In order to
prevent such disastrous consequences in the clinic, each batch of viral vectors should be
checked for mutants in long-term culture. Also, having a stable cell line producing
vectors of known sequence would be advisable.
Also, one of three LNCaP cultures transduced with RCR.Pro.GFP had decreasing
number of GFP expressing cells. Though not characterized, it is likely that the mutant
had a deletion of GFP. Logg et al have shown that GFP deletion is common after
multiple passages [Logg et al., 2001a]. Since we added fresh uninfected cells at a ratio of
about 1:1 every three days after the culture initially reached greater than 90% GFP
positive cells, percentage of GFP positive cells would decline over time as if GFP had
been deleted out.
Loss of intended function of transcriptionally targeted RCR and loss of the
transgene are real risks that must be taken into consideration in cancer gene therapy using
RCR vectors. But with the suicide gene that could function as a self-destruction
mechanism if the transcription regulatory elements lose their intended function due to
85

mutations and their relatively non-threatening pathogenic mechanisms in human, MLV
derived RCR vectors are still a good candidate for cancer gene therapy (discussed in
further detail in Chapter 5). The use of these vectors should rest on the risk-benefit ratio
– therapeutic benefit from efficient gene delivery and persistence versus the potential
adverse effects such as uncontrolled spread of mutant species.
86

CHAPTER 4
LONG-TERM EFFECTS OF PROMOTER STABILITY AND
SELECTIVITY IN SUICIDE GENE THERAPY

INTRODUCTION
Suicide genes such as the herpes simplex virus thymidine kinase and yeast
cytosine deaminase have been incorporated into gene therapy vectors to enhance the
elimination of transduced tumor cells and have shown promising results [Tai and
Kasahara, 2008; Tai et al., 2005; Wang et al., 2003a; Wang and Weng, 2002; Wang et al.,
2004; Wang et al., 1998; Wang et al., 1999]. For example, cytosine deaminase gene
(CD) converts the non-toxic 5-fluorocytosine (5-FC) to the toxic nucleotide analog 5-
fluorouracil (5-FU) killing replicating cells (figure 4.1). Suicide genes function not only
in a therapeutic role to in enhance the elimination of transduced tumor cells but suicide
genes can also function as a safeguard to terminate “runaway” mutant viruses spread by
elimination of all infected cells in active division. However, suicide genes with its pro-
drug can kill non-dividing normal cells as well by bystander effects. For example, MLV
won’t be able to infect non-dividing normal cells, but nucleotide analogs that the suicide
gene converted to its active form can get to surrounding normal cells via gap junctions or
even simply by diffusion. In addition to incorporating into DNA during replication
causing chain termination, nucleotide analogs can also incorporate into mRNAs affecting
normal cells as well. Therefore, controlled expression of the suicide gene in target tissue
87

CMV R U5 gag pol env U3 R U5 IRES CD
psi
5-fluorocytosine
(non-toxic)
5-fluorouracil
(toxic)
CMV R U5 gag pol env U3 R U5 IRES CD
psi
5-fluorocytosine
(non-toxic)
5-fluorouracil
(toxic)
CMV R U5 gag pol env U3 R U5 IRES CD
psi
CMV R U5 gag pol env U3 R U5 IRES CD
psi
5-fluorocytosine
(non-toxic)
5-fluorouracil
(toxic)
Figure 4.1: Schematic diagram of pRCR.U3.CD. CD converts
the non-toxic pro-drug 5-FC to the toxic metabolite 5-FU

88
is important making transcriptionally targeted vectors more desirable to use with this
approach. However, as shown in previous chapters, mutations do occur and can have
undesirable effects on transgene expression. Also, as shown in Chapter 3 and by Logg et
al, transgene can be deleted out as well [Logg et al., 2001a]. In this chapter, we explored
the effects of promoter strength and selectivity on targeting specific tissues in long-term
culture.

MATERIALS AND METHODS
Nomenclature
A “p” at the beginning of construct names designates the plasmid vectors,
whereas names without a “p” denote viruses derived from the corresponding plasmid
vector. All sequence coordinates are given relative to the 5’ end of the 5’ long terminal
repeat (LTR) in the plasmid.
Tissue Culture
MDA-MB 435 human breast cancer cells, U87 human glioma cells, and 293T
human embryonic kidney cells were cultivated in DMEM 4.5g glucose/ml with 10% fetal
bovine serum (FBS). LNCaP human prostate cancer cells were cultivated in RPMI 1640
with 10% FBS.

5-FC and Ganciclovir (GCV) Cytotoxicity Assay
MDA-MB 435 human breast cancer cells, U87 human glioma cells, and LNCaP
human prostate cancer cells were seeded in 96-well plates in triplicates. 24 hours after
seeding, varying concentrations of 5-FC (0.1 mM, 1 mM, and 10 mM) or ganciclovir
89

(GCV) (1 μg/mL and 10 μg/mL) were added to determine ideal concentration to be used
with vectors with cytosine deaminase (CD) or herpes simplex virus thymidine kinase
(HSV-tk). Cell viability assay from Promega was used to test cell viability on days 3, 6,
and 9.

Construction of Plasmids
All retrovirus vector plasmids in this study were derived from either pACE-ept
(pRCR.Pro.GFP), pACE-GFP (pRCR.U3.GFP) [Logg et al., 2002; Logg et al., 2001a], or
pRCR.Grp.GFP. For pRCR.U3.TK, pRCR.Grp.TK, and pRCR.Pro.TK herpes simplex
virus thymidine kinase (HSV-tk) gene was fused with internal ribosome entry site (IRES)
at the first ATG start site by overlap extension PCR. Fo pRCR.U3.CD, pRCR.Grp.CD,
and pRCR.Pro.CD, yeast cytosine deaminase (CD) gene was fused with IRES at the first
ATG start site by overlap extension PCR.

Vector Production
Retrovirus vector stocks were produced by transfection of 293T cells with vector
plasmids using Fugene 6 (Roche). Supernatant containing vectors was collected at 48,
60, and 72 hours post transfection and filtered using 0.45-micron filters. Filtered
supernatant was stored in –80 degrees Celsius until used.

Genomic DNA Isolation
Genomic DNA was isolated from various cell lines using the Genomic Prep Cell
and Tissue DNA Isolation Kit (Amersham). Cells were harvested in microcentrifuge
90

tubes and spun at 1500g for 3 minutes. These cells were washed with cold PBS twice
and lysed in cell lysis solution. After incubating at 65 degrees Celsius for 1 hour, the
lysate was treated with RNase in room temperature for 30 minutes and protein
precipitation solution was added. After discarding the precipitates, isopropanol
precipitation was used to isolate genomic DNA. DNA pellet was washed with 70%
ethanol and dried pellet was eluted in DNA hydration solution and stored in 4 degrees
Celsius until used.

Polymerase Chain Reaction
For construction of pRCR.U3.TK, pRCR.Grp.CD, pRCR.Grp.TK, pRCR.Pro.CD,
and pRCR.Pro.TK, Pfu Ultra DNA polymerase (Stratagene) was used (pACE-CD (same
as pRCR.U3.CD) was given to us by Dr. Chris Logg). 100ng of template DNA with CD
or HSV-tk in 50uL reactions containing 1X Pfu Ultra buffer, 0.2mM dNTP, 15pmol of
each primer, and 2.5 units of Pfu Ultra. The first step was to heat the PCR samples to 95
degrees Celsius for 2 minutes to activate the enzyme. This was followed by 30 cycles of
95 degrees Celsius for 30 seconds to denature the DNA, 56 degrees Celsius for 45
seconds for primer annealing to template, and then 72 degrees for 1 minute 15 seconds
for extension.
91

RESULTS

At least up to 1mM 5-FC and 1ug/ml of GCV in MDA-MB 435, U87, and LNCaP cells
are non-cytotoxic
In order to determine the appropriate level of 5-FC and GCV for cytotoxicity
assay, MDA-MBA 435 human breast cancer cells, U87 human glioma cells, and LNCaP
human prostate cancer cells were subjected to concentrations of 0.1 mM, 1mM, or 10
mM of 5-FC, and 1ug/mL or 10ug/mL of GCV. Results show that the cells can tolerate
at least up to 1 mM concentration of 5-FC and 1 ug/mL concentration of GCV (figures
4.2, 4.3, 4.4).

RCR.U3.CD is able to transduce various cancer cell lines and efficiently eradicates them
with its pro-drug 5-FC
About 10
5
MDA-MB 435 human breast cancer cells, U87 human glioma cells,
and LNCaP human prostate cancer cells were transduced with 100mL filtered supernatant
from pRCR.U3.CD transfection and these cells were cultured for two weeks to make sure
the vector has completely spread throughout the culture (figure 2.3). These cells were
collected after two weeks to confirm integration of RCR.U3.CD genome. Genomic PCR
for cytosine deaminase gene confirmed transduction and integration of RCR.U3.CD into
the hosts’ genome (figure 4.5). In 96-well plates, these transduced cells were seeded in
triplicates either as 100% transduced cells or 50% transduced cells – one to one ratio of
transduced and untransduced cells. 1ug/mL GCV was added about 24 hours after seeding
(day 0) and was not replenished throughout the rest of the assay. Results show that
92

A
B
0
20
40
60
80
100
120
Day 3 Day 6 Day 9
% V ia b le
0
20
40
60
80
100
120
Day 3Day 6Day 9
% V ia b le
A
B
0
20
40
60
80
100
120
Day 3 Day 6 Day 9
% V ia b le
0
20
40
60
80
100
120
Day 3Day 6Day 9
% V ia b le
Figure 4.2: At least up to 1mM 5-FC or 1ug/ml of GCV in MDA-MB 435 is
non-cytotoxic. (A) 5-FC in MDA-MB 435 at various concentrations. Black
bar is control without 5-FC, bar with slanted dashes is 0.1mM of 5-FC, dotted
bar is 1mM of 5-FC and white bar is 10mM 5-FC. (B) GCV in MDA-MB435
at various concentrations. Black bar is control without GCV, bar with slanted
dashes is 1ug/ml GCV, and white bar is what 10ug/mL. 5-FC and GCV were
added to the media 24 hours after seeding and was not added again throughout
the rest of the assay.
93

AB
0
20
40
60
80
100
120
Day 3 Day 6 Day 9
% V ia b le
0
20
40
60
80
100
120
Day 3Day 6Day 9
% V ia b le
AB
0
20
40
60
80
100
120
Day 3 Day 6 Day 9
% V ia b le
0
20
40
60
80
100
120
Day 3Day 6Day 9
% V ia b le
Figure 4.3: At least up to 1mM 5-FC or 1ug/ml of GCV in U87 is non-
cytotoxic (A) 5-FC in MDA-MB 435 at various concentrations. Black bar
is control without 5-FC, bar with slanted dashes is 0.1mM of 5-FC, dotted
bar is 1mM of 5-FC and white bar is 10mM 5-FC. (B) GCV in MDA-
MB435 at various concentrations. Black bar is control without GCV, bar
with slanted dashes is 1ug/ml GCV, and white bar is what 10ug/mL. 5-FC
and GCV were added to the media 24 hours after seeding and was not added
again throughout the rest of the assay.

94
A
B
0
20
40
60
80
100
120
140
Day 3 Day 6 Day 9
% V i a b l e
0
20
40
60
80
100
120
Day 3Day 6Day 9
% V i a b l e
A
B
0
20
40
60
80
100
120
140
Day 3 Day 6 Day 9
% V i a b l e
0
20
40
60
80
100
120
Day 3Day 6Day 9
% V i a b l e
Figure 4.4: At least up to 1mM 5-FC or 1ug/ml of GCV in LNCaP is non-
cytotoxic (A) 5-FC in MDA-MB 435 at various concentrations. Black bar
is control without 5-FC, bar with slanted dashes is 0.1mM of 5-FC, dotted
bar is 1mM of 5-FC and white bar is 10mM 5-FC. (B) GCV in MDA-
MB435 at various concentrations. Black bar is control without GCV, bar
with slanted dashes is 1ug/ml GCV, and white bar is what 10ug/mL. 5-FC
and GCV were added to the media 24 hours after seeding and was not added
again throughout the rest of the assay.

95
CD
+ - ab c
β-Actin
- + ab
c
CD
+ - ab c
β-Actin
- + ab
c
Figure 4.5: RCR.U3.CD infects various cell lines as confirmed by
genomic PCR. Confirmation of RCR.U3.CD was done by preparing
genomic DNA of transduced cells and performing PCR for cytosine
deaminase (CD) transgene. Positive control (+) for CD PCR is
pRCR.U3.CD plasmid PCR, negative control (-) is no template PCR,
(a) is from U87, (b) is from MDA-MB 435, and (c) is from LNCaP.

96
RCR.U3.CD, with its pro-drug 5-FC, efficiently kills over 50% of all transduced cancer
cells within four days even when only 50% of the cells were initially transduced (figure
4.6). In all cancer cells, less than 10% of transduced cells with 5-FC were viable by day
10.
Same experiment was done using pRCR.U3.TK and similar results were obtained
(figure 4.7).

RCR.Grp.CD is able to transduce various cancer cell lines and efficiently eradicates them
under normal uninduced condition with its pro-drug 5-FC
pRCR.Grp..CD and pRCR.Grp.TK were constructed replacing the GFP
expression cassette in pRCR.Grp.GFP with cytosine deaminase (CD) or thymidine kinase
(TK) (figure 4.8). About 10
5
MDA-MB 435 human breast cancer cells, U87 human
glioma cells, or LNCaP human prostate cancer cells were transduced three times, every
12 hours, with 100mL filtered supernatant from pRCR.Grp.CD transfection to make sure
that nearly all cells are transduced, and these cells were cultured for two weeks to make
sure the vector has completely spread throughout the culture (figure 2.11). These cells
were collected after two weeks to confirm integration of RCR.Grp.CD genome.
Genomic PCR for cytosine deaminase gene confirmed transduction and integration of
RCR.Grp.CD into the hosts’ genome (figure 4.9). In 96-well plates, these transduced
cells were seeded in triplicates either as 100% transduced cells or 50% transduced cells –
one to one ratio of transduced and untransduced cells. 1 mM 5-FC was added about 24
hours after seeding (day 0) and was not replenished throughout the rest of the assay.
Results show that, under normal uninduced conditions, RCR.Grp.CD, with its pro-drug 5-
97

0
20
40
60
80
100
120
02 4 6 8 10 12
Day
% Viable
Figure 4.6: RCR.U3.CD with its pro-drug 5-FC efficiently kills
various human cancer cells (A) MDA-MB 435, (B) LNCaP, and (C)
U87 without selection. 1mM 5-FC was added only once 24 hours
after seeding cells and Day 0 is the day in which the 5-FC was added.
Solid line with rectangles is control without RCR.U3.CD infection
and 5-FC; solid line with triangles is only with 5-FC and no
RCR.U3.CD infection; solid lines with diamonds is only with 100%
RCR.U3.CD infection and no 5-FC; dashed line with circles is 100%
RCR.U3.CD infected cells with 5-FC; and dashed line with rectangles
is 50% infected cells with 5-FC.

98
Figure 4.6, Continued
Figure 4.6 B
0
20
40
60
80
100
120
024 68 10 12
Day
% Viable
0
20
40
60
80
100
120
02 468 10 12
Day
% Viable
Figure 4.6 C

99
A
0
20
40
60
80
100
120
02 4 68 10 12
Day
% Viable
Figure 4.7: RCR.U3.TK with its pro-drug GCV efficiently kills
various human cancer cells (A) MDA-MB 435, (B) LNCaP, and (C)
U87 without selection. 1ug/ml 5-FC was added only once 24 hours
after seeding cells and Day 0 is the day in which the GCV was
added. Solid line with rectangles is control without RCR.U3.TK
infection and GCV; solid line with triangles is only with GCV and
no RCR.U3.TK infection; solid lines with diamonds is only with
100% RCR.U3.TK infection and no GCV; dashed line with circles
is 100% RCR.U3.TK infected cells with GCV; and dashed line with
rectangles is 50% infected cells with GCV.

100
Figure 4.7, Continued
Figure 4.7 B
0
20
40
60
80
100
120
02 468 10 12
Day
% Viable
0
20
40
60
80
100
120
02 468 10 12
Day
% Viable
Figure 4.7 C

101
CMV R U5 gag pol env Grp78 R U5 IRES CD or TK
psi
R U5 gag pol env Grp78 R U5 IRES CD or TK
psi
Grp78
A
B
MLV TATA
MLV TATA
ML V TATA
CMV R U5 gag pol env Grp78 R U5 IRES CD or TK
psi
R U5 gag pol env Grp78 R U5 IRES CD or TK
psi
Grp78
A
B
MLV TATA
MLV TATA
ML V TATA
CMV R U5 gag pol env Grp78 R U5 IRES CD or TK
psi
R U5 gag pol env Grp78 R U5 IRES CD or TK
psi
Grp78
A
B
MLV TATA
MLV TATA
ML V TATA
Figure 4.8: Schematic diagram of pRCR.Grp.CD and pRCR.Grp.TK
constructs. (A) shows pRCR.Grp.CD/TK construct with CMV
driving the initial transcription of viral genes gag, pol, and env and
GFP transgene. (B) shows the viral genome in its DNA form, called
provirus, after reverse transcription. 3’ Grp78 promoter has been
duplicated and replaced CMV promoter at the 5’ LTR.

102
+ - ab c
- + ab c
CD
β-Actin
+ - ab c + - ab c
- + ab c - + ab c
CD
β-Actin
CD
β-Actin
Figure 4.9: RCR.Grp.CD infects various cell lines as confirmed by
genomic PCR. Confirmation of RCR.Grp.CD was done by preparing
genomic DNA of transduced cells and performing PCR for cytosine
deaminase (CD) transgene. Positive control (+) for CD PCR is
pRCR.Grp.CD plasmid PCR, negative control (-) is no template PCR, (a)
is from U87, (b) is from MDA-MB 435, and (c) is from LNCaP.

103
FC, efficiently kills over 50% of all transduced U87 human glioma and LNCaP human
prostate cancer cells within four days even when only 50% of the cells were initially
transduced (figure 4.10) consistent with high basal level activity shown in Chapter 2.
Also, less than 10% of transduced cells with 5-FC were viable by day 10 under
uninduced conditions. However, RCR.Grp.CD, with 5-FC, took about a week to reach
50% killing under normal uninduced conditions in MDA-MB 435 and did not reach
complete eradication as shown in U87 and LNCaP. Same experiment was done using
pRCR.Grp.TK and similar results were obtained (figure 4.11).

RCR.Grp.CD expresses detectable levels of its transgene cytosine deaminase under
normal uninduced conditions as detected by RT-PCR
About 10
5
MDA-MB 435 human breast cancer cells, U87 human glioma cells, or
LNCaP human prostate cancer cells were transduced three times, every 12 hours, with
100mL filtered supernatant from pRCR.Grp.CD transfection to make sure that nearly all
cells are transduced, and these cells were cultured for two weeks under normal uninduced
condition as done in the previous section. These cells were further cultured for two
weeks to emulate long-term culture and these cells were collected for RNA prep. RT-
PCR for cytosine deaminase was done to confirm expression of cytosine deaminase,
which would explain efficacious cell killing with 5-FC under normal conditions. Results
show that by four weeks post-transduction under normal non-hypoglycemic condition,
there are detectable levels of cytosine deaminase in all cells transduced with
RCR.Grp.CD (figure 4.12).
These experiments demonstrate that Grp78 promoter with leaky expression is not
104

0
20
40
60
80
100
120
02 46 8 10 12
Day
% Viable
Figure 4.10: RCR.Grp.CD with its pro-drug 5-FC partially kills
various human cancer cells (A) MDA-MB 435, (B) LNCaP, and (C)
U87 under non-hypoglycemic condition. 1mM 5-FC was added only
once 24 hours after seeding cells and Day 0 is the day in which the 5-
FC was added. Solid line with rectangles is control without
RCR.Grp3.CD infection and 5-FC; solid line with triangles is only
with 5-FC and no RCR.Grp.CD infection; solid lines with diamonds
is only with 100% RCR.Grp.CD infection and no 5-FC; dashed line
with circles is 100% RCR.Grp.CD infected cells with 5-FC; and
dashed line with rectangles is 50% infected cells with 5-FC.

105
Figure 4.10, Continued
Figure 4.10 B
0
20
40
60
80
100
120
02 4 68 10 12
Day
% Viable
0
20
40
60
80
100
120
02 4 68 10 12
Day
% Viable
Figure 4.10 C

106
0
20
40
60
80
100
120
140
02 468 10 12
Day
% Viable
Figure 4.11: RCR.Grp.TK with its pro-drug GCV partially kills
various human cancer cells (A) MDA-MB 435, (B) LNCaP, and
(C) U87 under normal non-hypoglycemic condition. 1ug/mL
GCV was added only once 24 hours after seeding cells and Day 0
is the day in which the GCV was added. Solid line with
rectangles is control without RCR.Grp.TK infection and GCV;
solid line with triangles is only with GCV and no RCR.Grp.TK
infection; solid lines with diamonds is only with 100%
RCR.Grp.TK infection and no GCV; dashed line with circles is
100% RCR.Grp.TK infected cells with GCV; and dashed line
with rectangles is 50% infected cells with GCV.

107
Figure 4.11, Continued
Figure 4.11 B
0
20
40
60
80
100
120
02 46 8 10 12
Day
% Viable
0
20
40
60
80
100
120
02 46 8 10 12
Day
% Viable
Figure 4.11 C

108

CD
β-Actin
+ - a b c d
CD
β-Actin
+ - a b c d
CD
β-Actin
+ - a b c d
Figure 4.12: RT-PCR nearly 4 weeks after the initial RCR.Grp.CD
infection confirms expression of cytosine deaminase transgene in
RCR.Grp.CD infected cells in long-term culture under normal non-
hypoglycemic conditions in various cancer cell lines. Positive PCR
control (+) for CD was pRCR.Grp.CD and no template was added
for negative control. (a) is from RCR.U3.CD infected MDA-MB
435 cells as RT-PCR positive control, (b) is from RCR.Grp.CD
infected MDA-MB 435, (c) is from RCR.Grp.CD infected U87, and
(d) is from RCR.Grp.CD infected LNCaP.
109

a good candidate for MLV based RCR gene therapy due to its high level of basal level
activity. Furthermore, additional mutations can potentially lead to further activation of
the promoter but it’s more likely that the cells transduced with the vector would be
eradicated in the early stages before the rise of mutant species due to high basal level
activity inhibiting vector spread

RCR.Pro.CD is able to transduce various cancer cell lines and shows partial cell killing
even in non-androgen-responsive cancer cells with its pro-drug 5-FC
pRCR.Pro..CD and pRCR.Pro.TK were constructed replacing the GFP expression
cassette in pRCR.Pro.GFP with cytosine deaminase (CD) or thymidine kinase (TK)
(figure 4.13). About 10
5
MDA-MB 435 human breast cancer cells, U87 human glioma
cells, or LNCaP human prostate cancer cells were transduced with 100mL filtered
supernatant from pRCR.Pro.CD transfection and these cells were cultured for two weeks
to make sure the vector has completely spread throughout LNCaP culture (figure 3.3).
These cells were collected after two weeks to confirm integration of RCR.Pro.CD
genome. Genomic PCR for cytosine deaminase gene confirmed transduction and
integration of RCR.Pro.CD into the hosts’ genome (figure 4.14). In 96-well plates, these
transduced cells were seeded in triplicates either as 100% transduced cells or 50%
transduced cells – one to one ratio of transduced and untransduced cells. 1mM 5-FC was
added about 24 hours after seeding (day 0) and was not replenished throughout the rest of
the assay. Results show that RCR.Pro.CD, with its pro-drug 5-FC, efficiently kills almost
all LNCaP culture within two weeks (figure 4.15). However, like RCR.Grp.CD,
RCR.Pro.CD is also able to partially kill non-androgen-responsive cells with 5-FC.
110

CMV R U5 gag pol env Prob R U5 IRES CD or TK
ψ
R U5 gag pol env Prob R U5 IRES CD or TK Prob
A
B
ML V TATA
ML V TATA
MLV TATA ψ
CMV R U5 gag pol env Prob R U5 IRES CD or TK
ψ
R U5 gag pol env Prob R U5 IRES CD or TK Prob
A
B
ML V TATA
ML V TATA
MLV TATA ψ
Figure 4.13: Schematic diagram of pRCR.Pro.CD and pRCR.Pro.TK
constructs. (A) shows pRCR.Pro.CD/TK construct with CMV driving
the initial transcription of viral genes gag, pol, and env and GFP
transgene. (B) shows the viral genome in its DNA form, called provirus,
after reverse transcription. 3’ probasin promoter has been duplicated and
replaced CMV promoter at the 5’ LTR.
111

+
- ab c
- + ab c
CD
β-Actin
+
- ab c
+
- ab c
- + ab c - + ab c
CD
β-Actin
CD
β-Actin
Figure 4.14: RCR.Pro.CD infects various cell lines as confirmed by
genomic PCR. Confirmation of RCR.Pro.CD was done by preparing
genomic DNA of transduced cells and performing PCR for cytosine
deaminase (CD) transgene. Positive control (+) for CD PCR is
pRCR.Pro.CD plasmid PCR, negative control (-) is no template PCR, (a)
is from U87, (b) is from MDA-MB 435, and (c) is from LNCaP.

112
0
20
40
60
80
100
120
02 4 6 8 10 12
Day
% Viable
Figure 4.15: RCR.Pro.CD with its pro-drug 5-FC partially kills human
cancer cells (A) MDA-MB 435 and (C) U87 and efficiently eradicates (B)
LNCaP. 1mM 5-FC was added only once 24 hours after seeding cells and
Day 0 is the day in which the 5-FC was added. Solid line with rectangles is
control without RCR.Pro.CD infection and 5-FC; solid line with triangles is
only with 5-FC and no RCR.Pro.CD infection; solid lines with diamonds is
only with 100% RCR.Pro.CD infection and no 5-FC; dashed line with circles
is 100% RCR.Pro.CD infected cells with 5-FC; and dashed line with
rectangles is 50% infected cells with 5-FC.

113
Figure 4.15, Continued
Figure 4.15 B
0
20
40
60
80
100
120
02 4 68 10 12
Day
% Viable
0
20
40
60
80
100
120
02 4 68 10 12
Day
% Viable
Figure 4.15 C

114
0
20
40
60
80
100
120
02 46 8 10 12
Day
% Viable
Figure 4.16: RCR.Pro.TK with its pro-drug GCV partially
kills human cancer cells (A) MDA-MB 435 and (C) U87 and
efficiently eradicates (B) LNCaP. 1ug/mL GCV was added
only once 24 hours after seeding cells and Day 0 is the day in
which the GCV was added. Solid line with rectangles is
control without RCR.Pro.TK infection and 5-FC; solid line
with triangles is only with GCV and no RCR.Pro.TK
infection; solid lines with diamonds is only with 100%
RCR.Pro.TK infection and no GCV; dashed line with circles
is 100% RCR.Pro.TK infected cells with GCV; and dashed
line with rectangles is 50% infected cells with GCV.

115
Figure 4.16, Continued
Figure 4.16 B
0
20
40
60
80
100
120
02 4 68 10 12
Day
% Viable
0
20
40
60
80
100
120
02 46 8 10 12
Day
% Viable
Figure 4.16 C

116
Same experiment was done using pRCR.Pro.TK and similar results were obtained
(figure 4.16).

RCR.Pro.CD expresses detectable levels of its transgene cytosine deaminase in long-term
culture even in non-androgen responsive cancer cells as detected by RT-PCR
About 10
5
MDA-MB 435 human breast cancer cells, U87 human glioma cells, or
LNCaP human prostate cancer cells were transduced three times, every 12 hours, with
100mL filtered supernatant from pRCR.Pro.CD transfection to make sure that nearly all
cells are transduced, and these cells were cultured for extended period of 4 weeks. Cells
were harvested and RNA was isolated for RT-PCR for cytosine deaminase to confirm
expression of cytosine deaminase, which would explain partial cell killing with 5-FC in
MDA-MB 435 and U87. Results show that by four weeks after the first infection, there
are detectable levels of cytosine deaminase transcript in all cells transduced with
RCR.Pro.CD (figure 4.17)

DISCUSSION
We demonstrate in this chapter that suicide gene therapy coupled with an
appropriate promoter for the target tissue is very potent (e.g. probasin promoter for
androgen-responsive LNCaP prostate cancer cells). As demonstrated by using the Grp78
promoter, promoters with high basal level activity in most tissues should be used with
caution since they can express enough levels of the suicide gene efficiently killing
infected cells as efficiently as vectors with wild type LTR. This does not rule out
117

CD
β-Actin
+ - a b c d
CD
β-Actin
+ - a b c d
CD
β-Actin
CD
β-Actin
+ - a b c d
Figure 4.17: RT-PCR nearly 4 weeks after the initial RCR.Pro.CD
infection confirms expression of cytosine deaminase transgene in
RCR.Pro.CD infected cells in non-androgen responsive cancer cells in
long-term culture. Positive PCR control (+) for CD was pRCR.Pro.CD
and no template was added for negative control. (a) is from RCR.U3.CD
infected MDA-MB 435 cells as RT-PCR positive control, (b) is from
RCR.Pro.CD infected MDA-MB 435, (c) is from RCR.Pro.CD infected
U87, and (d) is from RCR.Pro.CD infected LNCaP.

118
promoters like Grp78 to be used since its inducibility could potentially have a greater
impact on target tissues like hypoglycemic/stressed tumor core under in vivo conditions
(unsuccessful experiments conducted in mouse tumor models with RCR.Grp.GFP
resulting in undetectable level of transduction in tumor cells – data not shown).
It is worth noting that even with probasin promoter, some cell death was observed
in MDA-MB 435 as well as U87 cells. Our RT-PCR shows that there are detectable
levels of the suicide gene transcript in all cells infected with the vector, but our analysis
does not demonstrate the level of expression. In order to better quantify the level of
expression over time, quantitative PCR should be done. Also, it is possible, perhaps even
likely, that there could have been vectors with mutations allowing leaky or even
constitutive expression of the suicide gene since vector infected cells were cultured for
two weeks before starting suicide gene/pro-drug cytotoxicity assay. In the case of Pmt, it
took the mutant 2 weeks to infect 10% of the culture. If this were the case in our suicide
gene assay, a small percentage of the cells would have been infected in the beginning of
the suicide gene/pro-drug cytotoxicity assay and would have been eradicated during the
first couple of days before they had a chance to spread.
In order to reduce the risk of escape mutants at the level of transcription
regulatory elements or the transgene, pro-drug of the suicide gene should be added early.
However, this would have an impact on the efficiency of the virus to deliver the gene.
Chain termination by nucleotide analogs activated by the suicide gene delivered by the
vector would affect the process of viral DNA replication as well. The potential problem
is that virus replication and spread may be prematurely attenuated at an early stage prior
to actual cell death, and further attenuated by the existence of bystander cells that have
119

not yet been infected but have absorbed chain-terminating nucleotide analogs from
adjacent infected cells through gap junctions. Furthermore, some nucleotide analogs can
be incorporated into mRNAs as well as DNA. This may also function to terminate virus
spread at an early stage, by disabling viral gene products still being produced in the
infected cells prior to completion of virus replication.
120

CHAPTER 5
DISCUSSION & FUTURE DIRECTIONS

Advances in gene delivery technologies have made possible the treatment of
malignancies at the genetic level and numerous clinical trials for cancer gene therapy
have initiated over the past couple of decades. In particular, vectors derived from
retroviruses have been widely studied as vehicles for gene transfer into tumor cells due to
their well-characterized biology and their innate characteristics that make them more
favorable for cancer gene therapy. Simple retroviruses, such as murine leukemia virus
(MLV), require cell division for infection and thus possess inherent selectivity for rapidly
dividing cells characteristic of tumor tissues. This unique property, and the ease with
which retroviral vectors can be manipulated due to their relatively well-characterized
genetics, provided much of the impetus for their use in experimental and clinical cancer
gene therapy studies.
Conventional MLV-based vectors used for the experimental human gene therapy
have generally been rendered replication-defective to address safety concerns – key viral
genes are removed and replaced with therapeutic genes, and these handicapped vectors
are produced in cells that trans-complement the defective viral elements. After the first
round of infection, these replication defective vectors are incapable of secondary
horizontal infections of adjacent cells due to the deletion of one or multiple essential viral
genes, thereby enhancing safety [Sinn et al., 2005; Weber et al., 2001].
121

However, this safety mechanism greatly reduces the efficiency of gene delivery.
Although, the initial results of in vivo models using defective retroviral vectors to
deliver genes into tumors were encouraging [Culver et al., 1994; Culver et al., 1990;
Ram et al., 1994; Short et al., 1990], these vectors have generally proven to be
therapeutically inefficacious in clinical trials. With intra-tumoral injection of
replication-defective virus vectors, penetration into the tissue appears to be diffusion-
limited, resulting in effective gene delivery within a range of only a few cell diameters
surrounding the injection site. In one of the largest clinical studies of cancer gene
therapy to date, cells engineered to produce replication-defective retrovirus vectors
expressing the Herpes simplex thymidine kinase suicide gene (HSV-tk) were injected at
a large number of sites within brain tumors that had been subjected to prior surgical
debulking. Quantification of the gene transfer efficiency from the injection
demonstrated that no more than 0.002% of the tumor cells in any examined patient had
been transduced [Rainov et al., 2000]. This demonstrates the need for vectors with
substantially higher level of transduction efficiency for any clinical efficacy.
On the contrary, replication-competent recombinant retrovirus (RCR) vectors
have shown promising results in its transduction and gene delivery efficiency both in
vitro and, more recently, in vivo. As shown in this paper (figure 1.3) as well as by Logg
et al. [Logg et al., 2001b], we have developed RCR vectors based on murine leukemia
virus (MLV) that are capable of transducing tumor cells in vitro as well as solid tumor in
vivo with very high efficiency. For example, U87 cells were inoculated with RCR
vectors with GFP marker at MOI of 0.05 and the entire population of cells were
completely transduced and expressed GFP within 2 weeks showing similar replication
122

kinetics as the wild type MLV (Figure 3.3C). In vivo experiments by Logg et al.
demonstrated that after the injection of RCR vectors with GFP marker into established
subcutaneous tumors, highly efficient spread of GFP marker was observed over a period
of 7 weeks, in some cases resulting in spread of the marker throughout the entire tumor.
These experiments conclusively demonstrated the significant advantages over standard
replication-defective vectors in efficiency of delivery both in culture and in vivo. Further
strengthening the attractiveness of the RCR vectors, Tai et al. have demonstrated in an in
vivo model that RCR vectors equipped with suicide genes, such as the yeast cytosine
deaminase gene, are able to prolong the survival rate of lab animals with administration
of their pro-drugs, such as 5-fluorocytosine. Cytosine deaminase gene converts nontoxic
5-flurocytosine (5-FC) to the highly toxic metabolite 5-fluorouracil (5-FU), one of the
most active anti-neoplastic agents in conventional cancer chemotherapy [Kievit et al.,
1999].

BENEFITS AND POTENTIAL SAFETY CONSIDERATIONS
RCR vectors have clear advantages over conventional replication defective
vectors. First, RCR vectors have an absolute requirement for cell division for initial
transduction and subsequent propagation. It has been well characterized that the RCR
lacks nuclear localization signals and can only infect cells that are actively dividing
[Miller et al., 1990]. In fact, this was precisely the reason why retrovirus vectors were
first utilized for cancer gene therapy [Culver, 1992]. This absolute selectivity for dividing
cells is responsible for a significant level of tumor selectivity for MLV-based RCR vector
propagation in tumor tissue. In most cases, adjacent normal parenchymal tissues would
123

be post-mitotic reducing the risk of inadvertently transducing normal quiescent cells,
whereas the tumor and the surrounding neovasculature would be actively dividing
serving as a target for the MLV-based RCR vectors. For example, Tai et al have shown
by immunohistochemistry of an in vivo glioma model that MLV spread was confined to
the tumor cells without spread to surrounding normal brain tissue, with a sharp
demarcation between tumor and post-mitotic peritumoral brain tissues, suggesting that
viral replication is highly selective for the rapidly dividing glioma cells [Tai et al., 2005].
Secondly, MLV is an integrating virus and with non-lytic lifecycle that enables
transduced cells to become producer cell of vectors. This characteristic in combination
with its selectivity for actively dividing cells makes MLV-based RCR vector an efficient
gene delivery vehicle for tumors without causing direct damage to post-mitotic quiescent
normal cells as a consequence of viral replication.
Thirdly, its ability to go through non-lytic replication resulting in reduced
immunogenicity can allow MLV to achieve greater persistence in tumors and high levels
of intra-tumoral gene transfer. Although other replicating viruses in current use as
oncolytic agents such as recombinant adenoviruses have been reported to achieve
selectivity for tumor cells by taking advantage of intrinsic defects in cellular defense
mechanisms such as p53 mutations, it is well-documented that most of these viruses can
also infect and lyse non-dividnig normal cells, and provoke highly robust anti-viral
immune responses [Yoon et al., 2001]. Also, in the case of oncolytic adenoviruses, it is
well documented that even conventional replication-defective adenoviral vectors are
rapidly eliminated due to Class I responses against low level “leaky” expression of viral
products [Yang et al., 1996a; Yang et al., 1996b; Yang et al., 1996c]. On the other hand,
124

MLV’s necessity to integrate and its non-lytic replication cycle resulting in reduced
immunogenicity also make persistent infection possible which provide opportunities to
treat recurring tumors with administration of pro-drugs in MLV-based RCR vector
suicide gene therapy.
Lastly, the biology of MLV is simple consisting of two long terminal repeats
(LTR) flanking 5’ and 3’ end of its three genes – gag, pol, and env – and very well
characterized as described earlier in introduction. This well characterized biology
enables us to manipulate MLV and regulate its replication and expression of exogenous
transgene. As used in this study, we have developed various RCR vectors derived from
MLV and other simple retroviruses, which retain all essential viral sequences to replicate
and thus possess the ability to transmit inserted exogenous transgene sequence while
efficiently propagating into actively dividing cells. We have also developed vectors
derived from MLV that replicate in tissue-specific manner demonstrating its potential to
be even more selective in certain tissues [Logg et al., 2002].
However, there are potential pitfalls in using RCR vectors and their safety,
including long-term effects, has been the center of attention. The foremost safety
consideration for replicating vectors is the possibility of uncontrolled spread of
replicating vectors resulting in insertional mutagenesis potentially leading to
carcinogenesis. This safety hazard was heightened especially after a report of fatal
lymphomas in 3 of 10 rhesus macaques after transplant of bone marrows cells heavily
contaminated with MLV [Donahue et al., 1992]. In this incident, however, the monkeys
were subjected to severe immune suppression by irradiation prior to bone marrow
transplantation and other experiments with less severe immune suppression showed no
125

evidence of MLV pathology [Cornetta et al., 1990]. More recent evidence of insertional
mutagenesis by MLV heightened the debate to another level when it was reported that 3
of 11 immunodeficient children developed T cell leukemia after transfer of IL-R gamma-
chain gene to hematopoietic progenitors by replication defective MLV [Check, 2003;
Fischer et al., 2004; Hacein-Bey-Abina et al., 2003]. As the vectors used were
replication-defective and there was no evidence of RCR reversion, insertional activation
of the LMO-2 proto-oncogene by the MLV vector, combined with a selective growth
advantage of the IL-R corrected cells, are likely to have contributed to leukemogenesis.
Contrary to gene replacement therapy, insertional mutagenesis caused by
defective or replication competent vectors do not pose a great threat in RCR-mediated
suicide gene therapy. The very rationale for the use of replication competent vectors for
cancer therapy is to selectively eradicate actively dividing cells. As the majority of
normal cells in vivo are quiescent, especially hematopoietic stem cells, MLV-based RCR
vectors selectively transduce actively dividing cells which is the primary hallmark of
proliferating cancer cells. For retroviral gene transfer in the gamma-chain deficiency trial,
hematopoietic stem cells were harvested and forced to replicate in culture by cytokine
stimulation which is not a normal physiological condition in vivo.
Further reducing the concerns, in vivo experiments from our lab supports the
notion that MLV-based RCR vectors selectively transduce tumor tissue [Wang et al.,
2006; Wang et al., 2003c]. After intratumoral administration, RCR vectors showed no
detectable spread to normal tissues by PCR analysis in short term studies in either
immunodeficient or immunocompetent tumor models. Some spread to bone marrow and
spleen was detected in immunodeficient nude mice using more sensitive real-time
126

quantitative PCR but this was not seen immunocompetent rat syngeneic tumor models.
The central nervous system also contains dividing normal cells, including neural
progenitor in the subventricular zone [Sanai et al., 2004], microglia, and immunocytes
but we have not detected significant infection of subventricular zone cells by
immunohistochemistry suing atntibodies against viral proteins, even when infected tumor
masses could be observed in the same section in adjacent areas [Tai et al., 2005],
suggesting that the vector cannot readily penetrate or diffuse through non-dividing
normal brain parenchyma. It is reasonable to assume that infection would take place if
tumor cells migrate and directly contact actively dividing cells in the CNS and cause
systemic viral spread by migrating out of the tumor mass. On the other hand, the
migration of viral antigen presenting cells could also stimulate immune response against
virus-infected cells. Supporting our results, Klatzman et al also reported similar results
with low level detection of RCR in immunodeficient nude mice but no such detection in
immunecompetent Balb/c mice [Solly et al., 2003]
In addition to the extremely low possibility of transducing normal cells in
immunocompetent subjects, the stable integration of a suicide gene itself constitute a
safety mechanism as the host cell would “self-destruct” with the presence of pro-drug and
spread would be inherently limited. Furthermore, anti-retroviral drugs such as AZT can
readily terminate replication of wild type MLV [Ruprecht et al., 1990] as well as MLV-
based RCR vectors [Ruprecht et al., 1990; Wang et al., 2003c]. The possibility of
transducing normal cells is an undesirable event nonetheless and every effort should be
made to minimize the possibility especially since cancer patients are often immuno-
compromised increasing their risk for systemic RCR spread.
127

Deletion of transgene is also a possibility as we previously reported [Logg et al.,
2001a] and raises some concerns as this transgene in suicide gene therapy provides the
mechanism for self-destruction described above if undesirable spread takes place. In our
previous study, RCR vectors containing inserts of 1.15 to 1.3 kb replicated with kinetics
slightly attenuated compared to the wild-type MLV and maintained their genomic
integrity over multiple serial infection cycles. However, multiple species of deletion
mutants were detected in later infection cycles eventually dominating the entire
population. Logg et al. also reported one instance of insert deletion through
recombination with an endogenous retrovirus
Also, as described in this paper, there are real risks of losing control of replication
and transcription of transgene via point mutations, deletions, and homologous
recombination events in long-term vector propagation. Thus, ultimately, the use of
MLV-based RCR vectors should rest on the risk-benefit ratio – therapeutic benefit from
efficient gene delivery and persistence versus the potential adverse effects such as
insertional mutagenesis. Laboratory research and clinical trials to date make clinical
application of MLV-based RCR in cancer gene therapy a promising option.

ADDITIONAL METHODS UNDER DEVELOPMENT
While the current method of choice is untargeted RCR vectors for specific
applications where tropism modification is less essential and the risk is well justified in
poor prognosis malignancies (e.g., glioblastomas as described above), there are new
methods in active development in order to improve the selectivity and safety profiles of
RCR vectors. Promising methods are, but not limited to, (1) physical targeting, (2)
128

transcriptional targeting, (3) hybrid-vector targeting, and (4) cellular delivery of vectors
using tumor-homing cells.

Targeting tumor cells by physical binding
Modifications to the vector genome physically altering its natural tropism towards
target antigens such as tumor antigens would further increase transduction efficiency as
well as its safety profile. The proof of concept for MLV was demonstrated by Kasahara
et al. by integration of a chimeric protein in the viral envelope [Kasahara et al., 1994] and
by Tai et al. by coating the viral envelop with antibody but these approaches have proven
extremely challenging. Production and sustainability of the integrity of physically altered
enveloped vectors are extremely challenging and their gene delivery efficiency beyond
binding seems to be decreased as well. A variety of naturally occurring retroviral
envelop sequences (e.g., ecotropic, amphotropic, GALV) can be used as the env gene
within the RCR genome but this approach has not provided significant improvements in
achieving efficient levels of targeted transduction [Galanis et al., 2001] as these naturally
occurring envelop sequences do not provide significant improvements in tumor
selectivity. However, similar approaches have proven more achievable in other vector
systems such as the adenovirus where its fiber knob can be stably modified to redirect
tropism [Biermann et al., 2001; Borovjagin et al., 2005]. Given the benefits of MLV-
based RCR vectors (e.g., selectivity, low immunogenicity) physical targeting of RCR
could bring added tumor selectivity and gene delivery efficiency and, hence, requires
further development.

129

Targeting tumor cells by transcriptional targeting
Once the transgene is delivered either by selective or non-selective transduction,
selective transcription of transgene can be achieved via replacement of the viral
enhancer/promoter elements with tissue-specific regulatory sequences [Logg et al., 2002;
Logg et al., 2001a]. Variety of tissue-specific promoters have been used in conventional
replication defective vectors and we have demonstrated in this paper and previously that
RCR vector replication can be stringently restricted to prostate-derived cells by
incorporation of the prostate-specific probasin promoter into the 3’ LTR [Logg et al.,
2002]. Replication of probasin-targeted RCR vectors was tested in human prostate
cancer in vitro and in vivo and these vectors have shown replication kinetics comparable
to wild type virus and were restricted to androgen receptor positive cells (figure 3.3). To
assess the safety of these vectors, potential dissemination and genotoxicity of these
vectors were tested by real-time PCR in immunocompetent and immunodeficient mice
after systemic viral injection or bone marrow transplantation from the donors transduced
with the vectors. High copy number of integrated untargeted RCR vector was detected in
the spleen and bone marrow, whereas the prostate specific vectors showed no detectable
integration in normal tissues. No malignant changes were observed in any mice received
BMT. Targeted vectors did not mediate transgene expression into human PBMCs
whereas untargeted vectors showed significant transduction. These results indicate that
transcriptional targeting of tumor cells could improve the safety profile of RCR vectors
(Kimura T, et al., manuscript submitted).

130

Targeting tumor cells by hybrid vectors
Soifer et al. have demonstrated that high-capacity helper-dependent adenovirus
vectors can be used as a first-stage carrier for production of secondary RCR vectors
[Soifer et al., 2002]. These vectors have been completely gutted, fully deleted of all
adenoviral genes, thus have a cloning capacity of up to 36kb providing great flexibility in
transgene cassette capacity – even multiple copies of full length RCR vectors – and offer
the potential advantage of relatively low immunogenicity for Class I-mediated cellular
response to adenoviral proteins. Thus, hybrid vectors based on helper-dependent
adenoviruses directing the in situ production of RCR vectors could represent an ideal
combination, with high titer production, potentially low immunogenicity, and targeting of
binding tropism via the first-stage adenovirus, with permanent transgene integration and
amplification of initial input titer by propagation strictly restricted to actively dividing
tumor cells via the second-stage RCR, whose production from the adenovirus genome in
infected cells could further be regulated by the use of tissue-specific or inducible
promoters.

Targeting tumors by cellular delivery of RCR vectors

Some cell types exhibit preferential homing to tumor sites in vivo and the
potential to use these cells as carriers to chaperone viral vectors, including oncolytic
viruses, to tumor cells in vitro and in vivo has been demonstrated.
Cytotoxic T lymphocytes (CTLs) in particular have the unique potential for this
developing approach to combine adoptive immunotherapy by tumor antigen-activated
CTLs along with their use as vehicles for tumor-selective systemic delivery of targeted
131

RCR vectors. In principle, these CTLs would allow better RCR penetration by migrating
through the tumor mass and to tumor foci infiltrating normal brain tissue, compared to
direct intra-tumoral injection which initially achieve only limited diffusion away from the
injection site.
Replication competent retrovirus vectors are good candidates for cancer gene
therapy due to some of its inherent characteristics – such as its low immunogenicity and
inability to infect non-dividing cells. Furthermore, future strategies combining desirable
characteristics of other vectors (e.g., gutted Ad) and increasing selectivity (e.g., physical
and transcriptional targeting) demonstrate that RCR vectors are promising candidates for
cancer therapeutics.
132

BIBLIOGRAPHY

Alizon M, Wain-Hobson S, Montagnier L, Sonigo P. 1986. Genetic variability of the
AIDS virus: nucleotide sequence analysis of two isolates from African patients. Cell
46(1):63-74.
Anderson LM, Swaminathan S, Zackon I, Tajuddin AK, Thimmapaya B, Weitzman SA.
1999. Adenovirus-mediated tissue-targeted expression of the HSVtk gene for the
treatment of breast cancer. Gene Ther 6(5):854-864.
Andriani F, Nan B, Yu J, Li X, Weigel NL, McPhaul MJ, Kasper S, Kagawa S, Fang B,
Matusik RJ, Denner L, Marcelli M. 2001. Use of the probasin promoter ARR2PB to
express Bax in androgen receptor-positive prostate cancer cells. J Natl Cancer Inst
93(17):1314-1324.
Bachrach E, Duch M, Pelegrin M, Dreja H, Pedersen FS, Piechaczyk M. 2003. In vivo
infection of mice by replication-competent MLV-based retroviral vectors. Methods Mol
Med 76:343-352.
Bandres JC, Ratner L. 1994. Human immunodeficiency virus type 1 Nef protein down-
regulates transcription factors NF-kappa B and AP-1 in human T cells in vitro after T-cell
receptor stimulation. J Virol 68(5):3243-3249.
Bassuk AG, Anandappa RT, Leiden JM. 1997. Physical interactions between Ets and NF-
kappaB/NFAT proteins play an important role in their cooperative activation of the
human immunodeficiency virus enhancer in T cells. J Virol 71(5):3563-3573.
Biermann V, Volpers C, Hussmann S, Stock A, Kewes H, Schiedner G, Herrmann A,
Kochanek S. 2001. Targeting of high-capacity adenoviral vectors. Hum Gene Ther
12(14):1757-1769.
Bohnlein E, Siekevitz M, Ballard DW, Lowenthal JW, Rimsky L, Bogerd H, Hoffman J,
Wano Y, Franza BR, Greene WC. 1989. Stimulation of the human immunodeficiency
virus type 1 enhancer by the human T-cell leukemia virus type I tax gene product
involves the action of inducible cellular proteins. J Virol 63(4):1578-1586.
133

Borovjagin AV, Krendelchtchikov A, Ramesh N, Yu DC, Douglas JT, Curiel DT. 2005.
Complex mosaicism is a novel approach to infectivity enhancement of adenovirus type 5-
based vectors. Cancer Gene Ther 12(5):475-486.
Bramblett D, Hsu CL, Lozano M, Earnest K, Fabritius C, Dudley J. 1995. A redundant
nuclear protein binding site contributes to negative regulation of the mouse mammary
tumor virus long terminal repeat. J Virol 69(12):7868-7876.
Brand K, Loser P, Arnold W, Bartels T, Strauss M. 1998. Tumor cell-specific transgene
expression prevents liver toxicity of the adeno-HSVtk/GCV approach. Gene Ther
5(10):1363-1371.
Chang SC, Erwin AE, Lee AS. 1989. Glucose-regulated protein (GRP94 and GRP78)
genes share common regulatory domains and are coordinately regulated by common
trans-acting factors. Mol Cell Biol 9(5):2153-2162.
Chang SC, Wooden SK, Nakaki T, Kim YK, Lin AY, Kung L, Attenello JW, Lee AS.
1987. Rat gene encoding the 78-kDa glucose-regulated protein GRP78: its regulatory
sequences and the effect of protein glycosylation on its expression. Proc Natl Acad Sci U
S A 84(3):680-684.
Chao CC, Lin-Chao S. 1992. A direct-repeat sequence of the human BiP gene is required
for A23187-mediated inducibility and an inducible nuclear factor binding. Nucleic Acids
Res 20(24):6481-6485.
Check E. 2003. Harmful potential of viral vectors fuels doubts over gene therapy. Nature
423(6940):573-574.
Chen KD, Hung JJ, Huang HL, Chang MD, Lai YK. 1997. Rapid induction of the Grp78
gene by cooperative actions of okadaic acid and heat-shock in 9L rat brain tumor cells--
involvement of a cAMP responsive element-like promoter sequence and a protein kinase
A signaling pathway. Eur J Biochem 248(1):120-129.
Chen LY, Chiang AS, Hung JJ, Hung HI, Lai YK. 2000a. Thapsigargin-induced grp78
expression is mediated by the increase of cytosolic free calcium in 9L rat brain tumor
cells. J Cell Biochem 78(3):404-416.

134

Chen X, Zhang D, Dennert G, Hung G, Lee AS. 2000b. Eradication of murine mammary
adenocarcinoma through HSVtk expression directed by the glucose-starvation inducible
grp78 promoter. Breast Cancer Res Treat 59(1):81-90.
Claessens F, Verrijdt G, Schoenmakers E, Haelens A, Peeters B, Verhoeven G, Rombauts
W. 2001. Selective DNA binding by the androgen receptor as a mechanism for hormone-
specific gene regulation. J Steroid Biochem Mol Biol 76(1-5):23-30.
Colby WW, Chen EY, Smith DH, Levinson AD. 1983. Identification and nucleotide
sequence of a human locus homologous to the v-myc oncogene of avian
myelocytomatosis virus MC29. Nature 301(5902):722-725.
Cornetta K, Moen RC, Culver K, Morgan RA, McLachlin JR, Sturm S, Selegue J,
London W, Blaese RM, Anderson WF. 1990. Amphotropic murine leukemia retrovirus is
not an acute pathogen for primates. Hum Gene Ther 1(1):15-30.
Creighton S, Huang MM, Cai H, Arnheim N, Goodman MF. 1992. Base mispair
extension kinetics. Binding of avian myeloblastosis reverse transcriptase to matched and
mismatched base pair termini. J Biol Chem 267(4):2633-2639.
Culver KW. 1992. The potential for genetic healing. Jama 268(13):1768.
Culver KW, Moorman DW, Muldoon RR, Paulsen RM, Jr., Lamsam JL, Walling HW,
Link CJ, Jr. 1994. Toxicity and immunologic effects of in vivo retrovirus-mediated gene
transfer of the herpes simplex-thymidine kinase gene into solid tumors. Cold Spring Harb
Symp Quant Biol 59:685-690.
Culver KW, Morgan RA, Osborne WR, Lee RT, Lenschow D, Able C, Cornetta K,
Anderson WF, Blaese RM. 1990. In vivo expression and survival of gene-modified T
lymphocytes in rhesus monkeys. Hum Gene Ther 1(4):399-410.
Dai JL, Burnstein KL. 1996. Two androgen response elements in the androgen receptor
coding region are required for cell-specific up-regulation of receptor messenger RNA.
Mol Endocrinol 10(12):1582-1594.
DesGroseillers L, Rassart E, Jolicoeur P. 1983. Thymotropism of murine leukemia virus
is conferred by its long terminal repeat. Proc Natl Acad Sci U S A 80(14):4203-4207.
135

Diaz RM, Eisen T, Hart IR, Vile RG. 1998. Exchange of viral promoter/enhancer
elements with heterologous regulatory sequences generates targeted hybrid long terminal
repeat vectors for gene therapy of melanoma. J Virol 72(1):789-795.
Donahue RE, Kessler SW, Bodine D, McDonagh K, Dunbar C, Goodman S, Agricola B,
Byrne E, Raffeld M, Moen R, et al. 1992. Helper virus induced T cell lymphoma in
nonhuman primates after retroviral mediated gene transfer. J Exp Med 176(4):1125-1135.
Dong D, Dubeau L, Bading J, Nguyen K, Luna M, Yu H, Gazit-Bornstein G, Gordon
EM, Gomer C, Hall FL, Gambhir SS, Lee AS. 2004. Spontaneous and controllable
activation of suicide gene expression driven by the stress-inducible grp78 promoter
resulting in eradication of sizable human tumors. Hum Gene Ther 15(6):553-561.
Dougherty JP, Temin HM. 1988. Determination of the rate of base-pair substitution and
insertion mutations in retrovirus replication. J Virol 62(8):2817-2822.
Drummond IA, Lee AS, Resendez E, Jr., Steinhardt RA. 1987. Depletion of intracellular
calcium stores by calcium ionophore A23187 induces the genes for glucose-regulated
proteins in hamster fibroblasts. J Biol Chem 262(26):12801-12805.
Emiliusen L, Gough M, Bateman A, Ahmed A, Voellmy R, Chester J, Diaz RM,
Harrington K, Vile R. 2001. A transcriptional feedback loop for tissue-specific
expression of highly cytotoxic genes which incorporates an immunostimulatory
component. Gene Ther 8(13):987-998.
Fernandez PM, Tabbara SO, Jacobs LK, Manning FC, Tsangaris TN, Schwartz AM,
Kennedy KA, Patierno SR. 2000. Overexpression of the glucose-regulated stress gene
GRP78 in malignant but not benign human breast lesions. Breast Cancer Res Treat
59(1):15-26.
Feuerman MH, Lee WT, Pattengale PK, Fan H. 1988. Comparison of three recombinant
murine leukemia viruses carrying the v-src oncogene of avian sarcoma virus: differences
in in vitro transformation and in vivo pathogenicity. Mol Carcinog 1(1):57-66.
Fischer A, Abina SH, Thrasher A, von Kalle C, Cavazzana-Calvo M. 2004. LMO2 and
gene therapy for severe combined immunodeficiency. N Engl J Med 350(24):2526-2527;
author reply 2526-2527.
136

Galanis E, Vile R, Russell SJ. 2001. Delivery systems intended for in vivo gene therapy
of cancer: targeting and replication competent viral vectors. Crit Rev Oncol Hematol
38(3):177-192.
Gazit G, Hung G, Chen X, Anderson WF, Lee AS. 1999. Use of the glucose starvation-
inducible glucose-regulated protein 78 promoter in suicide gene therapy of murine
fibrosarcoma. Cancer Res 59(13):3100-3106.
Gazit G, Kane SE, Nichols P, Lee AS. 1995. Use of the stress-inducible grp78/BiP
promoter in targeting high level gene expression in fibrosarcoma in vivo. Cancer Res
55(8):1660-1663.
Greenberg NM, DeMayo F, Finegold MJ, Medina D, Tilley WD, Aspinall JO, Cunha
GR, Donjacour AA, Matusik RJ, Rosen JM. 1995. Prostate cancer in a transgenic mouse.
Proc Natl Acad Sci U S A 92(8):3439-3443.
Greenberg NM, DeMayo FJ, Sheppard PC, Barrios R, Lebovitz R, Finegold M,
Angelopoulou R, Dodd JG, Duckworth ML, Rosen JM, et al. 1994. The rat probasin gene
promoter directs hormonally and developmentally regulated expression of a heterologous
gene specifically to the prostate in transgenic mice. Mol Endocrinol 8(2):230-239.
Gruters RA, Otto SA, Al BJ, Verhoeven AJ, Verweij CL, Van Lier RA, Miedema F.
1991. Non-mitogenic T cell activation signals are sufficient for induction of human
immunodeficiency virus transcription. Eur J Immunol 21(1):167-172.
Hacein-Bey-Abina S, Von Kalle C, Schmidt M, McCormack MP, Wulffraat N, Leboulch
P, Lim A, Osborne CS, Pawliuk R, Morillon E, Sorensen R, Forster A, Fraser P, Cohen
JI, de Saint Basile G, Alexander I, Wintergerst U, Frebourg T, Aurias A, Stoppa-Lyonnet
D, Romana S, Radford-Weiss I, Gross F, Valensi F, Delabesse E, Macintyre E, Sigaux F,
Soulier J, Leiva LE, Wissler M, Prinz C, Rabbitts TH, Le Deist F, Fischer A, Cavazzana-
Calvo M. 2003. LMO2-associated clonal T cell proliferation in two patients after gene
therapy for SCID-X1. Science 302(5644):415-419.
Hanecak R, Pattengale PK, Fan H. 1991. Deletion of a GC-rich region flanking the
enhancer element within the long terminal repeat sequences alters the disease specificity
of Moloney murine leukemia virus. J Virol 65(10):5357-5363.
137

Hannibal MC, Markovitz DM, Nabel GJ. 1994. Multiple cis-acting elements in the
human immunodeficiency virus type 2 enhancer mediate the response to T-cell receptor
stimulation by antigen in a T-cell hybridoma line. Blood 83(7):1839-1846.
Hou MC, Shen CH, Lee WC, Lai YK. 1993. Okadaic acid as an inducer of the 78-kDa
glucose-regulated protein in 9L rat brain tumor cells. J Cell Biochem 51(1):91-101.
Hu WS, Temin HM. 1990a. Genetic consequences of packaging two RNA genomes in
one retroviral particle: pseudodiploidy and high rate of genetic recombination. Proc Natl
Acad Sci U S A 87(4):1556-1560.
Hu WS, Temin HM. 1990b. Retroviral recombination and reverse transcription. Science
250(4985):1227-1233.
Huang J, Powell WC, Khodavirdi AC, Wu J, Makita T, Cardiff RD, Cohen MB, Sucov
HM, Roy-Burman P. 2002. Prostatic intraepithelial neoplasia in mice with conditional
disruption of the retinoid X receptor alpha allele in the prostate epithelium. Cancer Res
62(16):4812-4819.
Indraccolo S, Minuzzo S, Habeler W, Zamarchi R, Fregonese A, Gunzburg WH, Salmons
B, Uckert W, Chieco-Bianchi L, Amadori A. 2000. Modulation of Moloney leukemia
virus long terminal repeat transcriptional activity by the murine CD4 silencer in retroviral
vectors. Virology 276(1):83-92.
Ishimoto A, Takimoto M, Adachi A, Kakuyama M, Kato S, Kakimi K, Fukuoka K, Ogiu
T, Matsuyama M. 1987. Sequences responsible for erythroid and lymphoid leukemia in
the long terminal repeats of Friend-mink cell focus-forming and Moloney murine
leukemia viruses. J Virol 61(6):1861-1866.
Ito T, Kabuyama Y, Okazaki S, Kameda T, Murakami M, Iba H. 1998. Unstable
retrovirus mutants with acquired transforming activity: rapid changes in the number of
repeats of a specific junD polynucleotide segment. Nucleic Acids Res 26(21):4868-4873.
Kasahara N, Dozy AM, Kan YW. 1994. Tissue-specific targeting of retroviral vectors
through ligand-receptor interactions. Science 266(5189):1373-1376.
Kasper S, Matusik RJ. 2000. Rat probasin: structure and function of an outlier lipocalin.
Biochim Biophys Acta 1482(1-2):249-258.
138

Kasper S, Rennie PS, Bruchovsky N, Sheppard PC, Cheng H, Lin L, Shiu RP, Snoek R,
Matusik RJ. 1994. Cooperative binding of androgen receptors to two DNA sequences is
required for androgen induction of the probasin gene. J Biol Chem 269(50):31763-31769.
Kievit E, Bershad E, Ng E, Sethna P, Dev I, Lawrence TS, Rehemtulla A. 1999.
Superiority of yeast over bacterial cytosine deaminase for enzyme/prodrug gene therapy
in colon cancer xenografts. Cancer Res 59(7):1417-1421.
Kim YK, Lee AS. 1987. Transcriptional activation of the glucose-regulated protein genes
and their heterologous fusion genes by beta-mercaptoethanol. Mol Cell Biol 7(8):2974-
2976.
Kirn DH, McCormick F. 1996. Replicating viruses as selective cancer therapeutics. Mol
Med Today 2(12):519-527.
Klein R, Ruttkowski B, Knapp E, Salmons B, Gunzburg WH, Hohenadl C. 2006. WPRE-
mediated enhancement of gene expression is promoter and cell line specific. Gene
372:153-161.
Koong AC, Chen EY, Lee AS, Brown JM, Giaccia AJ. 1994. Increased cytotoxicity of
chronic hypoxic cells by molecular inhibition of GRP78 induction. Int J Radiat Oncol
Biol Phys 28(3):661-666.
Kubo S, Mitani K. 2003. A new hybrid system capable of efficient lentiviral vector
production and stable gene transfer mediated by a single helper-dependent adenoviral
vector. J Virol 77(5):2964-2971.
Kubo S, Saeki Y, Chiocca EA, Mitani K. 2003. An HSV amplicon-based helper system
for helper-dependent adenoviral vectors. Biochem Biophys Res Commun 307(4):826-
830.
Kuriyama S, Kikukawa M, Masui K, Okuda H, Nakatani T, Akahane T, Mitoro A,
Tominaga K, Tsujinoue H, Yoshiji H, Okamoto S, Fukui H, Ikenaka K. 1999. Cancer
gene therapy with HSV-tk/GCV system depends on T-cell-mediated immune responses
and causes apoptotic death of tumor cells in vivo. Int J Cancer 83(3):374-380.

139

Li M, Baumeister P, Roy B, Phan T, Foti D, Luo S, Lee AS. 2000. ATF6 as a
transcription activator of the endoplasmic reticulum stress element: thapsigargin stress-
induced changes and synergistic interactions with NF-Y and YY1. Mol Cell Biol
20(14):5096-5106.
Li Y, Golemis E, Hartley JW, Hopkins N. 1987. Disease specificity of nondefective
Friend and Moloney murine leukemia viruses is controlled by a small number of
nucleotides. J Virol 61(3):693-700.
Liu YZ, Latchman DS. 1997. The octamer-binding proteins Oct-1 and Oct-2 repress the
HIV long terminal repeat promoter and its transactivation by Tat. Biochem J 322 ( Pt
1):155-158.
Logg CR, Kasahara N. 2004. Retrovirus-mediated gene transfer to tumors: utilizing the
replicative power of viruses to achieve highly efficient tumor transduction in vivo.
Methods Mol Biol 246:499-525.
Logg CR, Logg A, Matusik RJ, Bochner BH, Kasahara N. 2002. Tissue-specific
transcriptional targeting of a replication-competent retroviral vector. J Virol
76(24):12783-12791.
Logg CR, Logg A, Tai CK, Cannon PM, Kasahara N. 2001a. Genomic stability of murine
leukemia viruses containing insertions at the Env-3' untranslated region boundary. J Virol
75(15):6989-6998.
Logg CR, Tai CK, Logg A, Anderson WF, Kasahara N. 2001b. A uniquely stable
replication-competent retrovirus vector achieves efficient gene delivery in vitro and in
solid tumors. Hum Gene Ther 12(8):921-932.
LoSardo JE, Cupelli LA, Short MK, Berman JW, Lenz J. 1989. Differences in activities
of murine retroviral long terminal repeats in cytotoxic T lymphocytes and T-lymphoma
cells. J Virol 63(3):1087-1094.
Lowe SL, Rubinchik S, Honda T, McDonnell TJ, Dong JY, Norris JS. 2001. Prostate-
specific expression of Bax delivered by an adenoviral vector induces apoptosis in LNCaP
prostate cancer cells. Gene Ther 8(18):1363-1371.

140

Lyubchenko TA, Wurth GA, Zweifach A. 2001. Role of calcium influx in cytotoxic T
lymphocyte lytic granule exocytosis during target cell killing. Immunity 15(5):847-859.
Mendelman LV, Petruska J, Goodman MF. 1990. Base mispair extension kinetics.
Comparison of DNA polymerase alpha and reverse transcriptase. J Biol Chem
265(4):2338-2346.
Miller DG, Adam MA, Miller AD. 1990. Gene transfer by retrovirus vectors occurs only
in cells that are actively replicating at the time of infection. Mol Cell Biol 10(8):4239-
4242.
Mitani K, Kubo S. 2002. Adenovirus as an integrating vector. Curr Gene Ther 2(2):135-
144.
Murphy SJ, Chong H, Bell S, Diaz RM, Vile RG. 2002. Novel integrating
adenoviral/retroviral hybrid vector for gene therapy. Hum Gene Ther 13(6):745-760.
Nemunaitis J. 1999. Oncolytic viruses. Invest New Drugs 17(4):375-386.
Parker R, Phan T, Baumeister P, Roy B, Cheriyath V, Roy AL, Lee AS. 2001.
Identification of TFII-I as the endoplasmic reticulum stress response element binding
factor ERSF: its autoregulation by stress and interaction with ATF6. Mol Cell Biol
21(9):3220-3233.
Parrott C, Seidner T, Duh E, Leonard J, Theodore TS, Buckler-White A, Martin MA,
Rabson AB. 1991. Variable role of the long terminal repeat Sp1-binding sites in human
immunodeficiency virus replication in T lymphocytes. J Virol 65(3):1414-1419.
Perrino FW, Preston BD, Sandell LL, Loeb LA. 1989. Extension of mismatched 3'
termini of DNA is a major determinant of the infidelity of human immunodeficiency
virus type 1 reverse transcriptase. Proc Natl Acad Sci U S A 86(21):8343-8347.
Pulsinelli GA, Temin HM. 1991. Characterization of large deletions occurring during a
single round of retrovirus vector replication: novel deletion mechanism involving errors
in strand transfer. J Virol 65(9):4786-4797.

141

Pulsinelli GA, Temin HM. 1994. High rate of mismatch extension during reverse
transcription in a single round of retrovirus replication. Proc Natl Acad Sci U S A
91(20):9490-9494.
Rainov NG, Kramm CM, Banning U, Riemann D, Holzhausen HJ, Heidecke V, Burger
KJ, Burkert W, Korholz D. 2000. Immune response induced by retrovirus-mediated
HSV-tk/GCV pharmacogene therapy in patients with glioblastoma multiforme. Gene
Ther 7(21):1853-1858.
Ram Z, Walbridge S, Oshiro EM, Viola JJ, Chiang Y, Mueller SN, Blaese RM, Oldfield
EH. 1994. Intrathecal gene therapy for malignant leptomeningeal neoplasia. Cancer Res
54(8):2141-2145.
Reddy RK, Dubeau L, Kleiner H, Parr T, Nichols P, Ko B, Dong D, Ko H, Mao C,
DiGiovanni J, Lee AS. 2002. Cancer-inducible transgene expression by the Grp94
promoter: spontaneous activation in tumors of various origins and cancer-associated
macrophages. Cancer Res 62(24):7207-7212.
Rennie PS, Bruchovsky N, Leco KJ, Sheppard PC, McQueen SA, Cheng H, Snoek R,
Hamel A, Bock ME, MacDonald BS, et al. 1993. Characterization of two cis-acting DNA
elements involved in the androgen regulation of the probasin gene. Mol Endocrinol
7(1):23-36.
Rhode BW, Emerman M, Temin HM. 1987. Instability of large direct repeats in
retrovirus vectors. J Virol 61(3):925-927.
Robbins PD, Ghivizzani SC. 1998. Viral vectors for gene therapy. Pharmacol Ther
80(1):35-47.
Rubinchik S, Wang D, Yu H, Fan F, Luo M, Norris JS, Dong JY. 2001. A complex
adenovirus vector that delivers FASL-GFP with combined prostate-specific and
tetracycline-regulated expression. Mol Ther 4(5):416-426.
Ruprecht RM, Mullaney S, Bernard LD, Gama Sosa MA, Hom RC, Finberg RW. 1990.
Vaccination with a live retrovirus: the nature of the protective immune response. Proc
Natl Acad Sci U S A 87(14):5558-5562.

142

Russell SJ. 1994. Replicating vectors for gene therapy of cancer: risks, limitations and
prospects. Eur J Cancer 30A(8):1165-1171.
Sanai N, Tramontin AD, Quinones-Hinojosa A, Barbaro NM, Gupta N, Kunwar S,
Lawton MT, McDermott MW, Parsa AT, Manuel-Garcia Verdugo J, Berger MS,
Alvarez-Buylla A. 2004. Unique astrocyte ribbon in adult human brain contains neural
stem cells but lacks chain migration. Nature 427(6976):740-744.
Shiu RP, Pouyssegur J, Pastan I. 1977. Glucose depletion accounts for the induction of
two transformation-sensitive membrane proteinsin Rous sarcoma virus-transformed chick
embryo fibroblasts. Proc Natl Acad Sci U S A 74(9):3840-3844.
Short MP, Choi BC, Lee JK, Malick A, Breakefield XO, Martuza RL. 1990. Gene
delivery to glioma cells in rat brain by grafting of a retrovirus packaging cell line. J
Neurosci Res 27(3):427-439.
Shuda M, Kondoh N, Imazeki N, Tanaka K, Okada T, Mori K, Hada A, Arai M,
Wakatsuki T, Matsubara O, Yamamoto N, Yamamoto M. 2003. Activation of the ATF6,
XBP1 and grp78 genes in human hepatocellular carcinoma: a possible involvement of the
ER stress pathway in hepatocarcinogenesis. J Hepatol 38(5):605-614.
Sinn PL, Sauter SL, McCray PB, Jr. 2005. Gene therapy progress and prospects:
development of improved lentiviral and retroviral vectors--design, biosafety, and
production. Gene Ther 12(14):1089-1098.
Soifer H, Higo C, Logg CR, Jih LJ, Shichinohe T, Harboe-Schmidt E, Mitani K,
Kasahara N. 2002. A novel, helper-dependent, adenovirus-retrovirus hybrid vector: stable
transduction by a two-stage mechanism. Mol Ther 5(5 Pt 1):599-608.
Solly SK, Trajcevski S, Frisen C, Holzer GW, Nelson E, Clerc B, Abordo-Adesida E,
Castro M, Lowenstein P, Klatzmann D. 2003. Replicative retroviral vectors for cancer
gene therapy. Cancer Gene Ther 10(1):30-39.
Song MS, Park YK, Lee JH, Park K. 2001. Induction of glucose-regulated protein 78 by
chronic hypoxia in human gastric tumor cells through a protein kinase C-
epsilon/ERK/AP-1 signaling cascade. Cancer Res 61(22):8322-8330.

143

Stuhlmann H, Jaenisch R, Mulligan RC. 1989. Transfer of a mutant dihydrofolate
reductase gene into pre- and postimplantation mouse embryos by a replication-competent
retrovirus vector. J Virol 63(11):4857-4865.
Tai CK, Kasahara N. 2008. Replication-competent retrovirus vectors for cancer gene
therapy. Front Biosci 13:3083-3095.
Tai CK, Wang WJ, Chen TC, Kasahara N. 2005. Single-shot, multicycle suicide gene
therapy by replication-competent retrovirus vectors achieves long-term survival benefit in
experimental glioma. Mol Ther 12(5):842-851.
Temin HM. 1981. Structure, variation and synthesis of retrovirus long terminal repeat.
Cell 27(1 Pt 2):1-3.
Temin HM. 1989. Retrovirus vectors: promise and reality. Science 246(4933):983.
Temin HM. 1993. Retrovirus variation and reverse transcription: abnormal strand
transfers result in retrovirus genetic variation. Proc Natl Acad Sci U S A 90(15):6900-
6903.
Temin HM. 1995. Genetics of retroviruses. Ann N Y Acad Sci 758:161-165.
Thiesen HJ, Bosze Z, Henry L, Charnay P. 1988. A DNA element responsible for the
different tissue specificities of Friend and Moloney retroviral enhancers. J Virol
62(2):614-618.
Turelli P, Mangeat B, Jost S, Vianin S, Trono D. 2004. Inhibition of hepatitis B virus
replication by APOBEC3G. Science 303(5665):1829.
Varda-Bloom N, Shaish A, Gonen A, Levanon K, Greenbereger S, Ferber S, Levkovitz
H, Castel D, Goldberg I, Afek A, Kopolovitc Y, Harats D. 2001. Tissue-specific gene
therapy directed to tumor angiogenesis. Gene Ther 8(11):819-827.
Vennstrom B, Sheiness D, Zabielski J, Bishop JM. 1982. Isolation and characterization of
c-myc, a cellular homolog of the oncogene (v-myc) of avian myelocytomatosis virus
strain 29. J Virol 42(3):773-779.
144

Vile R, Russell SJ. 1994. Gene transfer technologies for the gene therapy of cancer. Gene
Ther 1(2):88-98.
Vile RG, Russell SJ, Lemoine NR. 2000. Cancer gene therapy: hard lessons and new
courses. Gene Ther 7(1):2-8.
Wang A, Huang H, Li S. 2003a. Therapeutic effect of AdCMVCD/5-FC system and
metabolism of 5-FC in the treatment of human tongue squamous cell carcinoma. Chin
Med J (Engl) 116(2):248-252.
Wang B, Weng J. 2002. [Effects of adenovirus-mediated HSV-tk/GCV system on lens
epithelium]. Zhonghua Yan Ke Za Zhi 38(10):618-622.
Wang CJ, Li JJ, Wen XM, Deng L. 2004. [Effect of recombinant adenoviruses with
CD/TK fusion suicide gene on human hepatocellular carcinoma cells]. Zhong Xi Yi Jie
He Xue Bao 2(1):42-45.
Wang Q, Lu D, Xing Y, Xue J, Qiu X. 1998. Suicide effect on rat gliomas mediated by
recombinant adenovirus thymidine kinase/acyclovir system. Chin Med J (Engl)
111(6):483-487.
Wang S, Gao J, Lei Q, Rozengurt N, Pritchard C, Jiao J, Thomas GV, Li G, Roy-Burman
P, Nelson PS, Liu X, Wu H. 2003b. Prostate-specific deletion of the murine Pten tumor
suppressor gene leads to metastatic prostate cancer. Cancer Cell 4(3):209-221.
Wang W, Tai CK, Kershaw AD, Solly SK, Klatzmann D, Kasahara N, Chen TC. 2006.
Use of replication-competent retroviral vectors in an immunocompetent intracranial
glioma model. Neurosurg Focus 20(4):E25.
Wang WJ, Tai CK, Kasahara N, Chen TC. 2003c. Highly efficient and tumor-restricted
gene transfer to malignant gliomas by replication-competent retroviral vectors. Hum
Gene Ther 14(2):117-127.

145

Wang ZH, Zagzag D, Zeng B, Kolodny EH. 1999. In vivo and in vitro glioma cell killing
induced by an adenovirus expressing both cytosine deaminase and thymidine kinase and
its association with interferon-alpha. J Neuropathol Exp Neurol 58(8):847-858.
Watson DK, Reddy EP, Duesberg PH, Papas TS. 1983. Nucleotide sequence analysis of
the chicken c-myc gene reveals homologous and unique coding regions by comparison
with the transforming gene of avian myelocytomatosis virus MC29, delta gag-myc. Proc
Natl Acad Sci U S A 80(8):2146-2150.
Weber E, Anderson WF, Kasahara N. 2001. Recent advances in retrovirus vector-
mediated gene therapy: teaching an old vector new tricks. Curr Opin Mol Ther 3(5):439-
453.
Weiss RA, Mason WS, Vogt PK. 1973. Genetic recombinants and heterozygotes derived
from endogenous and exogenous avian RNA tumor viruses. Virology 52(2):535-552.
Yang Y, Haecker SE, Su Q, Wilson JM. 1996a. Immunology of gene therapy with
adenoviral vectors in mouse skeletal muscle. Hum Mol Genet 5(11):1703-1712.
Yang Y, Jooss KU, Su Q, Ertl HC, Wilson JM. 1996b. Immune responses to viral
antigens versus transgene product in the elimination of recombinant adenovirus-infected
hepatocytes in vivo. Gene Ther 3(2):137-144.
Yang Y, Su Q, Wilson JM. 1996c. Role of viral antigens in destructive cellular immune
responses to adenovirus vector-transduced cells in mouse lungs. J Virol 70(10):7209-
7212.
Yoon TK, Shichinohe T, Laquerre S, Kasahara N. 2001. Selectively replicating
adenoviruses for oncolytic therapy. Curr Cancer Drug Targets 1(2):85-107.
Yu D, Jia WW, Gleave ME, Nelson CC, Rennie PS. 2004. Prostate-tumor targeting of
gene expression by lentiviral vectors containing elements of the probasin promoter.
Prostate 59(4):370-382.

146

Yu H, Goodman MF. 1992. Comparison of HIV-1 and avian myeloblastosis virus reverse
transcriptase fidelity on RNA and DNA templates. J Biol Chem 267(15):10888-10896.
Zhang J, Temin HM. 1994. Retrovirus recombination depends on the length of sequence
identity and is not error prone. J Virol 68(4):2409-2414.
Zhang J, Thomas TZ, Kasper S, Matusik RJ. 2000. A small composite probasin promoter
confers high levels of prostate-specific gene expression through regulation by androgens
and glucocorticoids in vitro and in vivo. Endocrinology 141(12):4698-4710.

147

Abstract (if available)

Abstract Advances in techniques for gene transfer and expression have made feasible the treatment of malignancies at the genetic level by introduction of exogenous genes into tumor cells. Numerous clinical trials of cancer gene therapy have been initiated utilizing replication defective vectors only to face limited success due to low, often undetectable, level of transduction efficiency. Replication competent vectors have been developed more recently demonstrating efficient transduction as well as tissue specific expression. These studies demonstrate the potential benefits of utilizing replication competent retrovirus as efficient gene delivery vehicles but undermine the potential risks in their innate tendency to mutate potentially resulting in an uncontrolled spread of replication-competent mutant virus with adverse effects on normal cells. To understand the potential risks in persistent long-term infection of replication competent retrovirus vectors (RCR), our study focused on the genomic stability of the long terminal repeat (LTR) of murine leukemia virus (MLV) responsible for driving the expression of viral genes for replication since our vector designs to date manipulate this region to drive selective replication of vectors. After long term propagation of various RCR vectors in various tumor cell lines, we demonstrate here that there are regions of instability in the LTR but very low risk in having a mutant virus with adverse effects as the very reason for utilizing RCR vector is for efficient spread and the vectors are normally equipped with a self-destruction mechanism.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Engineering viral vectors for gene and cell targeting

PDF

Non-viral and viral hematopoietic progenitor cell gene therapy

PDF

Improving adeno-associated viral vector for hematopoietic stem cells gene therapy

PDF

Gene therapy for the prevention and treatment of leukemia

PDF

Effects of a GSK-3 inhibitor on retroviral-mediated gene transfer to human CD34+ hematopoietic progenitor cells

PDF

Entry and intracellular trafficking of measles virus glycoprotein pseudotyping lentivector for transduction in target cells

PDF

Engineering lentiviral vectors for gene therapy and DC-vaccine

PDF

Study of dendritic cell targeting by engineered lentivectors

PDF

Pseudotyped viral vectors: HIV gene therapy applications and basic studies of SARS-COV-2

PDF

Using engineered exosomes and gene-editing to target latent HIV

PDF

Rational selection of CRISPR/Cas9 guide RNAs for homology directed genome editing and its utility in the development of gene therapies

Asset Metadata

Creator Yoon, Laurent (author)

Core Title Genomic stability of transcriptionally targeted replication competent retroviral vectors

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Biochemistry and Molecular Biology

Degree Conferral Date 2008-05

Publication Date 04/21/2010

Defense Date 04/20/2005

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

Tag gene therapy,genomic stability,OAI-PMH Harvest,replication competent retrovirus

Language English

Advisor Cannon, Paula (committee chair), Kasahara, Noriyuki (committee member), Kohn, Donald B. (committee member)

Creator Email ltyoon@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m1164

Unique identifier UC1161488

Identifier etd-Yoon-20080421 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-64582 (legacy record id),usctheses-m1164 (legacy record id)

Legacy Identifier etd-Yoon-20080421.pdf

Dmrecord 64582

Document Type Dissertation

Rights Yoon, Laurent

Type texts

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

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu