University of Southern California Dissertations and Theses

Gene therapy for the prevention and treatment of leukemia

(USC Thesis Other)

Gene therapy for the prevention and treatment of leukemia

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

GENE THERAPY FOR THE PREVENTION
AND TREATMENT OF LEUKEMIA

by

Martina Blumenthal

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

August 2007

Copyright 2007 Martina Blumenthal
Acknowledgements

I thank my Ph.D. advisory committee, Dr. Donald B. Kohn, Dr. Axel H. Schönthal, Dr.
James Ou and Dr. Paula Cannon, for their guidance through the dissertation
process and for their approval of my work for publication.

I particularly thank my doctoral advisor and mentor Donald B. Kohn for the
extraordinary opportunity to work on gene therapy and leukemia in his laboratory. Dr.
Kohn was an excellent teacher and supervisor who helped to develop my scientific
judgment by giving me a tremendous amount of freedom and responsibility while
guiding me with his valuable and endless advice. I thank him for his constant
support, encouragement and reassurance that made my doctoral work at Childrens
Hospital a remarkable experience that I will never forget.

I specially acknowledge my co-worker Dianne Skelton who assisted me on
numerous experiments with self-less devotion. She has been a good friend
throughout the years and her positive attitude always makes me smile; my
dissertation would not have been the same without her. I also thank Denise
Carbonaro for her indispensable help and honest opinion on diverse topics; the
vector core for cloning and producing some of the viral vectors presented in this
work; the FACS core for cell sorting and technical assistance; and the animal facility
for animal housing and care.

ii
It was a pleasure and honor to be part of the Kohn laboratory. I met many intelligent
and successful people here, and the whole team always stood for the highest level
of productivity achieved through excellence, respect and companionship. The
extended network of the BMT division is a perfect fit for this group because it
consists of equally talented people that have not forgotten the importance of having
fun. I am grateful for the big social events we had together as well as the daily lunch
gatherings. Thank you all for making work a place it was a joy to come to every day.

I express my highest thanks to my parents, Irmhild and Werner Blumenthal, my
brother Sebastian and my grandparents Emmi and, in loving memory, Rolf
Blumenthal. Thank you for your love, your emotional and financial support, and your
sacrifice of letting me move across the globe to Southern California. Alexander
Zarzuela, thank you for your care and support over here, especially through the
toughest and final stages of my dissertation. You all were there for me when I
needed you the most and believed in me when I despaired- I love you all and hope I
will make you proud.

In grateful appreciation,
Martina Blumenthal, Ph.D.

iii
Table of Contents

Acknowledgements ....................................................................................................ii
List of Tables ..............................................................................................................v
List of Figures............................................................................................................vi
Abstract ................................................................................................................... viii
Chapter 1- Introduction.............................................................................................. 1
Chapter 2- Suicide Gene Therapy for Leukemia..................................................... 15
2.1 Introduction................................................................................................ 15
2.2- Materials and Methods ............................................................................. 19
2.3 Results....................................................................................................... 25
2.4 Discussion ................................................................................................. 48

Chapter 3- IL-12 Adjuvant Therapy for Leukemia Cell Vaccination......................... 53
3.1- Introduction............................................................................................... 53
3.2 Material and Methods ................................................................................ 56
3.3- Results ..................................................................................................... 63
3.4- Discussion ................................................................................................ 80

Chapter 4- Leukemia Cell Vaccination after Bone Marrow Transplantation............ 87
4.1- Introduction............................................................................................... 87
4.2- Material and Methods............................................................................... 91
4.3- Results ..................................................................................................... 97
4.4- Discussion .............................................................................................. 113

Chapter 5- Concluding Remarks ........................................................................... 118
5.1- Gene therapy for congenital hematological diseases............................. 118
5.2- Immunotherapy with modified cancer cell-based vaccines .................... 121

Bibliography........................................................................................................... 126

iv
List of Tables

Table 1. In vitro Ganciclovir-resistant clones from Ba/F3 pools .............................. 39
Table 2. Evaluation of Ganciclovir resistant clones isolated from
transduced Ba/F3 pools in vitro and Ganciclovir-resistant
populations recovered from leukemic mice in vivo.................................... 43

Table 3. Relative p35/p70 ratio in heterogeneously transduced pools
and cloned BM185 cells as determined by p35 and p70 ELISA ............... 67

Table 4. Number of irradiated cells administered combined as a
prophylactic vaccine (Fig.3.4a).................................................................. 73

Table 5. Number of irradiated cells administered combined as a
prophylactic vaccine (Fig.3.4b).................................................................. 76

v
List of Figures

Fig 1.1: Persistent proviral integration into hematopoietic stems cells
and mature blood cells ................................................................................. 2

Fig 1.2: MND vector structure and packaging ........................................................... 5
Fig 2.1: Schematic overview over retroviral MND vector constructs ....................... 25
Fig 2.2: Characterization of Ba/F3 leukemia ........................................................... 28
Fig 2.3: In vitro and in vivo sensitivity of transduced Ba/F3 clones
to Ganciclovir.............................................................................................. 32

Fig 2.4: Dose response of Ba/F3 clones to Ganciclovir, Acyclovir
and Foscarnet............................................................................................. 34

Fig 2.5: In vitro and in vivo sensitivity of transduced Ba/F3 pools to
Ganciclovir.................................................................................................. 36

Fig 2.6: Vector integrity analysis of GCV-R clones obtained from pool 1................ 41
Fig 2.7: Vector integrity analysis of GCV-R clones obtained from pool 2................ 45
Fig 2.8: In vivo challenge with transduced pools 2 and analysis of
re-isolated GCV-resistant populations........................................................ 47

Fig 3.0: Schematic overview over the MND-scIL12 cloning procedure................... 59
Fig 3.1: Southern blot of IL-12 clones...................................................................... 64
Fig 3.2: IL-12 secretion and bioactivity.................................................................... 65
Fig 3.3: Rejection of IL-12 expressing live leukemia ............................................... 70
Fig 3.4: Leukemia rejection after prophylactic vaccination...................................... 74
Fig 3.5: Leukemia rejection after therapeutic vaccination ....................................... 79
Fig 4.1: Schematic overview of experimental design .............................................. 98
Fig 4.2: Irradiation dose titration and post-transplant survival ............................... 100
vi
Fig 4.3: Reconstitution of leukocyte subpopulations in various
peripheral organs...................................................................................... 104

Fig 4.4: Functional analysis of splenic effector cell populations............................ 107
Fig 4.5: Survival of challenge after vaccination post-BMT..................................... 110
Fig 4.6: BcrAbl-transduced bone marrow.............................................................. 112

vii
Abstract

Gene therapy is an approach to treat human diseases by stable transfer of
exogenous genes. Stem cell-based gene therapy for congenital blood disorders is
accompanied by the risk for leukemia development conferred by insertional
oncogenesis. The safety of retroviral gene transfer may be increased by including a
suicide gene in the therapeutic vector to eliminate adverse events. Mice were
challenged with leukemia cells transformed by proviral integration of a murine
oncogene that simultaneously expressed the HSV-TK suicide gene or hyperactive
mutants thereof (SR39 and sc39). After treatment with the drug substrate
Ganciclovir, leukemia developed in mice given clonal cells expressing HSV-TK, but
not SR39 or sc39. In vitro Ganciclovir resistance was observed in heterogeneously
transduced leukemia pools, and single nucleotide changes or partial loss of the
suicide gene were identified as mechanisms of drug escape. However, Ganciclovir
treatment resulted in 80-100% survival of mice challenged with partially resistant
leukemia pools expressing modified HSV-TK variants with improved biological
activity.

Retroviral gene transfer can be exploited for immunotherapy for cancer by
transducing autologous tumor cells with immunostimulatory molecules. Interleukin-
12 is a potent adjuvant for cell-based tumor vaccines, and to avoid toxic side effects
of recombinant IL-12 administration, murine acute lymphoblastic leukemia (ALL)
cells were stably transduced with IL-12 expressing vectors. Local IL-12 secretion of
viii
cellular vaccines efficiently rejected primary leukemia challenges in mice and was
slightly superior to systemic IL-12 administration in immunologic memory formation,
but anti-tumor effects were highly sentitive to IL-12 dosage. We provide evidence
that a combination of transgenic IL-12 expression in conjunction with a “triple”
immunomodulatory transgene expression panel (CD80, GM-CSF and CD40L) may
present the most potent cellular vaccination strategy to induce primary tumor
rejection and long-lasting memory formation. Our “triple” transduced leukemia
vaccine was also shown to be highly efficient in leukemia rejection one and three
months after syngeneic bone marrow transplantation. Survival rates were similar or
improved compared to untransplanted mice although there were slight deficiencies
in numerical immune reconstitution early after transplantation. Vaccination in a post-
BMT setting might therefore be beneficial to fight minimal residual disease and
prevent leukemia relapse after chemotherapy.
ix
Chapter 1- Introduction

Gene therapy
The goal of gene therapy is the stable transfer of genetic material into patients’ cells
to add, augment or repair genes in order to cure a broad range of congenital
disorders, infectious diseases and cancer. Gene therapy has particularly been
studied for hematological diseases such as severe combined immunodeficiencies
(SCID) or leukemia because blood cell are easily isolated, manipulated ex vivo and
re-introduced into the patient’s body. Furthermore, hematopoietic stem cells (HSCs)
are the best characterized adult stem cells of any human and murine tissue and
known to give rise to all lineages of the blood cell compartment. Targeting HSCs for
gene transfer therefore results not only in long-term propagation of the therapeutic
gene within the stem cell population for the lifespan of a patient, but also in the
transmission of the transgene to all mature hematopoietic lineages that are
constantly replenished from the stem cell pool (Fig1.1). Gene therapy has been
successful in clinical trials to treat a variety of congenital lymphoid and myeloid
immunodefiencies by targeting HSCs with the corrected version of the defective
gene, such as X-linked SCID (Cavazzana-Calvo 2000; Hacein-Bey-Abina 2002),
Aminodeaminase (ADA)-deficient SCID (Bordignon 1995; Kohn 1995, Kohn 1998;
Aiuti 2002) and X-linked chronic granulomatous disease (CGD) (Ott 2006).

1

Fig 1.1: Persistent proviral integration into hematopoietic stems cells and mature
blood cells
Hematopoietic stem cells (HSC) that are transduced with a viral vector propagate the
stably integrated transgene either by self-renewal or by differentiation into intermediate
progenitor cells (CLP= common lymphoid progenitor; CMP= common myeloid progenitor;
Pro-T/B = pro-B or T cell progenitors; GMP = granulocyte/ macrophage progenitor;
MEP= megakaryocyte/ erythrocyte progenitor) that give rise to mature cells of the
lymphoid (NK cells, T cells, B cells) or myeloid lineage.

Recombinant retroviruses as gene delivery systems
The most common approach for transgene delivery is the usage of recombinant
retroviruses. Murine oncoretroviruses like the Moloney murine leukemia virus (MLV)
have been widely used as gene transfer vehicles for stable integration into
mammalian cells. The typical retroviral genome constists of the gag, pol and env
genes encoding structural core proteins, enzymes including integrase, proteinase,
reverse transcriptase, and surface envelope proteins (cf Fig 2). Viral vectors have
been engineered in the laboratory to be replication incompetent by removing the
2
viral genes from the genome and replacing them with 1-2 exogenous transgenes.
Viral particles are produced in “packaging” cell lines that are supplied in trans with
plasmids carrying the viral genes necessary for virus assembly, but once the vector
is delivered to its cellular target it is not able to produce any further virus by
replication because of the missing viral components. Instead, the modified genome
carrying the transgene(s) integrates stably into the host DNA and is transmitted to
daughter cell generations upon every mitotic division.

Increasing LTR-directed transcription from modified retroviral vectors
The 5’ long terminal repeat (LTR) of the Moloney MLV contains enhancer repeats
and a TATA box and serves as a very potent constitutive promoter that can be
utilized to drive transgene expression in retroviral vector constructs without the
requirement of additional internal promoter sequences. However, the native MLV
LTR tends to get transcriptionally silenced in murine hematopoietic cells through
cytosine methylation (Challita 1994), and a serious of alteration have been made to
improve expression. First, the MLV-LTR sequence was replaced by the sequence
found in the myeloproliferative sarcoma virus (MPSV), a closely related murine
oncoretrovirus that spontaneously acquired infectivity of murine HSCs. A single
base pair substitution creating a novel binding site for the Sp1 transcription factor in
the LTR was found to be the molecular cause of the altered infectivity. Second, the
primer binding site (PBS) was substituted with the sequence derived from the
dl587rev endogenous murine retrovirus. The PBS allows binding of a tRNA to the
RNA transcript as a primer for the initiation of reverse transcription, but also harbors
a repressor binding site in the MLV sequence that is absent in the dl587rev PBS.
3
Third, the negative control region (NCR) that acts as a repressive element by
binding the dual function YY-1 transcription factor has been removed to allow
improved expression in murine HSCs. Those three alterations that were made by
Challita et al (1995) resulted in our current laboratory standard MND (MPSV LTR,
NCR deleted, dl587 PBS) retroviral vector that is optimized for expression in murine
HSCs and was used to transduced murine leukemia in this study.

Genotoxicity and methods to improve the safety of gene therapy
The major disadvantage of the use of viral vectors for gene therapy is that despite
all applied alterations the vector structure is still reminiscent of the highly
transforming oncoretroviral origin, and proviral integration has been shown to cause
transforming events known as insertional oncogenesis. In fact, the gene therapy trial
in France that cured 9 out of 10 infants with X-linked SCID had to report severe
adverse events a few years after successful treatment. Two patients developed a T
cell acute leukemia-like syndrome that was caused by proviral integration in the
vicinity and transactivation of the T cell oncogene LMO2 (Hacein-Bey Abina 2003),
and a third patient presented with lymphoproliferative disorder shortly thereafter
(Check 2005).

A large effort has been made since to improve the safety of vectors. The inclusion of
chromatin insulators can potentially shield the flanking chromatin region from the
effects of proviral enhancer/promoters, albeit no evidence of success has been
demonstrated so far and further research is needed. Lentiviral vectors have bee
studied extensively to replace gamma-retroviral vectors for gene therapies and have
4

Fig 1.2: MND vector structure and packaging
A)The genome of the Moloney murine leukemia virus (MLV) consists of the 5’ and 3’ long-
terminal repeats (LTR), the Ψ packaging region, the viral genes gag, pol and env and splice
donor and acceptor sites (D, A). In the MND vector, the MLV 5’LTR is replaced by the
myeloproliferative sarcoma virus (MPSV) LTR with the negative control region (NCR)
deleted and the dl578 primer-binding site (PBS) added. The MLV packaging sequence ( Ψ) is
retained to allow packaging into viral particles in the producer cell line (see below); viral
genes are replaced by the therapeutic gene of interest which in addition eliminates the splice
acceptor site and increases vector stability. After proviral integration into the target cell
genome, the vector mRNA is transcribed from enhancer/promoter elements in the LTR
region and the therapeutic protein is transcribed in a cap-dependent manner.
B) The viral particle consist of the single-stranded, positive vector mRNA (including the 5'
cap and 3' PolyA inside the virion) that becomes reverse transcribed into double stranded
DNA after entering the host cell, structural core proteins encoded by the gag

gene,
accessory proteins like a protease (Pro), integrase (Int) and the reverse transcriptase (RT)
which are all encoded by pol, and the surface envelope proteins (env).
C) The MND vector is replication incompetent because viral genes have been deleted from
its sequence. Gag/Pol and Env components are provided in trans by packaging and
envelope plasmids that are co-transfected into a producer/ packaging cell line together with
the vector plasmid. All viral proteins and the vector RNA are assembled into viral particles at
the cell membrane and released through a budding process into the culture supernatant
where they can be collected, concentrated and utilized for transduction of the target cell
population.
5
A

B

C

6
shown several advantages, such as enhanced transduction efficiencies especially in
non-dividing HSCs, greater genomic stability and the capacity to carry larger
transgenes (reviewed in Chang 2007). Retroviral vector integration is generally
targeted to actively transcribed genes (reviewed in Baum 2006), but other than
oncoretroviral vectors that tend to integrate near the 5’ end of active genes (Wu
2003), lentiviral vectors integrate all across transcriptional units without positional
preferences within the gene sequence (Schröder 2002). It has been argued that the
reduced propensity to integrate near promoter regions might reduce genotoxic side
effects, but this remains hypothetical at this point especially because transactivation
of oncogenes by proviral enhancers may still occur over vast distances. Although
lentiviruses seem to have many advantages over oncoretroviral systems, the clinical
use is still limited because of the great biosafety concerns against the use of HIV-
derived lentiviral vectors. HIV is a potent human pathogen and it is imperative to
prove the complete absence of recombination competent lentivirus formation in a
clinical setting. To date, lentiviral gene transfer has been tested for safety in a
Phase I study monitoring T cells of HIV/AIDS patients (Levine 2006) and has been
approved for a total of eight clinical trials, whereas retroviral vectors have been used
for gene therapy in over 290 clinical studies since 1989 (source: The Journal of
Gene Medicine Clinical trial site at http://www.wiley.co.uk/genetherapy/clinical).
Gene transfer to hematopoietic stem cells can also be achieved with non-viral
delivery systems, but stable expression levels are currently too low to meet the
needs of high efficiency transduction and transgene persistence for clinical
applications (Hollis 2006).

7
In this study we introduce an alternative approach to increase the safety of
commonly used retroviral vectors that utilizes a suicide gene included as a
secondary transgene in the therapeutic vector. Hyperproliferative clones emerging
in a patient as a result of insertional oncogenesis can potentially be eliminated by
treatment with a drug substrate that becomes converted into a toxic substance by
the suicide gene, thus eliminating transduced cells carrying a proviral insert while
leaving other tissues inert. This technology should be viewed as a “fail-safe”
mechanism, and one major disadvantage is apparently the concomitant loss of the
gene-corrected graft upon suicide drug induction. But in the light of the high cure
rate achieved with gene therapy for severe combined immunodeficiencies that is
compromised by the high risk for adverse events, a suicide gene therapy approach
is reasonable to preserve the benefits of gene therapy while abrogating the most
severe side effect. The efficiency of suicide gene therapy to prevent leukemia
development was studied in a novel model of insertional oncogenesis in mice and is
presented in Chapter 2.

Retroviral transduction of mature lymphocytes
Originally, gene therapy was developed for the treatment of inherited disorders by
replacing defective genes with their normal counterparts. But retroviral gene transfer
can be used as a broad tool for many applications profiting from the addition,
replacement or even subtraction of endogenous genes. Subsequently, a wide array
of clinical conditions has been added to the target field of gene therapy, for example
malignant, cardiovascular, neurologic and infectious diseases, and the list is
constantly growing. The second part of this report focuses on the genetic
8
manipulation of acute lymphoblastic leukemia (ALL) cells for whole cell cancer
vaccinations aimed at enhancing the recipient’s immune response to leukemia cells
persisting as minimal residual disease. Employing viral gene transfer for
immunotherapy for leukemia differs from stem cell-based therapies in the important
aspect that transgenes persist only for a limited amount of time in mature target cell
populations which are not capable of self-renewal. Moreover, in the setting of whole
cell tumor vaccinations, transduced cells injected into test animals or ultimately into
a patient on treatment are irradiated prior to infusion to prevent malignant expansion.
In this regard, the strict precautions that have to be made for gene transfer into stem
cell populations to avoid genotoxic side effects are of lesser concern for retroviral
targeting of more differentiated cell types like ALL blasts.

Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph
+
ALL)
Acute Lymphoblastic Leukemia (ALL) is the most frequent type of childhood cancer
(23%). The overall success rate in the cure of ALL is 70-80% with current
chemotherapy regiments (reviewed in Jones 2005). The remaining 20-30% of
patients that respond poorly to chemotherapy often fall into a special cytogenic
subgroup defined by the Philadelphia (Ph) chromosome, a t(9;22) chromosomal
translocation found in leukemic blasts. The presence of the Ph chromosome is a
predictive marker of poor prognosis with a high risk of treatment failure and relapse,
and less then 25% of patient achieve 5-year event-free survival after chemotherapy
and bone marrow transplantation. The failure of conventional therapies to
completely eradicate Ph
+
ALL necessitates the development of alternative or
9
synergistic strategies such as immunotherapy to compliment current forms of
treatment.

BCR-ABL: a tumor specific antigen
The translocation from chromosome 9 to 22 in Ph
+
ALL blasts results in the
expression of a novel fusion oncogene, BCR-ABL. ABL is the human homolog of the
murine Ableson tyrosine kinase, and its N-terminal fusion to the BCR signaling
molecule maintains ABL in an active conformation state (reviewed in Wong 2004).
The dysregulated kinase acts on a diverse range of signaling pathways of which
activation of the Ras-MAPK pathway appears to be the primary oncogenic event
crucial for cellular transformation. The BCR gene on chromosome 22 spans a break
point cluster region, and depending on the exact point of chromosome breakage two
chimeric products can be found, one resulting in a p210 BCR-ABL protein and one
in a smaller p185 protein. P210 is found in 90% of chronic myeloid leukemia (CML)
and 33% of adult ALL cases, but the majority of childhood ALL expresses the more
aggressive p185 form associated with poor disease outcome. The BCR-ABL fusion
oncogene has been of great interest to cancer immunologists for two reasons: 1) the
fusion region between the BCR and ABL proteins creates a novel immunogenic
epitope that elicits tumor-specific immune responses; 2) continued expression of
BCR-ABL is absolutely required for maintenance of the phenotype which prevents
the antigen from escaping immune recognition by secondary mutations
(immunoediting).

10
Cancer immunology
Tumor cells are derived from normal tissue and most of their antigens are subject to
self-tolerance. The greatest challenge of anti-cancer immunotherapy is therefore to
specifically eradicate abnormal cells while preserving healthy tissue. The presence
of the novel fusion gene BCR-ABL makes Ph
+
ALL a great candidate for
immunotherapy because cancer cells can be easily distinguished from regular cells.
The immunosurveillance theory postulates that many tumors fail to arise because
the immune system efficiently eradicates abnormal cells at an early stage. For
example, more than 30% of healthy individuals have detectable circulating levels of
high-avidity p210 specific cytotoxic T lymphocytes that efficiently prevent the
expansion of Ph
+
cells that could result in CML (Rezvani 2003). Tumors are thought
to be able to progress if transformed cells evolve to evade the immune system and
expand in an environment of immunologic tolerance. Tolerance can be induced by a
variety of mechanisms that probably act in concert, e.g. down regulation of
costimulatory molecules on cancer cells necessary for T cell stimulation; secretion
of anti-inflammatory cytokines in the tumor microenvironment; or presence of
suppressive regulatory T cells. The main goal of immunotherapy is therefore to
overcome tolerance and sensitize the immune system to the malignant cells by
presenting tumor-specific antigens and providing necessary co-stimulatory signals
(reviewed in Ryan 2007).

Tumor cell vaccination
The goal of immunotherapy for cancer is to break tolerance and induce tumor-
specific immunity. All types of cancer vaccines must provide a single or multiple
11
tumor antigens plus a strong adjuvant to activate both an innate and adaptive
immune response specific for the tumor and ultimately establish long-lasting anti-
tumor memory. The source of tumor antigens presented to host lymphocytes varies
greatly, including known immunogenic peptides or proteins, plasmid DNA or viruses
encoding tumor antigens, or whole cell tumor cells. Using lysed or irradiated
autologous tumor cells as a vaccine that are certain to contain the relevant antigens
unique for the patient is of great advantage because only a small number of tumor
epitopes are actually identified. While the isolation of cells from solid tumors has
certain limitations, large scale isolation of leukemia cells and vaccine manufacture
are clinically feasible (Haining 2005).

Early studies have shown that the systemic delivery of immunogens alone induces
specific tolerance. Many immunologic manipulations have been investigated since
to stimulate antigen presenting cells (APCs) and T lymphocytes. Co-administration
of microbial adjuvants or haptens (BCG, diphtheria toxin, keyhole limpet
hemocyanin), systemic injections of cytokines (IL-2, IL-4, IL-12, GM-CSF), gene
modification of cancer cells expressing costimulatory molecules or cytokines (CD80,
CD86, CD40L, GM-CSF) have all been shown to activate APCs, natural killer (NK)
cells, cytotoxic T lymphocytes (CTLs) and to promote the survival of antigen-specific
T cells to varying extents. Clinical trials of cellular immunotherapy have been
promising for certain types of cancer such as melanoma (Dranoff 2003), renal cell
carcinoma (Su 2003), lymphoma (Kwak 2000), colon cancer (Uyl de Groot 2005),
CML (Cathcart 2004), but failed yet to show clinical benefits for other malignancies
including ALL (Haining 2005).
12
A murine model for Ph
+
ALL
Our lab has previously established a murine model for immunotherapy in
Philadelphia chromosome-positive (Ph
+
) ALL. Balb/c bone marrow was transformed
with the human p185 Bcr-Abl oncogene to form the clonal pre-B leukemia cell line
BM185 (Stripecke 1998). Intravenous injection of 5000 BM185 cells was highly
leukemogenic and led to 100% mortality within 3 weeks, with high leukemia burden
detected in the blood, spleen and bone marrow. BM185 cells were then stably
transduced with various immunomodulatory molecules and tested for their efficacy
to protect mice from leukemia development by a) immunorejection after a challenge
with live modified leukemia cells; b) prophylactic immunity of an irradiated tumor cell
vaccine prior to wild type leukemia challenge; or c) therapeutic immunity of a
vaccine administered to eradicate pre-established leukemia. Immunorejection was
observed in 20-30 % of mice challenged with 5000 live BM185 expressing either
CD80, granulocyte macrophage colony-stimulating factor (GM-CSF), or CD40L
alone, but survival was increased to 70% when cells expressed all three
immunomodulatory molecules at once.

CD80 (B7.1) is a costimulatory molecule usually expressed on antigen presenting
cells that is required for the proper activation of naïve T cells after recognition of
antigenic peptides presented by major histocompatibility complex (MHC) molecules
on the surface of the APC. When stably expressed as an adjuvant on tumor cells,
CD80 has been shown to stimulate CD4 and CD8 T cells and to induce tumor-
specific CTL activity (Stripecke 1999, Gruber 2002). GM-CSF is a critical growth and
differentiation factor for dendritic cells (DCs), the most potent subgroup of
13
professional APCs. GM-CSF enhances the efficiency of tumor cell vaccines by
recruiting DCs to the site of its secretion and has been widely used in pre-clinical
(Chen 2000; Wu 2007) and clinical studies (Sloan 2000; Nemunaitis 2006; Simons
2006). Engagement of the CD40 molecule by CD40L (CD154) expressed on
leukemia cells can further enhance antigen cross-presentation on the APC through
up-regulation of co-stimulatory molecules like CD80 and stimulation of cytokine
production. Signaling through the CD40 receptor may replace the requirement for
CD4 helper T cells as shown in CD4 depletion studies, and it is crucial for the
activation of NK cell activity (Gruber 2002). CD80, GM-CSF and CD40L are thought
to act synergistically to stimulate all three effector arms of the adaptive immune
system (CD4, CD8 and NK cells), and because the combination of all three
molecules was superior to any other combinations tested in our ALL model, the
“TripleVax” BM185 clone (BM185-CD80/GM-CSF/CD40L) has been used in all
subsequent studies aimed at testing the effects of Interleukin-12 (IL-12) addition to
the vaccination regiment or the efficacy of vaccination after bone marrow
transplantation.
14
Chapter 2- Suicide Gene Therapy for Leukemia

2.1 Introduction

Gene therapy is a promising approach for the treatment of inherited hematopoietic
disorders. Nine out of ten infants treated for X-linked SCID (or γ
C
chain deficiency)
in a clinical trial showed immune restoration 2-5 months after gene transfer (Hacein-
Bey-Abina 2002). Unfortunately, three of the patients subsequently developed a
lymphoproliferative disease caused by the random insertion of the viral construct in
the vicinity of the human T cell oncogene LMO2 (Hacein-Bey-Abina 2003). It is
speculated that the retrovirally encoded γc chain transgene might have a co-
transforming role in leukemogenesis by providing a subtle proliferative or anti-
apoptotic advantage in conjunction with LMO-2 activation leading to clonal
expansion of rare leukemic precursors (Kohn 2003; Davé 2004; Shou 2006);
however, no adverse events have been observed in other clinical trials using the γc
transgene for the correction of X-linked SCID (Gaspar 2004) or in a trial for ADA-
deficient SCID (Aiuti 2002).

One possible approach to retain the benefits of gene therapy but to eliminate the
risk of insertional oncogenesis is to introduce a suicide gene such as the Herpes
simplex virus type 1 thymidine kinase (HSV1-TK) into the therapeutic vector. An
engrafted cell that proliferates upon vector integration readily expresses the suicide
gene and can be selectively eliminated by treatment with the prodrug Ganciclovir
that becomes converted to a toxic metabolite by HSV-TK. Retroviral HSV-TK
15
transfer has already been used in clinical trials for in situ transduction of solid
tumors (Moolten 1994; Ram 1997; Wildner 1999; Satoh 2005), and HSV-TK
expressing donor T lymphocytes have been successfully eradicated by Ganciclovir
in a murine GvHD model (Cohen 1997) and in human clinical trials of allogeneic
marrow transplants (Bonini 1997; Tiberghien 2001). The goal of this study was to
determine whether HSV-TK is a feasible tool to eradicate rapidly growing leukemia
cells.

Retroviral insertion into the host genome is a random process and the probability for
the natural occurrence of insertional oncogenesis is very low, between 10
-6
and 10
-8

per insertion event (Baum 2003). We developed a novel leukemia model of
insertional oncogenesis in which vector integration drives cell transformation by
delivering a cellular oncogene; by analogy to the clinical events, the vector must be
present and transcriptionally active for the leukemia to persist. The IL-3 dependent
murine pro-B leukemia cell line Ba/F3 has been described to proliferate factor-
independently after being transduced with a constitutively active mutant of the
murine proto-oncogene c-Kit
D814V
(Kitayama 1995). We designed retroviral vectors
to express the c-Kit
D814V
oncogene from the MLV-LTR followed by the HSV-TK
suicide gene expressed from an internal ribosome entry site (IRES).

HSV-TK, however, may be a suboptimal suicide gene candidate to eliminate
insertional oncogenesis for three reasons. First, the high drug doses required for a
complete response of the HSV-TK/Ganciclovir system can have unwanted
myelosuppressive side effects (Qasim 2002). Second, a subset of HSV-TK
16
transduced cells might be refractory to Ganciclovir treatment and persist as
Ganciclovir-resistant leukemia similar to viral resistance observed in clinical isolates
of Herpes viruses in patients with prolonged therapeutic Ganciclovir treatment for
viral infections (Limaye 2000). And third, HSV-TK is known to contain cryptic
splicing donor and acceptor sites flanking the active site of the kinase; truncated
splicing variants packaged into retroviral particles in producer cells represent a small
percentage of vector integrants in a heterogeneously transduced target cell pool that
are unable to produce the cytotoxic metabolite and are drug resistant (Garin 2001).
To address these potential limits of the HSV-TK suicide system, we included two
improved HSV-TK variants in our study. SR39, an HSV-TK mutant with an
increased affinity for Ganciclovir showed improved tumor cell killing in vivo
compared to wild type HSV-TK (Black 2001) and may circumvent GCV cytotoxicity
by responding to lower drug doses. We created a splice corrected version of the
SR39 mutant, sc39, by introducing silent point mutations into the cryptic splice
donor and acceptor sites of the SR39 gene. These mutations have been reported to
abrogate aberrant splicing in the wild type HSV-TK gene (Chalmers 2001), but to
our knowledge have never been applied in combination with the hyperactive mutant
SR39. We hypothesize that elimination of leukemia cells in our murine model of
retroviral insertional oncogenesis will be more efficient using the HSV-TK based
mutants SR39 and splice corrected sc39 which have a higher affinity for their drug
substrate Ganciclovir and therefore respond to lower systemic GCV doses
compared to wild-type HSV-TK.

17
We show that the hyperactive TK variants SR39 and sc39 expressed in clonal
Ba/F3 cell lines conferred protection from leukemia development in mice challenged
by intravenous injection of leukemia cells and then treated with Ganciclovir, in
contrast to wild type HSV-TK transduced Ba/F3 cells which were not eradicated with
Ganciclovir in vivo. Although Ganciclovir resistance in heterogeneously transduced
Ba/F3 pools was observed for all three TK constructs in vitro and correlated with
diminished elimination of leukemia cells in vivo, a high level of protection from
leukemia development was still achieved using the hyperactive TK variants SR39 or
sc39 as a suicide gene in this novel model of insertional oncogenesis.
18
2.2- Materials and Methods

Mice
Male Balb/c mice, 6 −8 weeks old, were purchased from Harlan Bioproducts for
Science (Indianapolis, IN) and maintained in our animal care facility under standard
conditions. All experiments involving mice were reviewed and approved by the
CHLA Institutional Animal Care & Use Committee (IACUC) (Childrens Hospital Los
Angeles, Los Angeles, CA).

Cell culture and cell lines
Ba/F3 cells, a murine pre-B leukemia cell line strictly dependent on murine IL-3
(German National Resource Centre for Biological Material, Braunschweig,
Germany), was cultured in RPMI 1640 medium supplemented with 10% fetal bovine
serum (FBS), 2 mM L-glutamine, penicillin/streptomycin (100 units/ml) and 5 ng/ml
recombinant mouse IL-3 (Biosource, Camarillo CA) and subcultured every 3-5 days.
The human embryonic kidney cell line 293T (obtained from American Type Culture
Collection- ATCC, Manassas, VA) was maintained in Dulbecco’s modified essential
medium (DMEM) supplemented with 10% FBS, 2 mM L-glutamine and penicillin/
streptomycin (100 units/ml) and subcultured at 80% confluency. The mouse
embryonic fibroblast cell line 3T3 (ATCC) was maintained in DMEM supplemented
with 10% newborn calf serum (NCS), 2 mM L-glutamine and penicillin/ streptomycin
(100 units/ml) and subcultured at 80% confluency.

19
Plasmid constructs
All of our retroviral vectors were of the Moloney murine leukemia virus (Mo-MuLV)-
based MND-series that contains the long terminal repeat (LTR) from the
Myeloproliferative Sarcoma Virus with alterations in the LTR enhancer, primer
binding site and negative control region and that has been previously described to
minimize gene silencing and achieve enhanced expression in murine hematopoietic
cells (Challita 1995; Robbins 1997; Robbins 1998; Halene 1999). The murine c-
kitD814V oncogene (Jahn 2002) was cloned into a multicloning site of the MND
backbone, followed by the encephalomyocarditis virus E5m internal ribosome entry
site (IRES) and a suicide gene. The suicide gene consisted of either the wild type
Herpes simplex virus type-1 thymidine kinase (HSV-TK) cDNA, the SR39 mutant
generated by semi-random mutagenesis with increased prodrug sensitivity (Black
2002) and a splice-corrected variant of the SR39 gene. To generate the splice-
corrected (sc) SR39 gene, we mutated the cryptic splice donor and acceptor sites
using a site-directed mutagenesis kit (Stratagene, La Jolla CA) as described
elsewhere (Chalmers 2001).

Retroviral vector production and transduction of Ba/F3 cells
Retroviral vectors were produced by transient tri-transfection of 293T cells with
vector plasmid, pHIT60 packaging plasmid and pHIT123 ecotropic envelope plasmid
by DNA/ calcium phosphate co-precipitation (Soneoka 1995). Culture supernatants
were collected twice at 24 and 48 h post-transfection and concentrated 100-fold by
filtration through Centricon columns (Millipore, Billerica MA). The viral titers were
20
determined by transduction of 3T3 cells followed by Southern blotting and were
found to be in the range of 1x106-1x107 IU/ml.
To be transduced at an MOI of ~1, Ba/F3 cells were diluted to a concentration of
1x104 cells/ml, spinoculated in the presence of virus for 1h at 2500 rpm in 3 ml of
medium containing 5 ng/ml IL-3, resuspended in the same transduction medium and
plated onto retronectin-coated 6 well plates overnight. Transduced pools were
washed with PBS and expanded in culture medium with IL-3 for 7 days before IL-3
was omitted from the medium to enrich for transduced cells only. Cells were viably
cryopreserved as pools or sorted by flow cytometry to obtain single cell clones.
Transduced Ba/F3 cells were cultured IL-3 independently in RPMI 1640 with 10%
FBS, L-glutamine and pen/strep. Clonal cell lines were obtained by flow cytometry
using an automated cell deposition unit (ACDU) to deposit single cells sorted for
viability by forward and side scatter into wells of a 96 well U-bottom plate (Costar
Corning) containing 180 μl of RPMI supplemented with 20% FBS, L-glutamine and
pen/strep without IL-3. After 10-14 days, visibly growing colonies were expanded in
regular culture medium with 10% FBS and no IL-3.

Cytotoxicity assay
Transduced Ba/F3 clones or pools were plated in triplicate as 1000 cells in 100 μl
medium with the indicated concentrations of Ganciclovir (Cytovene®-IV, Roche
Laboratories, Nutley NJ), Acyclovir (Zovirax®, GlaxoSmithKline, Greenville NC) or
Foscarnet (Foscavir®, Astra Zeneca LP, Wilmington DE) on 96 well assay black
plates (Costar Corning, Corning NY). After five days of incubation at 37°C in a
humidified 5% CO2 tissue culture incubator, 10 μl CellTiter96 AQueous One
21
Solution (Promega, Madison WI) were added per well and incubated for an
additional 4 hours before the absorbance at 450 nm was measured using an ELISA
plate reader. After subtraction of background values, percent cell viability was
calculated as the mean absorbance of test samples over the mean absorbance of
untreated samples x 100%.

ACDU colony assay for Ganciclovir resistance
Heterogeneously transduced Ba/F3 pools were washed and resuspended in PBS,
and sorted by flow cytometry. Cells were gated on the live cell population by
forward and side scatter on a FACS Vantage flow cytometer and sorted using a
single cell/well instrument setting on an automated cell deposition unit (ACDU) into
96 well plates containing 180 μl tissue culture medium supplemented with 20 %
FBS and with or without 10 μM Ganciclovir. Plates were incubated for 10 days at
37°C in a humidified, 5% CO2 tissue culture incubator and visually inspected for
colony growth. The cloning efficiency was calculated as the percentage of the
number of colonies observed over the total number of wells seeded. Ganciclovir
resistance was calculated as the percentage of number of colonies obtained under
Ganciclovir treatment over the number of colonies obtained without treatment
(untreated control). Ganciclovir resistant clones were expanded in the continuous
presence of 10 μM Ganciclovir and genomic DNA was extracted for further analysis.

In vivo leukemia challenge and Ganciclovir administration
For in vivo leukemia challenges, non-transduced Ba/F3 parental cells and
transduced Ba/F3 clones or pools were harvested, washed twice in Hanks Balanced
22
Salt Solution (HBSS) and resuspended in HBSS with 50 U/ml of heparin. Leukemia
challenges were administered by injecting 5000 live cells in 100 μl volume into the
tail vein of 6-8 week old mice. Ganciclovir was administered at a dose of 50 mg/kg
in 100 μl PBS as intraperitoneal injections once a day for five consecutive days
starting day 7 after challenge. Mice were monitored daily for general health and
appearance and euthanized by CO2 narcosis at the onset of disease or after 60
days of survival. Peripheral blood smears were obtained post mortem and stained
with Wright-Giemsa for the presence of leukemic blasts, and spleen weights were
recorded.

Histology
One Balb/c mouse challenged with 5x103 cK/TK-transduced Ba/F3 cells was not
treated with Ganciclovir and sacrificed at the onset of disease. Brain, kidney, liver
and spleen tissues were collected from the test animal and a naïve control animal,
fixed in 10% Neutral Buffer Formalin (10% NBF; Sigma) for 16 hours, washed with
H2O, dehydrated in 70% ethanol, embedded in paraffin, sliced and stained with
Hematoxylin & Eosin (H&E) Stain. All embedding, slicing and staining procedures
were performed by the Histocore Facility of the Research Immunology/BMT
Department at Childrens Hospital Los Angeles.

Southern blot
Twelve μg genomic DNA digested with the single cutter enzyme Nde-1 or Spe-1
were separated on an agarose gel by electrophoresis, blotted onto a nylon
membrane (Amersham, Buckinghamshire, England), hybridized with a radioactively
23
labeled probe spanning 1106 bp of the HSV-TK suicide gene, and exposed to
Kodak X-Omat film.

Polymerase chain reaction
For PCR to detect the previously reported common HSV-TK splicing event [19], we
used the following primers specific for the regions flanking the splice donor and
acceptor sites within the TK gene sequence: forward F5977 5’-GCA GCA AGA AGC
CAC GGA AG-3’ and reverse R6521 5’-CGA TGT GTC TGT CCT CCG G-3’. The
temperature protocol for 50 μl reactions containing 100-300 ng genomic DNA, 2.5
mM MgCl2, 5% DMSO and 0.25 U Taq Polymerase was as follows: 5 min at 95°C;
35 cycles for 45 sec at 95°C, 45 sec at 55°C and 45 sec at 72°C; 7 min at 72°C.

Sequencing of Ganciclovir-resistant TK genes
A nested set of PCR fragments was generated from the genomic DNA from one
Ganciclovir-resistant clone by using a high-fidelity Pfu Turbo® Polymerase
(Stratagene) and the following primer pairs: F5832 5’- GGA CGT GGT TTT CCT
TTG -3’ and R6193 5’- GCG GTG TTG TGT GGT GTA -3’; F6098 5’- CGC GAC
GAT ATC GTC TAC -3’ and R6568 5’- GCG GTC GAT GTG TCT GTC -3’; F6451
5’- CCT TAT GGG CAG CAT GAC -3’ and R6899 5’- GGA TAA AGA CGT GCA
TGG -3’; F6822 5’- ACG GCG ACC TGT ATA ACG -3’ and R7252 5’- CCC CCT
TTT TCT GGA GAC -3’. The PCR fragments were cloned into a pPCR
BlueScriptTM cloning vector using a cloning kit (Stratagene, Cedar Creek TX) and
sequenced with the T7 and T3 sequencing primers (AnaGen Sequencing, Atlanta
GA)
24
2.3 Results

Vector design and production
We designed and constructed three MND-based retroviral vectors (Fig. 2.1) that
were bicistronic for the murine cKit
D814V
oncogene (cK) and a suicide gene. The
suicide gene consisted of either the wild type HSV-TK sequence (TK), the
hypersensitive SR39 mutant (39), or splice corrected SR39 (sc), the latter
comprising the SR39 sequence in which we introduced two silent point mutations by
PCR-site directed mutagenesis to abrogate cryptic splicing as previously described
(Chalmers 2001). The enhancer/promoter element in the MND vector 5’-LTR drove

Fig 2.1: Schematic overview over retroviral MND vector constructs
The long terminal repeats (LTR) that flank the bicistronic coding regions in the MND
vector are based on the Moloney Murine Leukemia Virus sequence but harbor a
retroviral LTR U3 region enhancer/promoter (MND) and a primer binding site substitution
to minimize silencing of proviral gene expression. The LTR promoter drives the
expression of the murine cellular oncogene mcKitD814V (cK) followed by an internal
ribosome entry site (IRES) that promotes synthesis of the respective suicide gene
product. (A) The suicide gene in the MND-cK/TK vector is the wild type Herpes simplex
virus -type 1 thymidine kinase (TK). (B) The MND-cK/39 vector contains the
hypersensitive TK mutant sr39TK (39); the shaded area marks the 5 amino acid
substitutions in the active site. (C) The splice-corrected TK (sc) in the MND-cK/sp vector
is based on the sr39TK sequence with two silent point mutations in the cryptic splicing
donor and acceptor site (indicated by arrow heads) that abrogate aberrant splicing
events.
25
translation of the mcKit
D814V
protein in the conventional, Cap-dependent manner,
while the suicide genes were designed to be translated from an internal ribosome
entry site (IRES). All three vector constructs were pseudotyped with the MLV
ecotropic envelope and produced by transiently transfected 293T cells. Viral titers
were determined by transduction of murine 3T3 cells and analyis for copy number
by either TaqMan or Southern Blot (not shown).

Factor-independent growth of a clonal Ba-TK cell line
To determine whether expression of the c-Kit
D814V
oncogene in our construct was
sufficient to transform Ba/F3 leukemia cells as previously shown (Kitayama 1995),
Ba/F3 cells were transduced with MND-cK/TK at an MOI of ~1 and IL-3 was
withdrawn from the culture medium several days after transduction. From the
outgrowing cell population, a clonal cell line was obtained by single cell sorting
utilizing a flow-cytometry-based automated cell deposition unit (ACDU). A clonal Ba-
TK cell line and untransduced parental Ba/F3 cells were seeded at identical
numbers in the presence or absence of IL-3 and proliferation was monitored by
manual counting of viable cells under trypan blue exclusion up to 4 days after
seeding (fig 2.2a). Ba-TK cells without IL-3 treatment proliferated at a rate similar to
IL-3 treated transduced and untransduced cells, while parental Ba/F3 cells died
rapidly in response to IL-3 withdrawal. Transgenic cKitD814V expression was thus
sufficient to drive IL-3 independent proliferation.

26
Leukemogenesis and cell dose optimization and disease history in mice
Next we wanted to determine if Ba-TK cells were leukemogenic in mice and what
dosage would lead to 100% animal death within the first 3-4 weeks post-challenge.
In an escalating dose experiment (Fig2.2b), syngeneic immunocompetent Balb/c
mice were challenged with 1x10
3
, 1x10
4
or 1x10
5
Ba-TK cells or 1x10
5
parental
Ba/F3 cells as a negative control. As expected, untransduced Ba/F3 cells were not
leukemogenic even at the high dose given, presumably because systemic IL-3
levels were not sufficient to support in vivo survival of this cell line. In contrast, IL-3
independent Ba-TK cells led to leukemia development in a dose-dependent manner,
with a medium survival time of 19 days at the highest dose, 21 days at the medium
dose and 27 days at the lowest dose. Based on this pilot experiment, we set the
optimal leukemia cell dosage to 5x10
3
cells per injection for all future in vivo
experiments.

To have a better understanding of the disease history in challenged animals, one
mice injected with 5x10
3
Ba-TK mouse was sacrificed at the onset of disease on day
21 after challenge which was typically characterized by a ruffled appearance,
hunched posture and hind limb paralysis; an age-matched naïve mouse was
sacrificed as a control. Brain, kidney, liver and spleen tissues were fixed, sliced and
stained with Hematoxylin & Eosin (H&E) which stains nuclei in a blue/violet color,
cytoplasm in pink and red blood cells in orange/red (Fig 2.2c). Stained sections of
the brain and kidney looked similar between control and leukemia samples. In the
liver of the challenged animal, a slight increase of darker violet cells can be
observed compared to the control samples, marking cells with a high
27

Fig 2.2: Characterization of Ba/F3 leukemia
A) Growth Curve in vitro. Untransduced parental Ba/F3 cells (wt) or Ba/F3 transduced with
the MND-cK/TK vector were seeded at 1x10
5
cells/ ml at day 0 in the presence or absence
of 10 ng/ml mrIL-3 and counted on days 1, 2, and 4 after initial seeding under Trypan Blue
exclusion.
B) Dose optimization study in vivo. Immunocompetent Balb/c mice were challenged i.v. with
1x105 parental Ba/F3 cells or increasing doses of Ba-cK/TK clone #4 (1x103, 104, 105) and
observed for survival for 60 days after challenge (n=5).
C) History of leukemia in various tissue types. One Balb/c mouse challenged with 5x103 Ba-
TK.4 leukemia cells was sacrificed at the onset of disease (day 21 post-challenge); an age-
matched naïve Balb/c mouse was used as a control. Brain, kidney, liver and spleen tissue
from each animal was fixed and stained with H&E stain; 5x and 20x magnifications are
shown.
D) Optimization of Ganciclovir schedule. Balb/c mice were challenged with 5x103 Ba-TK
cells and treated i.p. with 50 mg/kg Ganciclovir (GCV) once a day on days 7-11, 14-18, 21-
25 after challenge or left untreated (not GCV control) and observed for event-free survival for
60 days (n=5).
28
A
0
5
10
15
20
25
30
35
01 24
Days in Culture
Viable Cell Number [10
5
/ml]
wt [+ IL3]
wt [- IL3]
cK/TK [+ IL3]
cK/TK [-IL3]
B
0
20
40
60
80
100
0 102030 40506
Days after Challenge
Survival [%]
0
10e5 Ba/F3 wt
10e3 Ba-cK/TK
10e4 Ba-cK/TK
10e5 Ba-cK/TK
C

D
0
20
40
60
80
100
0 102030 405060
Days after Challenge
Survival [%]
no GCV control
GCV day 7-11
GCV day 14-18
GCV day 21-25

0
100
200
300
400
500
600
700
800
900
1000
No GCV
Control
GCV day
7-11
GCV day
14-18
GCV day
21-25
Spleen Weight [mg]

29
nucleus/cytoplasm ratio that is characteristic for immature leukemic blasts. The most
drastic increase of leukemic presence, however, was observed in the spleen where
virtually the whole organ was filled with leukemia cells in the challenged animal. This
is in accordance with the known observation that the spleen as the largest
hematopoietic organ provides a supportive environment for leukemic expansion
(Shaked 2005).

Optimization of the in vivo GCV administration schedule
Initial in vitro cytotoxicity assays demonstrated that Ba-TK proliferation could be
inhibited in a dose-dependent manner when treated with increasing concentrations
of GCV (1 nM-100 μM; not shown). To establish a schedule of in vivo GCV
administration, Balb/c mice were challenged with 5x10
3
Ba-TK cells and treated with
50 mg/kg GCV, a dose comparable to the mean exposure to Ganciclovir in humans
after standard infusion, once a day for 5 consecutive days starting either day 7, 14,
or 28 after challenge (Fig 2d, left panel). None of the mice survived the challenge in
spite of GCV treatment, indicating that Ba-TK cells were not sensitive enough to the
administered dose to respond with complete regression. However, when spleen
weights were analyzed (Fig 2d, right panel) it became apparent that splenomegaly
caused by leukemia cell infiltration was reduced 2-fold in the early treatment group
(days 7-11 post-challenge) compared to the later treatment group (days 21-25) or
the challenged, but untreated control group (no GCV control). This indicated that the
greatest inhibitory effect on Ba-TK proliferation in vivo was expected to be achieved
with early GCV administration, and the treatment schedule chosen for all future
30
challenges was 50 mg/kg GCV i.p. on days 7-11 post challenge, based on this
experiment.

Differential response of clonal Ba/F3 cell lines expressing three different
suicide genes to Ganciclovir in vitro
Ba/F3 cells were transduced with all three vector constructs (cf. Fig 2.1) at an MOI
of ~1 and IL-3 was withdrawn from the culture medium several days after
transduction. Factor-independent cell pools were subjected to FACS-assisted single
cell sorting by an automated cell deposition unit (ACDU). Three monoclonal cell
lines derived from each Ba-TK, Ba-39 or Ba-sc population were tested for GCV
sensitivity in vitro by treating them with a wide range of GCV concentrations
(0.0001-100 μM) and measuring cell survival with a colorimetric assay. The
inhibitory dose IC
50
(leading to 50% of cell death) was found to be between 0.1-1 μM
for Ba-TK clones and about 10-fold lower for Ba-39 and Ba-sc cells, indicating that
cell lines expressing either one of the SR39 variants were more sensitive to
Ganciclovir than cells expressing wild type TK. The dose-response curves for
representative clones Ba-TK.4, Ba-39.2 and Ba-sc.2 that we chose for further
analysis are shown in Fig.2.3a. The viability of untransduced Ba/F3 cells
(supplemented with IL-3) was not affected even by high doses of GCV. The
presence of 1-2 proviral copies per cell was confirmed by Southern blot (Fig. 2.3b).

31
A
0
20
40
60
80
100
0 0.0001 0.001 0.01 0.1 1 10 100
GCV Concentration [ μM]
Viability [%]
Ba/F3 (+IL-3)
cK/TK
cK/39
cK/sc

B

C
0
20
40
60
80
100
0 1020 304050 6
Days after Challenge
Survival [%]
0
Ba/F3 (-)
cK/TK (-)
cK/39 (-)
cK/sc (-)
Ba/F3 (+)
cK/TK (+)
cK/39 (+)
cK/sc (+)
GCV
↓↓↓↓↓

Fig 2.3: In vitro and in vivo sensitivity of transduced Ba/F3 clones to Ganciclovir.
(A) Cytotoxicity Assay. Monoclonal Ba/F3 populations stably transduced with MND-cK/TK
( □), MND-cK/39 ( ∆) or MND-cK/sp ( ●) were seeded in triplicate in 100 μl RPMI with
10%FBS in a 96 well plate at 1 x10
4
cells per well and treated with the indicated
concentrations of Ganciclovir for five days. As a control, non-transduced Ba/F3 cells
supplemented with 8 ng/ml murine recombinant IL-3 ( ▬) and treated with Ganciclovir
were included in the experiment. Wells were incubated with CellTiter 96 dye for 4h and
absorbance was measured at 450 nm. Percent cell viability is shown as the percentage of
mean absorbance of treated cells over the mean absorbance of untreated (0 μM GCV)
cells. (B) Southern Blot. Genomic DNA from parental Ba/F3 cells (Ba) or the Ba/F3
clones Ba-TK.4, Ba-39.2 and Ba-sc.2 was digested with the single cutter enzymes Nde-1
or Spe-1 and hybridized with a 1.1 kb probe spanning the entire coding region of the
suicide gene. (C) Survival Plot. Immunocompetent, syngeneic Balb/c mice were
challenged intravenously with 5,000 live Ba/F3 leukemia clones and either left untreated
((-), open symbols) or treated with 50 mg/kg GCV ((+), filled symbols) intraperitoneally for
five consecutive days beginning one week after challenge. Mice were monitored for
survival for 60 days post challenge and euthanized at the onset of disease.
32
We subsequently challenged 10 mice each with 5,000 cells of either the Ba/F3
clones Ba-TK.4, Ba-39.2, and Ba-sc.2 or untransduced Ba/F3 cells given
intravenously on day 0 (Fig. 2.3C). Five mice of each cohort were treated with 50
mg/kg GCV i.p. once a day on 5 consecutive days (days 7-11 post-challenge); the
five remaining mice of each cohort were left untreated. All transduced cell lines were
highly leukemogenic leading to 100% mortality between days 19-27 and seemed not
to be rejected by the recipients’ immune system. Mice challenged with untransduced,
parental Ba/F3 cells did not develop leukemia and survived the entire observation
period of 60 days, which demonstrates that leukemogenicity was strictly dependent
on vector integration. GCV treatment of mice challenged with Ba-TK.4 prolonged
survival for about a week, but all mice in this cohort eventually succumbed to
disease. In contrast, GCV treatment of mice given Ba-39.2 and Ba-sc.2 cells
protected 100% (5/5, respectively) of challenged animals. This experiment shows a
superiority of clonal cell lines expressing the hypersensitive TK variants SR39 and
sc39 to respond to clinically relevant systemic GCV levels in vivo, compared to cells
expressing the wild type HSV-TK.

Unresponsiveness of clonal BaF/3 cell lines to Foscarnet
To further evaluate the clinical feasibility of suicide gene therapy for leukemia, we
wanted to address whether patients receiving transduced autologous grafts
expressing HSV-TK may be limited in treatment options for Herpes infections such
as CMV, Herpes virus type 1 or 2 or Varicella zoster. These patients cannot be
treated with antiviral nucleoside analogs such as Ganciclovir or the related drug
Acyclovir, because in addition to infected cells, gene-corrected graft cells will also
33
be eliminated. We hypothesized that Foscarnet, an organic analog of inorganic
pyrophosphate that targets viral DNA polymerase rather than the thymidine kinase,
and which is often used to treat Ganciclovir and Acyclovir-resistant viral strains, will
selectively eliminate infected cells without affecting graft cells. To test this
hypothesis, clonal Ba/F3 cell lines expressing wt TK, sr39 or splice corrected 39
were treated with increasing concentrations of Ganciclovir, Acyclovir and Foscarnet,
and the percent viability compared to untreated controls was measured (Fig 2.4).

Fig 2.4: Dose response of Ba/F3 clones to Ganciclovir, Acyclovir and Foscarnet
Untransduced parental Ba/F3 cells (A) or the clonal cell lines Ba-TK.4 (B), Ba-39.2 (C)
and Ba-sc.2 (D) were seeded on 96 well plates at 1 x104 cells per well in triplicate, and
cells were treated with the indicated concentrations of Ganciclovir (black squares),
Acyclovir (gray triangles) or Foscarnet (white circles) for 5 days. CellTiter96 dye was
added for 4 hours and absorbance was measured at 450 nm. Percent cell viability was
calculated as the percentage of the mean absorbance of treated cells over the mean
absorbance of untreated (0 μM GCV) cells.

34
Untransduced parental Ba/F3 cells not expressing HSV-TK showed some
nonspecific cytotoxicity at high concentrations of Ganciclovir and Acyclovir but
otherwise showed no dose-response. All of the other cell lines generally were more
sensitive to Ganciclovir than Acyclovir which has been previously reported, and
again, cells expressing the hyperactive TK variants SR39 and sc-39 were eliminated
at lower drug concentrations compared to the wt TK. Importantly, all cell lines
including transduced clones survived in the presence of Foscarnet, even at high
doses, indicating that Foscarnet treatment may be used to treat herpes virus
infections in patients who have engrafted cells that express HSV-TK or one of its

In vitro drug sensitivity of transduced Ba/F3 pools
The use of cell clones pre-characterized to be Ganciclovir sensitive in vitro neglects
the possible occurrence of suicide escape mutants that may develop within a
heterogeneously transduced pool due to vector rearrangements, splicing events or
epigenetic silencing. Transduced pools (cryopreserved prior to the single cell
sorting that resulted in the clonal cell lines used in our previous experiments) were
thawed and similarly treated in vitro with a range of Ganciclovir concentrations for 5
days. Fig. 2.5A shows that pools expressing the SR39 or sc39 suicide gene were
more sensitive to Ganciclovir treatment than pools expressing the wild type HSV-TK
gene, whereas untransduced control cells showed cell death only at high drug
concentrations of 100 μM. These data indicate that at least during short-term drug
treatment, transduced pools were eradicated as efficiently as cell clones because
the IC
50
values were found to be similar between pools and clones containing the
same vector (Figs. 2.3A and 2.5A).
35

Fig 2.5: In vitro and in vivo sensitivity of transduced Ba/F3 pools to Ganciclovir
(A) Cytotoxicity Assay. Heterogeneously transduced Ba/F3 pools expressing MND-cK/TK ( □),
MND-cK/39 ( ∆) or MND-cK/sp ( ●) were seeded in triplicate in 100 μl RPMI with 10%FBS in a
96 well plate at 1 x104 cells per well and treated with the indicated concentrations of
Ganciclovir for five days. As a control, non-transduced Ba/F3 cells supplemented with 8
ng/ml murine recombinant IL-3 ( ▬) and treated with Ganciclovir were included. Wells were
incubated with CellTiter 96 dye for 4h and absorbance was measured at 450 nm. Percent
cell viability is shown as the percentage of mean absorbance of treated cells over the mean
absorbance of untreated (0 μM GCV) cells.
(B) Southern Blot. Genomic DNA from the transduced Ba/F3 pools Ba-TK.p1, Ba-39.p1 and
Ba-sc.p1 was digested with the multiple cutter enzyme HindIII. The released 1.2 kb band
was hybridized with a 1.1 kb probe spanning the entire coding region of either suicide gene.
As a control, genomic DNA from a clone with two proviral integrants was mixed with DNA of
untransduced Ba/F3 cells at various ratios to obtain a copy number standard of 0.5, 1 and 2
copies per cell.
(C, D) Survival Plots. Immunocompetent, syngeneic Balb/c mice were challenged
intravenously with 5,000 live Ba/F3 of (C) pool 1 or (D) pool 2 or untransduced parental
Ba/F3 cells and either left untreated (-) or treated i.p. with 50 mg/kg GCV (+) for five
consecutive days one week after challenge. Mice were monitored for survival for 60-90 days
post challenge and sacrificed at the onset of disease. (C) Plot shows summarized data from
two independent experiments (total n=10 per experimental arm). (D) Splenocytes from
animals TK(+).2 and 3 (*), 39(+).1 (§) and sc(+).1 (#) were isolated immediately post mortem
and grown in tissue culture.
36
A
0
20
40
60
80
100
0.001 0.01 0.1 1 10 100
Canciclovir Concentration [ μM]
Cell Viability [%]
Ba/F3 wt
Ba-TK.p
Ba-39.p
Ba-sc.p
B

C
0
20
40
60
80
100
0 102030 40 50 6
Days after Challenge
Survival [%]
0
Ba/F3 (-)
cK/TK (-)
cK/39 (-)
cK/sp (-)
Ba/F3 (+)
cK/TK (+)
cK/39 (+)
cK/sp (+)

D

37
Analysis of the genomic DNA of transduced pools for viral integrants by Southern
Blot using an enzyme which cuts at least twice in the proviral sequence revealed
that all transduced pools grown in the absence of exogenous IL-3 harbored on
average between 1-2 copies per cells (Fig. 2.5B).

Transduced Ba/F3 pools and in vitro Ganciclovir resistance
To analyze the frequency of Ganciclovir resistance in transduced Ba/F3 pools in
vitro, we subjected the same pools that were used to derive clonal cell lines (now
referred to as pool 1 or p
1
) to another round of ACDU single cell sorting into medium
either lacking or containing 10 μM Ganciclovir. The percentage of Ganciclovir
resistant clones obtained from each pool is summarized in Table 1. While we did not
observe resistant clones in the Ba-TK.p
1
or Ba-39.p
1
pools, approximately 2.5% of
the cells from the splice corrected Ba-sc.p1 pool grew under treatment condition. As
a positive control for our ACDU Ganciclovir resistance assay, we included a GCV-
resistant Ba/F3 culture (Ba-39.p
1
-R) that was bulk-selected as a whole pool with 10
μM GCV for 21 days prior to ACDU sorting; over 80% of this population was
observed to grow in the presence of GCV in the sorting experiment compared to
untreated plates. In addition, we freshly transduced Ba/F3 cells with all three MND
vector constructs and selected once more for IL-3 independent Ba/F3 pools
(referred to as pool 2 or p2), and the ACDU Ganciclovir resistance assay was
repeated. In this experiment, about 14% of Ganciclovir-resistant outgrowth was
observed for Ba-TK.p2, 1% for Ba-39.p2 and around 5% for Ba-sc.p2. All these cells
presumably contain a viral integrant because no colonies were found if parental
Ba/F3 cells were cultured in the absence of IL-3, indicating that spontaneous IL-3
38
independent cell growth of untransduced cells was probably not responsible for
colony growth in heterogeneously transduced pools.

Table 1. In vitro Ganciclovir-resistant clones from Ba/F3 pools

Sorted Cells % Resistant Clones
a
% Cloning Efficiency
b
Ba/F (-IL3) 0.00 (0/0) 0.00 (0/384)
Ba/TK pool 1 0.00 (0/57) 9.90 (57/576)
Ba/39 pool 1 0.00 (0/201) 34.90 (201/576)
Ba/sc pool 1 2.47 (4/162) 28.13 (162/576)
ACDU #1
Ba/sc-R pool 1 83.33 (25/30) 5.21 (30/576)
Ba/F (-IL3) 0.00 (0/0) 0.00 (0/96)
Ba/F (+IL3) 97.87 (92/94) 48.96 (94/192)
Ba/TK pool 2 13.93 (28/201) 69.79 (201/288)
Ba/39 pool 2 1.10 (2/181) 62.85 (181/288)
ACDU #2
Ba/sc pool 2 4.62 (9/195) 67.71 (195/288)
a
number of clones obtained under treatment with 10 μM GCV over number of clones
obtained without treatment
b
number of clones obtained without treatment over total number of wells seeded

Survival of mice challenged with Ba/F3 pools with or without Ganciclovir
treatment
Immunocompetent, syngeneic Balb/c mice were challenged with 5,000 live Ba/F3
leukemia cells from pool 1 (Fig. 2.5C). Similarly to Ba/F3 clones used in the
previous in vivo challenge, all transduced Ba/F3 pools were found to be highly
leukemogenic in mice in the absence of Ganciclovir treatment, whereas a control
group injected with parental Ba/F3 cells survived the challenge. When treated with
50 mg/kg Ganciclovir for five consecutive days one week after challenge, mice with
39
leukemia cells expressing wt TK showed prolonged survival, but 70% eventually
succumbed to the disease. Balb/c mice challenged with Ba-39.p
1
pools were 100%
protected from leukemia development after Ganciclovir treatment. However, Ba-
sc.p
1
cells which showed about 2.5% resistance in vitro were not eliminated by
Ganciclovir administration in vivo, and 70% of the animals expired almost as rapidly
as the untreated group.

When the in vivo challenges were performed with transduced Ba/F3 cells from pool
2 (Fig. 2.5D), Ganciclovir treatment significantly prolonged the survival of mice
challenged with TK expressing cells (Ba-TK(+)), but the survival rate eventually
declined to 20% over a time period of 60 days. The cause of death was
development of leukemia as confirmed by splenomegaly and presence of leukemic
blasts in the peripheral blood (not shown). In contrast, in both groups challenged
with hyperactive TK variants and treated with Ganciclovir (Ba-39(+) and Ba-sc(+)),
the survival rate did not drop below 80% and the majority of animals was protected
from leukemia development. We were able to isolate splenic leukemia cells from
Ganciclovir treated, moribund animals (animals TK(+).2, TK(+).3, 39(+).1, and
sc(+).1) and re-grow them in tissue culture for further analysis.

Vector integrity analysis of GCV-resistant clones from pool 1
Ganciclovir resistant clones obtained from the Ba-sc.p
1
pool were isolated for further
analysis as summarized in Table 2. We first analyzed genomic DNA for potential
splicing events around the previously described cryptic splice sites using the PCR
primers F5799-R6521 specific for regions flanking the corrected splice sites within
40
the vector sequence (Fig. 2.6A). The transgene was present and unspliced in all
independently isolated clones (Fig. 2.6B). We also performed a Southern Blot
analysis with a single cutter enzyme and a DNA probe spanning the full-length
SR39 gene (Fig. 2.6C). Comparing the unique integration sites of the vector, we
found that all Ganciclovir-resistant clones and even the bulk-selected pool (not
shown) were of monoclonal origin, i.e. a mutant vector had transduced a single cell
that survived Ganciclovir selection and expanded within the population to be re-
isolated several times from the Ba-sc.p
1
pool.

C

Fig 2.6: Vector integrity analysis of GCV-R clones obtained from pool 1
(A) Schematic diagram showing the location of the 545 bp PCR product flanking the cryptic
splice donor and acceptor sites (upper row) and enzyme restriction sites plus 1106 bp
probe for Southern Blotting (lower row). (B) PCR products of genomic DNA isolated from
Ba/sc-R clones 1-8 using primers F5977 and R6521.Positive control (pos): Ba/TK.4 DNA;
negative control (neg): Ba/F3 DNA; p1: DNA from Ba-sc pool sc.p1. (C) Southern Blot.
Genomic DNA from Ba/F3 cells (wt) Ba/sc-R clones 1-8 was digested with the single cutter
enzyme Nde-I (lanes 1-9); in addition, DNA from wt cells and clones 1 and 6 was digested
with the single cutter enzyme Spe-I (lanes 10-12). Bands were hybridized with a 1.1 kb
probe specific for the sc39 gene.

To determine possible vector rearrangements or mutations we amplified overlapping
DNA fragments from the integrated sc39 gene for DNA sequence analysis and
discovered a single base pair insertion towards the 3’ end of the sc39 gene. At the
41
amino acid level, expression of this mutant transgene gives rise to a full-length
protein but with the last 50 amino acids of the 350 aa full length sequence altered
due to the frameshift the insertion causes (not shown).

Vector integrity analysis of GCV-resistant clones from pool 2
In contrast to pool 1 where we obtained Ganciclovir-resistant colonies only from the
Ba-sc.p1 population, all transduced populations from pool 2 gave rise to Ganciclovir-
resistant clones, albeit at different frequencies (Table 1). PCR analysis of genomic
DNA from isolated clones using the primers flanking the cryptic splice sites revealed
the specific 545 bp band in only four out of 20 Ganciclovir-resistant Ba-TK.p
2
-R
clones and none in any of the Ba-39.p
2
-R or BA-sc.p
2
-R clones, indicating that some
vector rearrangements must have occurred (Fig. 2.7A). (Refer to table 2 for a
comparison of vector analysis between pool 1 and 2).

To further determine whether the loss of the specific gene product in clones isolated
from pool 2 in vitro was due to splicing events, we performed PCR reactions with a
nested set of four primer pairs that covered the whole suicide gene sequence (Fig.
2.7B). Of the 20 Ba-TK.p
2
-R clones, we chose to analyze four clones with the 545
bp fragment missing (TK.p
2
-R.1-4) plus the four clones with the fragment intact
(TK.p2-R.14, 18-20). The Ba-TK.p2-R clones 1-4 and both Ba-39.p
2
-R clones were
completely negative for any of the nested PCR products (Fig. 2.7C); thus the whole
suicide gene must have been lost in these cells. The TK clones Ba-TK.p
2
-R.14 and
18-20 must have retained the full TK sequence because specific bands were
detected with all four nested PCR primer sets. Three out of eight sc-R clones were
42
completely negative for all PCR sequences, but five clones showed a clear 430 bp
band for the last nested PCR primer pair. This suggests that splicing had taken
place within the scSR39 sequence somewhere upstream of last nested primer set
(d); whether any of the (corrected) cryptic splice sites were involved, however, is
inconclusive from these data.

Table 2. Evaluation of Ganciclovir resistant clones isolated from transduced Ba/F3
pools in vitro and Ganciclovir-resistant populations recovered from leukemic mice in
vivo

Ba/F3 Pool Isolated Clones
PCR
F5799-R6521
PCR
a, b, c, d
Southern
Blot
Other
In vitro:
TK pool 2 TK.p
2
-R.1-13, 15-17 Absent Absent a-d Short
fragment

TK.p
2
-R.14,18-20 Intact Intact a-d Full-length
and short

SR39 pool 2 39.p
2
-R.1-2 Absent Absent a-d Short
fragment

sc pool 1 sc.p
1
-R.1-8 Intact N.D. Monoclonal 1 bp
insert
sc pool 2 sc.p
2
- R.1-3, 5, 7 Absent Absent a-c Full-length
and short

sc.p
2
-R.4, 6, 8 Absent Absent a-d Short
fragment

In vivo:
TK pool 2 Mouse # TK(+).2, 3 Intact Intact a-d N.D. IC
50
=
1-3
μM

SR39 pool 2 Mouse # 39(+).1 Absent Absent a-d N.D. IC
50
=
30
μM

sc pool 2 Mouse # sc(+).1 Absent Absent a-d N.D. IC
50
=
10
μM

N.D. = not determined

43
Since large parts of the TK suicide genes seemed to be missing in many of the
Ganciclovir-resistant clones from pool 2, we wanted to determine whether the
cKit
D814V
oncogene was still detectable in the genomic DNA of those clones. We
performed a Southern Blot analysis of selected resistant clones from pool 2 (Ba-
TK.p
2
-R.6, 7, 19, 20, Ba-39.p
2
-R.1,2, and Ba-sc.p
2
-R.5-8) using the restriction
enzyme Xba-1 that cuts at least twice in the proviral DNA and a probe that
hybridizes to the cKit sequence. Fig. 2.7D shows a small 2 kb band that represents
a fragment derived from the endogenous murine cKit gene which is also detected in
the negative control lane loaded with DNA from untransduced Ba/F3 cells. The
negative control however lacks a larger 6 kb band which represents the proviral
insert as shown in the positive control lane loaded with Ba-TK.4 DNA. All samples of
Ganciclovir-resistant clones probed positive for the cKit transgene which was not
unexpected because these cells grew IL-3 independently and spontaneous IL-3
independent growth in the absence of vector transduction was never observed in
vitro (Table 1). For some of the clones that showed missing suicide gene sequences
by PCR (Ba-TK.p
2
-R.6,7, Ba-39.p
2
-R.1,2, and Ba-sc.p
2
-R.6) the restriction fragment
on the Southern Blot was smaller than the expected 6 kb band size, confirming that
parts of the vector insert must have been deleted.

44
A B

C

D

Fig 2.7: Vector integrity analysis of GCV-R clones obtained from pool 2
(A) PCR products of genomic DNA isolated from GCV resistant clones TK-R.1-20, 39-
R.1,2 and sc-R.1-8 using primers F5977 and R6521.Positive control (+): Ba/TK.4 DNA;
negative control (-): Ba/F3 DNA; pools: DNA from transduced pools p
2
. (B) Schematic
diagram showing the location of the nested PCR primer sets a: F5382/R6193;
b:F6098/R6568; c: F6451/R6899; and d: F6822/R7252. (C) PCR products of genomic
DNA isolated from selected GCV resistant clones as in (A) using the nested PCR primers
shown in (B). Positive control (+): Ba/TK.4 DNA; negative control (-): Ba/F3 DNA.
(D) Upper panel: Schematic figure of the location of Xba-1 restriction sites and the 462bp
cKit probe; the HSV-TK vector is missing the additional Xba-1 upstream of the cKitD824V
gene and released bands are expected to be larger. Lower panel: Southern blot of
genomic DNA isolated from Ganciclovir-resistant clones (Pool 2), digested with the double
cutter Xba-1 and hybridized with a 462 bp cKit probe; negative control (-): Ba/F3 DNA;
positive control (+): Ba-TK.4 DNA. The probe labels endogenous cKit at a band size of 2
kb.

45
Analysis of re-isolated GCV-resistant leukemia cells from mice for vector
integrity
Drug resistant leukemia cells were isolated from Ganciclovir treated, moribund
animals challenged with heterogeneously transduced Ba/F3 cells from pool 2 mice
(TK(+).2, TK(+).3, 39(+).1 and sc(+).1; Fig 2.5D) and re-grown in tissue culture for
further analysis. A PCR reaction performed with the primers flanking the cryptic TK
splice sites showed that the specific sequence was detectable in genomic DNA
isolated from both TK(+) challenged animals but not the 39(+) or sc(+) challenged
mice (Fig. 2.8A). Fig. 2.8B shows a PCR analysis using our nested PCR primer sets
(Fig. 7B) that reveals the presence of full length TK sequence in both TK(+).2 and
TK(+).3 populations, but not in cells isolated from mice 39(+).1 or sc(+).1. The PCR
patterns correlate with some of the single cell clones analyzed in Fig. 2.7C, (e.g.
Ba-TK.p
2
-R.14, 18-20, Ba-39.p
2
-R.1,2 and Ba-sc.p
2
-R.1-3; see also Table 2), but it
has to be noted that cells isolated from mice with leukemia are not necessarily
monoclonal and could have been of oligoclonal origin.

To determine whether the re-isolated cells lost their sensitivity to Ganciclovir, an in
vitro cytotoxicity assay was performed (Fig. 2.8C). Previously characterized
Ganciclovir-sensitive clones (as in Fig. 2.3A) were used in parallel as a positive
control and showed the characteristic dose-response curve with an IC
50
of less than
0.01 for Ba-39.2 and Ba-sc.4 clones and less than 0.1 for Ba-TK.4 clones (left
panel). In contrast, the re-isolated cells from mice TK(+).1,2, 39(+).1 and sc(+).1 all
had an IC
50
in the range of 1-30, which is approaching the value used to define viral
resistance in clinical CMV isolates (IC
50
>12 μM) (right panel). Overall, the GCV
46
sensitivity of cells re-isolated from mice that developed leukemia despite GCV
administration was greatly diminished compared to clones expressing functional
suicide genes, regardless of whether cells were transduced with wt TK or the
hyperactive mutants.

A B

C
0
20
40
60
80
100
0.001 0.01 0.1 1 10 100
Ganciclovir Concentration [uM]
Cell Viability [%]
TK(+).2
TK(+).3
39(+).1
sc(+).1

0
20
40
60
80
100
0.001 0.01 0.1 1 10 100
Ganciclovir Concentration [uM]
Cell Viability [%]
Ba/F3 wt
TK.4
39.2
sc.2

Fig 2.8: In vivo challenge with transduced pools 2 and analysis of re-isolated
GCV-resistant populations
(A) PCR analysis of re-isolated GCV-resistant populations TK(+).2 and 3 (*), 39(+).1
(§) and sc(+).1 (#) using the splice site flanking PCR primers F5977 and R6521.
Positive control (+): Ba/TK.4 DNA; negative control (-): Ba/F3 DNA. (B) PCR analysis
of re-isolated GCV-resistant populations TK(+).2 and 3 (*), 39(+).1 (§) and sc(+).1 (#)
using the nested PCR primer sets a: F5382/R6193; b:F6098/R6568; c: F6451/R6899;
and d: F6822/R7252. Positive control (+): Ba/TK.4 DNA; negative control (-): Ba/F3
DNA. (C) Cytotoxicity Assay. Left panel: re-isolated TK(+).2,3, 39(+).1 and sc(+).2
populations were seeded in triplicate in 100 μl RPMI with 10%FBS in a 96 well plate at
1 x104 cells per well and treated with the indicated concentrations of Ganciclovir for
five days. Wells were incubated with CellTiter 96 dye for 4h and absorbance was
measured at 450 nm. Percent cell viability is shown as the percentage of mean
absorbance of treated cells over the mean absorbance of untreated (0 μM GCV) cells.
Right panel: Untransduced Ba/F3 cells ( ▬; supplemented with IL-3) and Ba/F3 clones
expressing MND-cK/TK ( □), MND-cK/39 ( ∆) or MND-cK/sp ( ●) were treated with the
same range of GCV concentrations as an intra-experimental control.

47
2.4 Discussion

In the wake of the initial observation of insertional oncogenesis in a clinical trial, one
suggestion made to potentially eradicate transformed cells was to include a suicide
gene in the vector with the therapeutic gene. While the ability of the HSV-TK
suicide gene approach to induce cytotoxicity had been widely demonstrated in
transgenic mice, with tumors and donor lymphocytes (Moolten 1994; Ram 1997;
Cohen 1997; Bonini 1997; Wildner 1999; Tiberghien 2001; Satoh 2005) , the
efficacy in eradicating leukemia cells growing under transcriptional control of the
vector had not been studied. We produced bicistronic retroviral vectors carrying
variants of the HSV-TK suicide gene and a transforming oncogene, that transform a
pre-B cell leukemia line to become factor-independent (Kitayama 1995) and cause a
reproducible murine model. The goal of our study was to determine whether
Ganciclovir treatment would be sufficient to eradicate rapidly dividing HSV-TK
transduced leukemia cells in a novel murine model for insertional oncogenesis.

Our results show that Ganciclovir treatment had a protective effect on mice
challenged with transduced leukemia cells expressing either one of the
hypersensitive TK mutants used in our study (SR39 or sc39), but not with
transduced leukemia cells expressing the wild type HSV-TK gene. Cells expressing
the SR39 and sc39 mutants were over 10-fold more sensitive to Ganciclovir than
cells expressing the wild type HSV-TK protein; these findings imply that the
systemic levels of Ganciclovir after the in vivo administration regimen that was used
may be sufficient to eradicate cells with hypersensitive TK variants but not high
48
enough to eradicate cells with wild type HSV-TK. The I.V. dose of 50 mg/kg in mice
is comparable to the mean exposure to Ganciclovir in humans based on AUC
values following a standard 5 mg/kg single I.V. infusion during the maintenance
phase of antiviral therapy for cytomegalovirus infections (source: Roche
Laboratories 2000). We therefore propose that the use of a hypersensitive TK
suicide gene is a reasonable approach for selective ablation of insertional
oncogenesis and superior to the use of wild type HSV-TK. Moreover, we found that
treatment of transduced cells with an alternative antiviral agent, Foscarnet, did not
affect survival of HSV-TK gene modified cells; therefore the loss of the therapeutic
graft due to the clinical need to administer antiviral treatment for Herpes virus
infections could be circumvented by the use of Foscarnet instead of Ganciclovir.

One of the major drawbacks of the HSV-TK suicide system is the potential for
development of drug resistance. Mutations in the thymidine kinase gene of Herpes
viruses occur naturally in patients undergoing prolonged Ganciclovir treatment and
lead to resistance to anti-viral drug therapy (Morfin 2002). Mutations of HSV-TK
sequences have also been documented in various suicide gene models (Garin
2002; Frank 2004). In order to assess the occurrence of escape mutants in our
experimental model, we switched from using well-defined transduced cell clones to
heterogeneously transduced Ba/F3 pools. An in vitro single-cell cloning assay
resulted in the outgrowth of Ganciclovir-resistant clones only from the pool
generated by transduction with sc39, and molecular analysis revealed a monoclonal
origin of the Ganciclovir resistant cells in the sc39-transduced pool, i.e. only a single
aberrant viral transcript integrated into a target cell genome and the resulting clone
49
expanded under selective pressure in culture and was re-isolated independently
several times. Mice challenged with this pool harboring the resistant sc39
expressing sub-clone died rapidly of leukemia, despite Ganciclovir treatment. The
sc39 transgene in the resistant clone was of full length and not spliced, but DNA
sequencing showed a single T nucleotide insertion at a TTT repeat towards the 3’
end of the gene (data not shown) that leads to randomization of the last 50 amino
acids of the sc39 kinase. The mutation occurred not within the catalytic core of the
protein but upstream of the codon for aa336 which has been identified as a hot spot
for TK mutations in clinical HSV isolates associated with Acyclovir-resistance
(Morfin 2002). Thus it is likely that protein function is diminished by the single bp
insertion we identified, though this was not formally established for instance by
reconstruction of a vector containing the mutant sc39 gene and dose-inhibition
experiments.

In our second transduction experiment (Pool 2), we found Ganciclovir-resistant
clones in all three populations transduced with vectors containing either TK, SR39
or sc39. Pool 2 was transduced at a higher MOI than Pool 1 so that the probability
of picking up aberrant vector transcripts may have been increased. The frequency of
in vitro Ganciclovir resistance was relatively low for the SR39 vector (1.10%) or the
sc39 vector (4.62%) compared to HSV-TK (13.93%), and in vivo challenges with
these leukemia cell pools expressing either one of the hypersensitive mutants still
led to 80% survival with Ganciclovir treatment. Analysis of the integrity of the vector
provirus in clones selected for in vitro Ganciclovir resistance suggested an
oligoclonal origin of the escape mutants with redundant isolation. Thus, the
50
frequency of suicide gene escape vector variants is probably over-estimated by the
measured frequencies of resistant clones. The results further indicated that drug
resistance may occur by several mechanisms as some clones still contained full
length inserts whereas other clones contained truncated versions of the suicide
genes or were lacking the entire thymidine kinase sequence.

We produced a splice-corrected version of the SR39 gene to eliminate a cryptic
splicing event that has been commonly observed in retroviral vectors with the HSV-
TK gene, using site-directed mutagenesis of key base pairs which has been shown
to abrogate the cryptic splicing (Chalmers 2001). Interestingly, we did not observe
this splicing event in either the wild-type HSV-TK or the SR39 genes using PCR to
amplify the relevant region, and therefore it is not evident whether the mutagenesis
of the splice donor and acceptor in the SR39 gene had any benefit. Instead, several
patterns of deletions of the different HSV-TK genes were seen, suggesting that
splicing may occur at multiple sites, which may differ for different transgenes, vector
backbones, promoters, and elements such as the internal ribosome entry site
(IRES).

A recent study suggested that HSV-TK expressing lymphoma cells injected into
mice resulted in GCV-resistant relapses due to genetic and epigenetic alteration of
the retroviral insert (Frank 2004). In that model, the EL-4 lymphoma cells were fully
transformed prior to vector transduction and thus could remain tumorigenic even
with complete loss or epigenetic silencing of the retroviral vector and hence suicide
gene expression. However, proliferative clones resulting from insertional
51
oncogenesis are unlikely to completely lose or silence their vector insert because
clonal proliferation strictly depends on the continual transcriptional activity of the
provirus. We show in our experimental model of insertional oncogenesis by
Southern Blot analysis that the cKit
D814V
transgene remained present in all
Ganciclovir- resistant clones because malignant proliferation strictly depended on
expression of the oncogene. Thus, the model of insertional oncogenesis described
here may better reflect the challenges to elimination of transformed clones that a
suicide gene approach would face in the clinical setting.
The frequency of naturally-occurring insertional oncogenesis has been reported to
be low using replication-incompetent retroviral vectors (Li 2002; Woods 2006;
Montini 2006). The frequency of drug escape in transduced leukemia cells was also
found to be low in our murine model, in the range of 1-15% as an upper limit. Thus,
in most potential cases of insertional oncogenesis by a vector carrying a therapeutic
transgene and a suicide gene, the suicide gene would be intact and the clone would
be sensitive to Ganciclovir. Overall, we find that the use of an improved,
hypersensitive HSV-TK variant is sufficient to eradicate rapidly dividing leukemia
cells with clinically relevant intravenous dosages of Ganciclovir and suicide gene
ablation of insertional oncogenesis in clinical gene therapy trials is a feasible, if
imperfect, approach.
52
Chapter 3- IL-12 Adjuvant Therapy for Leukemia Cell
Vaccination

3.1- Introduction

Interleukin-12 (IL-12) as an adjuvant for ALL vaccination
IL-12 is a proinflammatory cytokine that has been shown to reverse immune
tolerance into specific immunity against cancer. It is produced by mature dendritic
cells that have been stimulated by CD4 T cells through CD40/CD40L interactions or
by pathogens and viruses through CD4-independent pathways. IL-12 drives CD4 T
cell differentiation towards the T
H
1 phenotype that supports a cell-mediated immune
response and directly activates CTLs, NK cells and NKT cells. Stimulation of naïve
CD8 T cells through the T cell receptor (TCR) and costimulatory receptors leads to
limited proliferation and may result in tolerance, and third signal cytokines like IL-12
or IFN- α are necessary to induce clonal expansion, full effector function, long term
survival and immunologic memory (Mescher 2006).

The major mechanism of tumor rejection is cytotoxic lysis of the malignant cells.
Given its strong activation of specific cellular immune responses, recombinant
human IL-12 protein has been first studied as a single agent for cancer treatment
with limited success because of the severe toxicity of high systemic doses of IL-12
that unexpectedly resulted in deaths in a phase II clinical trial (Leonard 1997).
However, IL-12 as an adjuvant either co-administered with cellular vaccines or
delivered locally by transduced tumor cells could eliminate pre-established tumors in
53
some animal models (Fuji 1999; Adris 2000; Nanni 2001; De Giovanni 2004) and
induced specific CTL responses in advanced stage melanoma patients (Lee 2001).

In our murine model of Ph
+
ALL it was previously shown that systemic IL-12 protein
administration elicited a potent antileukemic immune response (Gruber 2005).
Recombinant IL-12 alone was sufficient to eradicate pre-existing leukemia, but mice
were not protected from subsequent challenges. Immunological memory was only
established when IL-12 was given in combination with irradiated TripleVax but not
wt BM185 cells, indicating that the delivery of tumor-specific antigen plus co-
stimulatory molecules is necessary for the development of memory cells in this
tumor model.

My studies in continuation of this project were to determine whether local delivery of
IL-12 by transduced BM185 cells could effectively induce rejection of mouse Ph
+
ALL
while avoiding the administration of potentially toxic doses of recombinant IL-12.
The two specific aims of my studies were a) to determine the optimal retroviral
vector design for IL-12 gene delivery and b) to compare the efficacy of IL-12
transduced cell vaccines (IL-12Vax) to systemic IL-12 administration, TripleVax and
combinations thereof.

The vector design for IL-12 gene therapy has to be carefully addressed to maximize
effective cytokine secretion. IL-12 is a heterodimer consisting of two disulfate-linked
subunits, p35 and p40. The p35 subunit is constitutively expressed in vivo while p40
expression is tightly regulated. Over-expression of p40 can result in the secretion of
54
monomeric p40 and p40 homodimers which both have been proposed as natural
inhibitors of IL-12 (Gillessen 1995). A retroviral vector for IL-12 therefore must not
only simultaneously deliver the p35 and p40 subunits, but both genes must be
expressed in a near-equimolar fashion to maximize biologic efficiency and avoid
competitive inhibition of IL-12 activity. Two bicistronic constructs expressing the first
subunit from the retroviral promoter region and the second subunit from an internal
ribosomal entry site (IRES) in alternative order (p40/p35 and p35/p40) were
compared to a single chain IL-12 molecule (scIL-12) in which both subunits were
connected through a (Gly
4
Ser)
3
linker as previously reported (Lieschke 1997). We
hypothesized that the p35/p40 construct would produce the most biological active
form of IL-12 because IRES-dependent second gene expression is usually about 4-
6 fold lower than the first gene (Mizuguchi 2000) and the formation of p40
homodimers should therefore be minimal. However, the order of gene arrangement
did not seem to be important because both bicistronic vectors were equally superior
to the single chain construct, and a highly secretory BM185 clone transduced with
the p35/p40 vector was selected as “IL-12Vax” for subsequent in vivo experiments.

55
3.2 Material and Methods

Mice
Male Balb/c mice, 6-8 weeks old, were purchased from Jackson Laboratories (Bar
Harbor, ME) and kept under standard housing conditions at the Childrens Hospital
Los Angeles (CHLA) Animal Care Facility. All vaccination and leukemia challenge
protocols were reviewed and approved by the CHLA Institutional Animal Care & Use
Committee (IACUC).

Cell lines
The BM185 cell line was previously created by transformation of Balb/c bone
marrow with the human p185 BCR-ABL oncogene (Stripecke 1999) and selection
for a clone with a pre-B cell leukemia phenotype. Wild type BM185 cells have been
further transduced in our laboratory with murine CD80, murine GM-CSF, and murine
CD40L (Gruber 2002) and a clone expressing high levels of all three
immunomodulators termed “Triple Vax” was used for vaccination analysis. All
BM185 cells were cultured in R10M medium (RPMI1640 supplemented with 10%
fetal bovine serum (FBS), 2 mM L-glutamine, 100 U/ml penicillin/streptomycin and
50 μM β-mercaptoethanol) in a 5% CO2 humidified atmosphere. NIH-3T3 fibroblasts
(ATCC, Manassas VA) were cultured in DMEM medium supplemented with 10%
newborn calf serum (NCS; Gibco), 2 mM L-glutamine and 100 U/ml
penicillin/streptomycin, and 293T cells (ATCC) in DMEM medium with 10% FBS, 2
mM L-glutamine and 100 U/ml penicillin/streptomycin.

56
Vector constructs
The sk-mp35 and sk-mp40 plasmids encoding the cDNA for the murine IL-12
subunits were kindly provide by Paul Robbins, University of Pittsburgh, and the
bicistronic vectors MND-p35/p40 and MND-p40/p35 were previously constructed by
the Research Immunology/BMT vector core at Childrens Hospital Los Angles. All
murine IL-12 plasmids utilized the MND retroviral backbone which is based on the
Moloney murine leukemia virus (MoMLV) with modifications in the long terminal
repeat (LTR) that minimize gene silencing and enhance expression in hematopoietic
lineages (Challita 1995). Gene transcription in all constructs was driven from the
internal enhancer/promoter element in the MND 5’LTR region. In the p35/p40
bicistronic vector, the first p35 protein translation was cap-dependent and the
second p40 protein translation was initiated by the encephalomyocarditis virus
(EMCV) internal ribosome entry site (IRES); the same applies for the p40/p35
bicistronic vector albeit in alternative positions. The single chain IL-12 fusion protein
was constructed by 2-round PCR amplification of both subunits with synthetic
primers introducing the polylinker sequence (see Fig 3.0) as previously described
(Jiang 1999). In brief, the MP40-SK plasmid was amplified with the forward primer
F714p40 and the reverse primer R1739Lp40; the MP35-SK plasmid was amplified
with the forward primer F801Lp35 and the reverse primer R1411p35 shown below:

57
F714p40: 5’- TCTAGAGGATCGATCCCCACCAT -3’ (23 bp)
R1739Lp40: 5’- TGGAATGACCCTAGATCCGCCGCCACCCGACCCACCAC
CGCCCGAGCCACCGCCACCGGATCGGACCCT -3’ (69 bp)
F801Lp35: 5’- AGGGTCCGATCCGGTGGCGGTGGCTCGGGCGGTGGTG
GGTCGGGTGGCGGCGGATCTAGGGTCATTCCA-3’ (69 bp)
R1411p35: 5’- CGCTCTAGAACTAGTGGATCAATG-3’ (24bp)
Underlined = (Gly
4
Ser)
3
linker sequence

High-fidelity PCR was performed under the following conditions: 94°C for 5 min;
94°C for 1 min, 55°C for 1.5 min, 68°C for 1.5 min (35 cycles); and 72°C for 7 min.

The single-chain fragment for mIL-12 (scIL-12) was constructed by linkage of the
gel-purified p35 and p40 cDNA fragments in a second round of PCR with the
primers F714p40 and R1411p35 under the following conditions: 94°C for 5 min;
94°C for 1 min, 61°C for 1.5 min, 72°C for 1.75 min (30 cycles); and 72°C for 7 min.
The 1.7 kb scIL-12 PCR product was directly cloned into the pCR2.1-TOPO plasmid
using a TOPO TA cloning kit (Invitrogen, Carlsbad CA), control digested with PstI
and PvuII, sequenced (AnaGen Sequencing, Atlanta GA) and cloned into the MND
vector backbone using the restriction enzymes NcoI and EcoRI.

Vector supernatant production
Retroviral supernatants were prepared as previously described (Soneoka 1995) by
transfecting 293T cells with MND vector plasmid, pHIT60 packaging plasmid and
pHIT123 ecotropic envelope plasmid by calcium phosphate-DNA co-precipitation.
58
p40
L
p35
L
1. PCR round
L
p40 p35
Linker Linker
2. PCR round
PCR products
p40 p35 Linker
Fusion protein
sk-mp40 sk-mp35
Cloning
p40 p35
Linker
pCR2.1-scIL12
Sequencing
Digest
p40 p35
Linker
Nco-I
EcoRI (blunt)
X2 5‘LTR 3‘LTR
Nco-I Xho-I (blunt)
Ligation
3‘LTR
p40 p35 Linker 5‘LTR
MND-scIL12

Fig 3.0: Schematic overview over the MND-scIL12 cloning procedure
PCR fragments of p40 and p35 were generated with primers creating complimentary
overhangs encoding for a (Gly
4
Ser)
3
polylinker. Isolated PCR products were joined in a
second round of PCR. The purified fusion protein was subcloned into the PCR2.1
plasmid, verified by DNA sequencing and cloned into the multicloning site X2 of an MND
backbone plasmid with the restriction enzymes Nco-1 and EcoR1.

59
Cells were induced with 10 mM sodium butyrate 16 hours after transfection.
Retroviral supernatants were collected twice at 24 and 48h after transfection,
concentrated 100-fold by filtration through Centricon columns (Millipore, Billerica
MA) and titered in transduced 3T3 cells by Southern blot hybridization with a
specific probe for p35.

Retroviral transduction and single cell cloning of BM185 cells
BM185 cells were transduced twice with concentrated supernatant at a multiplicity of
infection (MOI) of < 1. Cells were expanded, tested for IL-12 secretion by p40 ELISA
and viably frozen. Transduced pools were thawed and sorted by flow cytometry
using an automated cell deposition unit (ACDU; Beckman-Dickinson) as single cells
into U-bottom 96-well plates containing 150 μl R20 medium (RPMI supplemented
with 20% FBS, L-glutamine, and Pen/Strep) and 2500-5000 irradiated (3000 rads)
wtBM185 feeder cells. Clones were expanded, screened for IL-12 secretion by p40
ELISA and viably frozen.

Southern Blots
Genomic DNA of BM185 clones was obtained by phenol chloroform extraction, and
15 μg of DNA were digested with the single cutter restriction enzyme SphI,
separated on a 0.8% agarose gel and blotted onto a nylon membrane (Micron
Separation Inc, Westborough MA) by 20x SSC wet transfer. The blot was hybridized
with a
32
P-dCTP labeled (Random Primer Labeling Kit, Stratagene) Pst-1 probe (472
bp) specific for the p40 subunit, washed and exposed to Kodak X-Omat film.

60
Enzyme-linked immunosorbent assay (ELISA)
BM185 clones were grown at 1x10
6
cells/ml for 24h and centrifuged; IL-12
containing supernatants were collected. IL-12 concentration was measured with the
IL-12 +P40 or the IL-12 +p70 Immunoassay Kit (BioSource International Inc.,
Camarillo CA) at different dilutions (undiluted, 1:2, 1:10, 1:40) and at least in
duplicate according to the manufacturer’s instructions. All values were calculated
using the standard mean of replicates.

IL-12 Bio-Assay (Splenocyte proliferation assay)
Splenocytes were harvested from naïve Balb/c mice, washed in PBS and treated
with red blood cell lysing buffer (Sigma-Aldrich, St. Louis MO). In a 96-well format,
1x10
4
splenocytes were seeded in 100 μl IMDM medium supplemented with 5%
FBS, L-glutamine and pen/strep and sub-mitogenic amounts of PMA (20 ng/ml) and
stimulated with 100 μl IL-12 supernatant (see ELISA method) or a standard curve
using recombinant murine IL-12 (rmIL-12; R&D Systems, Minneapolis MN) for 4
days in a 5%CO
2

humidified atmosphere at 37°C. Splenocyte proliferation was
measured by [
3
H]-thymidine incorporation after 16 hours of incubation.

Leukemia challenges, vaccinations and IL-12 treatment
For leukemia rejection studies, Balb/c mice were injected intravenously (i.v.; tail vein
injections) with 2000 syngeneic transduced or untransduced (wtBM185) cells and
observed for survival. For prophylactic vaccination studies, mice were injected
subcutaneously (s.c.) with 1x10
6
irradiated (3000 rad?) wtBM185, tripleVax or IL-
12Vax cells on day -14 and day -7 prior to challenge with 5000 live wtBM185 cells.
61
For therapeutic vaccination studies, mice were first challenged i.v. with 1000 live
wtBM185 cells and subsequently vaccinated s.c. with 1x10
6
TripleVax or IL-12Vax
cells on days 1, 5, and 12 after challenge. Recombinant murine IL-12 was given s.c.
were indicated at 2.5 μg/mouse on days 0,1,2,3,4 and 14,15,16, 17, 18 post
challenge as previously described (Gruber 2005). All mice were observed for
survival after leukemia challenge for 60 days and moribund animals were sacrificed
at the onset of disease. Long term survivors (>60 days) were re-challenged with
5000 live wtBM185 cells and observed for survival for an additional 60 days. Spleen
weights of deceased mice were recorded and peripheral blood smears were stained
for leukemic blasts with Wright-Giemsa staining.

Cytotoxic T lymphocyte (CTL) assay
Splenocytes were harvested from vaccinated or control animals and stimulated with
irradiated BM185 cells in the presence of murine recombinant IL-2 (10U/ml; source)
for 6 days at 37°C, under 5% CO
2
and humidified conditions. Splenocyte effector
cells were then co-cultured with
51
Cr-labeled BM185 target cells at various
effector:target (E:T) ratios for 4 h and % specific target cell lysis was measured as
the mean values of released chromium minus spontaneous release devided by total
lysis minus spontaneous release.

62
3.3- Results

Cloning of IL-12 transduced BM185 cells
Parental BM185 cells (wt-BM185) were transduced with retroviral MND vectors
encoding both murine IL-12 subunits as depicted in fig 3.1a: the bicistronic p35/p40
construct was designed to drive p35 translation cap-dependent and p40 translation
from in internal ribosome entry site (IRES); the bicistronic p40/p35 construct drives
p40 translation cap-dependent and p35 translation IRES-dependent; and the
monocistronic single chain (sc) IL-12 construct encodes for a fusion protein in which
the p40 and p35 subunits have been artificially linked by a (Gly
4
Ser)
3
linker
sequence.

Transduced pools were sorted as single cells by ACDU into 96 well plates
containing irradiated wt-BM185 feeder cells. About 20-50 colonies were expanded
for each IL-12 construct and screened for IL-12 secretion by ELISA (data not
shown). The following IL-12 positive clones were continued to be grown in culture:
p35.3, p35.10, p35.21, p35.24 (p35/p40 construct); p40.1, p40.14, p40.15 (p40/p35
construct); and sc.6, sc.15, sc.26, sc.28 (sc construct). The number of vector
integrants per clone was determined by Southern blot analysis using the restrictions
enzyme Sph1 that cuts only once within the proviral sequence and releases
fragments of 2.4 kb or higher depending on the unique integration site (fig3.1A). The
autoradiography in fig3.1b shows that each clone contained a single randomly
integrated proviral copy. A 7kb band visible in all samples including untransduced
63

Fig 3.1: Southern blot of IL-12 clones
Genomic DNA was isolated from untransduced wt-BM185 cells (wt), and BM185 cells
transduced with the 35/40 vector (clones 3,10,21, 24), the 40/35 vector (clones 1.14.15)
or the scIL12 vector (clones 6, 15,26,28). DNA was digested with the single cutter
enzyme SphI, electrophorezed and hybridized with a radioactively labeled probe specific
for the p40 subunit. The 7 kb band represents an endogenous p40 Sph1 fragment.

wt-BM185 control cells represents a band released from the endogenous murine
p40 subunit after genomic DNA digestion.

IL-12 secretion levels and bioactivity
The rationale behind our vector design was to minimize the production of inhibitory
monomeric p40 subunits or p40 homodimers. We hypothesized that cap-dependent

64
A
0
5
10
15
20
25
30
p35 pool
p40 pool
sc pool
p35.3
p35.10
p35.21
p35.24
p40.1
p40.14
p40.15
sc.6
sc.15
sc.26
sc.28
wt
Secreted IL-12 [pg/ml/ 24h/ 10
6
cells]
p40 ELISA
p70 ELISA

B
0
500
1000
1500
2000
2500
3000
3500
p35.3
p35.10
p35.21
p35.24
p40.1
p40.14
p40.15
sc.6
sc.15
sc.26
sc.28
wt
IL-12 Units/ml

Fig 3.2: IL-12 secretion and bioactivity
(A) BM185 pools transduced with IL-12 vectors (p35, p40, sc pool) or BM185 clones
were grown at 10
6
cells/ml for 24 hours; supernatants were collected and IL-12 content
was measured by an ELISA kit with either both capture and enzyme-linked antibodies
specific for p40 (p40 ELISA; light borders) or with antibodies specific for p40 and p35
(p70 ELISA; bold borders). Untransduced wt-BM185 cells (wt) were used as a negative
control. Data shown are summarized from 6 independent experiments (p40 ELISA) or 3
independent experiments (p70 ELISA), respectively. (B) The same supernatants
collected in A) were used to stimulate splenocytes from naïve Balb/c mice in the
presence of submitogenic amounts of PMA (20 ng/ml) for 4 days. Splenocyte
proliferation was measured by [
3
H]-thymidine incorporation and active units were
extrapolated from a standard curve using rmIL-12. Data shown are summarized from 3
independent experiments.

65
translation of the p35 units would increase the relative expression of p35 over IRES-
dependent p40 translation and therefore enhance the formation of p70 heterodimers
which consist of the p35 and p40 subunits linked by a disulfide bond. Moreover, we
wanted to compare the efficiency of both bicistronic murine IL-12 vectors in alternate
order to the single chain IL-12 fusion gene. To test the ratio of p35:p40 expression,
we performed two types of enzyme-linked immunosorbent assays (ELISA). Both
p40 and p70 ELISA kits (BioSource) contained 96 well plates coated with anti-p40
antibodies to capture IL-12 present in supernatants from transduced cells. The
difference between the two assays was that the p40 ELISA kit used a horse radish
peroxidase (HRP) enzyme-linked soluble antibody to a different epitope of p40 and
therefore detected p70 heterodimers, p40 homodimers and p40 monomers,
whereas the p70 ELISA kit utilized an enzyme-linked antibody to the p35 subunit to
only detect the presence of p70 heterodimers after extensive washing and
chromogen substrate incubation.

Transduced BM185 pools and clones were grown at 1x10
6
cells/ml for 24 h, and
supernatants were used for both p40 and p70 ELISA (fig 3.2a). Compared to an
untransduced wt-BM185 control that lacked IL-12 secretion, all three transduced
pools secreted moderate levels of IL-12 as measured by both ELISA kits. A lack of
selectable markers in our IL-12 vector constructs did not allow for pools to be
enriched for transduced cells prior to cloning. Pools therefore represent a
heterogeneous population of transduced and non-transduced cells at an unknown
ratio, and the low IL-12 levels can most likely be attributed to a sub-fraction of cells
secreting IL-12 at a high level. The amount of IL-12 secretion in clonal populations
66
exceeded those of the transduced pools for all three vector constructs. The ratio of
p35: p70 was skewed towards p40 monomer and/or homodimer production in both
bicistronic vectors regardless of the transgene order (Table 3). However, the
p35/p70 ratio was similar in all sc clones and the sc pool which validates the
comparison between both ELISA assays since the sc construct was monocistronic
and subunit detection would be expected to be equivalent. The overall IL-12 p70
production was higher in the sc clones compared to p35 and p40 clones, and overall
p40 production was highest in the p35 clones, slightly lower in the p40 clones and
lowest in the sc clones.

Table 3. Relative p35/p70 ratio in heterogeneously transduced pools and cloned
BM185 cells as determined by p35 and p70 ELISA

Pools p35/p70 Clones p35/p70

p35 pool 2.89 p35.3 3.56
p35.10 2.67
p35.21 3.23
p35.24 2.18

p40 pool 4.39 p40.1 3.06
p40.14 3.57
p40.15 3.63

sc pool 1.55 sc.6 1.87
sc.15 1.41
sc.26 1.02
sc.28 1.73

To determine whether the IL-12 protein production levels correlate to biological
activity, we performed an IL-12 bioassay (fig 3.2b). Splenocytes from naïve Balb/c
67
mice were stimulated with the same supernatants used for the ELISA assays and in
the presence of submitogenic amounts of PMA. Proliferation was measured after
four days of stimulation, and active units were derived by comparison to a standard
curve with recombinant IL-12 after normalization to unstimulated splenocyte
proliferation. IL-12 supernatants from sc clones showed very little biological activity,
almost comparable to supernatants collected from untransduced wt-BM185. In
contrast, all p35 and p40 clones showed activity above background levels.

Based on the above findings, we concluded that the p35/p40 vector was probably
the most effective construct because the biological activity was found to be the
highest and secretion levels were consistently high. For further experiments, we
opted against the use of p35.10, the clone with the highest production and activity
levels and chose p35.3 instead, a clone with average values. The rational behind
this choice was that any result obtained in this study should be theoretically
reproducible with any new transduction and not be solely based on data acquired
with an outlier. All animal work shown below was conducted with live p35.3 cells,
termed IL12-BM185, or with irradiated p35.3 cells for vaccination, termed IL-12Vax.

In vivo effects of IL-12 on the rejection of live leukemia cells
In an initial pilot experiment, immunocompetent Balb/c mice were challenged with
either 2000 live wt-BM185 or IL12-BM185 cells (fig 3.3a). As consistently observed
in our laboratory in the past, injection of wt-BM185 led to rapid leukemia
development within the first three weeks after challenge. The disease was typically
characterized by a ruffled appearance of the fur, hind limb paralysis, splenomegaly
68
and the presence of leukemic blasts in the peripheral blood as determined by
Wright-Giemsa staining of blood smears taken post mortem (data not shown).
Leukemia can usually be detected in the blood, spleen, and bone marrow (Gruber
2005). A challenge with IL12-BM185 cells was generally rejected, resulting in 80%
long-term survival (n=5). The survivors were re-challenged with 5000 live wt-BM185
cells on day 60 and completely rejected this secondary challenge that was
otherwise fatal in a naïve control arm injected at the same time (re-challenge
control). This indicates that animals initially receiving IL12-BM185 cells were not
only able to mount a primary immune response to Ph
+
ALL leukemia, but also
developed memory against a secondary challenge that lacked IL-12 expression.

The experiment was repeated in an attempt to delineate the effects of IL-12
expressed locally by leukemia cells versus IL-12 given systemically as a
recombinant protein (fig 3.3b). Balb/c mice (n=10) were challenged, this time with
5000 live wt-BM185, IL12-BM185, or with wt-BM185 with concomitant systemic
recombinant IL-12 therapy (2.5 μg s.c. on days 0-4 and 14-18 post challenge). An
initial survival rate of 90-100% was achieved after primary challenge if IL-12 was
provided either locally (IL12-BM185) or systemically (wt-BM185 + IL12). After 60
days, all survivors were re-challenged with 5000 wt-BM185 leukemia cells, but this
time almost no rejection of the secondary was observed except in 2 out of 10 mice
initially challenged with IL12-BM185. Thus, the local production of IL-12 by the
tumor cells seemed to have some advantage over systemic IL-12 treatment
regarding the formation of immunologic memory, but the 100% survival rate of re-
challenged animals achieved in the initial experiment (fig 3.3a) could not be
69
reproduced. The difference between both leukemia challenge studies was the initial
dose of leukemia cells, 2000 and 5000, respectively, and it can be speculated that
the blood concentration of IL-12 cytokine is crucial for the delicately balanced
processes involved in immunological memory formation.

Fig 3.3: Rejection of IL-12 expressing live leukemia
A) Balb/c mice were challenged i.v. with 2000 live wt-BM185 cells (dotted line) or 2000 IL12-
BM185 cells (black triangles). Long term survivors and naïve control mice (re-challenge
control; dotted line) were re-challenge with 5000 live wt-BM185 cells given i.v. on day 60
after initial challenge and observed for survival for an additional 60 days.
B) Balb/c mice were challenged i.v. with 5000 live wt-BM185 cells and left untreated (dotted
line) or treated with 2.5 μg murine recombinant IL-12 on days 0-4 and 14-18 post challenge
(open circles); an additional cohort was challenged with 5000 IL12-BM185 cells (black
triangles). Long term survivors were re-challenged with 5000 live wt-BM185 cells on day 60
after initial challenge including a naïve control cohort (re-challenge; dotted line). Data shown
are from two independent experiments (n=15).
C) Peripheral blood was drawn periodically from the tail veins of 2 mice from each cohort
represented in A (wtBM185, IL12BM185 or wtBM185+ s.c. IL-12 treatment on the days
indicated) or untreated naïve control mice. Serum levels of IL-12 were measured by IL12
+p40 ELISA.
70
A
0
20
40
60
80
100
0 20 40 60 80 100 120
Days after Challenge
Survival [%]
wt-BM185
IL12-BM185
Re-challenge control
↑
Re-challenge
wt-BM185

B
0
20
40
60
80
100
0 20 40 60 80 100 120
Days after Challenge
Survival [%]
wt-BM185
wt-BM185 + IL12
IL12-BM185
re-challenge control
↑
Re-Challenge
wt-BM185

C
0
100
200
300
400
500
600
0 4 7 11141821 25 28 32
Days after Challenge
Serum IL-12 [pg/ml]
Naïve
wt-BM185
IL12-BM185
wt-BM185 + IL12

71
For 35 days after the initial challenge performed in fig3.3b, peripheral blood was
drawn from two mice of each cohort every 3-4 days (fig 3.3c). Mice challenged with
wt-BM185 showed baseline levels of IL-12 (approximately 80 pg/ml) comparable to
those of naïve mice for the first 14 days after challenge. Levels increased drastically
to over 500 pg/ml only at the onset of leukemia, presumably as a result of
inflammatory responses. In mice treated with systemic IL-12 administration, serum
levels were high immediately after the time point of injection (days 1-4 and 14-18, up
to 250-400 pg/ml) and then sank back to normal values within two weeks thereafter.
Mice challenged with IL12-BM185 showed a modest increase of IL-12 levels to 120
pg/ml within the first two weeks after challenge. These results indicate that systemic
IL12 administration drastically increases circulating IL-12 levels in the peripheral
blood up to 5 fold over the baseline levels, whereas IL-12 production from IL12-
BM185 leukemia cells is more locally confined and raises blood levels of IL-12 only
modestly and temporarily.

Prophylactic vaccination before the establishment of leukemia
After observing a lack of immunological memory in response to a live challenge with
IL-12 expressing leukemia cells, we wanted to determine whether IL-12 secreting
irradiated BM185 cells are effective as a cancer vaccine prior to a subsequent
challenge with wild type leukemia cells. Our laboratory conducted extensive studies
on the efficiency of genetically modified BM185 cells engineered to express
immunomodulatory molecules as prophylactic cancer vaccines. The vaccine
established to induce rejection of subsequent leukemia challenges most efficiently
consisted of BM185 cells transduced to express CD80, GM-CSF and CD40L
72
simultaneously (Gruber 2005), hence termed “TripleVax”. Our goal was to compare
TripleVax to IL12Vax given either alone or in combination and determine the effects
on primary leukemia challenge rejection and on immunologic memory formation.
Balb/c mice (Fig 3.4a) were vaccinated with 2x10
6
irradiated cells total according to
the following scheme:

Table 4. Number of irradiated cells administered combined as a prophylactic vaccine
(Fig.3.4a)

Treatment Group wt-BM185 TripleVax IL12Vax

Triple/wt

1 x 10
6

1 x 10
6

-
Triple/IL12 - 1 x 10
6
1 x 10
6

IL12/wt 1 x 10
6
- 1 x 10
6

TripleVax - 2 x 10
6
-
Challenge Control - - -
Re-Challenge
Control

- - -

Mice vaccinated with IL12Vax or the Triple/IL12Vax combination were less
protected from subsequent leukemia challenges (10% survival) then mice that
received TripleVax in the absence of IL-12 expressing cells (30-80% survival). None
of the IL12Vax immunizations protected mice from secondary challenges, whereas
a memory response resulted in survival after the secondary challenge in TripleVax-
treated mice. Interestingly, a dose-dependent effect was observed among TripleVax
recipients: 80% of mice injected with a double dose of TripleVax (2 x10
6
, TripleVax
group) mounted a memory response, whereas only 20% survived up to day 120
after the initial challenge after receiving 1x10
6
TripleVax cells mixed with 1 x10
6
inert
wtBM185 cells (Triple/wt group). The observed dosage effect stresses the

73
Fig 3.4: Leukemia rejection after prophylactic vaccination
A) Balb/c mice were vaccinated s.c. with 1x10
6
TripleVax + 1x10
6
wtBM185 (Triple/wt; black
triangles), 1x10
6
TripleVax + 1x10
6
IL-12Vax (Triple/IL-12; white triangles); 1x10
6
IL-12Vax +
1x10
6
wtBM185 (IL12/wt; grey triangles) or 2x10
6
TripleVax (TripleVax; black squares) on
days -14 and -7 prior to i.v. challenge with 5x10
3
live wtBM185 cells on day 0 (n=10). Long
term survivors (>60 days) were re-challenged with 5x10
3
live wtBM185 cells at day 60 after
initial challenge. Control mice received no vaccination and either challenge only or re-
challenge only (dotted lines)
B) Balb/c mice were vaccinated s.c. with 1x10
6
TripleVax + 1x10
6
wtBM185 filler cells
(TripleVax; black squares), 1x10
6
TripleVax + 1x10
5
IL-12Vax + 9x10
5
wtBM185 (Triple/IL-
12 10:1; black triangles); 1x10
6
TripleVax + 3x10
5
IL-12Vax + 7x10
7
wtBM185 (Triple/IL12
3:1; grey triangles) or 1x10
6
TripleVax + 1x10
6
IL12Vax(Triple/IL-12 1:1; white triangles) on
days -14 and -7 prior to challenge. Peripheral blood was drawn on days -11 and -4 (3 days
post vaccination) to determine serum IL-12 levels. Mice were challenged i.v. with 5x10
3
live
wtBM185 cells on day 0 (n=10). Long term survivors (>60 days) were re-challenged with
5x10
3
live wtBM185 cells at day 60 after initial challenge. Control mice received no
vaccination and either challenge only or re-challenge only (dotted lines).
C) Peripheral blood was drawn from the tail veins of 3 mice from each vaccination cohort (wt
Vacc, TripleVax, Triple/IL12 10:1, 3:1 and 1:1; see Fig A) 3 days post vaccination. Serum
levels of IL-12 were measured by IL12 +p40 ELISA.
D) Long term survivors were sacrificed on day 120 after initial challenge (60 days after re-
challenge), splenocytes were stimulated with irradiated wtBM185 cells for 6 days and a
chromium release assay with
51
Cr-labeled live wtBM185 cells was performed at various
effector:target cell (E:T) ratios. Splenocytes from a naïve mouse were used as a negative
control.

74
A
0
20
40
60
80
100
0 204060 80 100 1
Days after Challenge
Survival [%]
20
Triple/wt
Triple/IL12
IL12/wt
TripleVax
challenge control
rechallenge control
↑
Re-challenge
wt-BM185

B
0
20
40
60
80
100
0 2040 6080 100 1
Days after challenge
Survival [%]
20
TripleVax
Triple/IL12 10:1
Triple/IL12 3:1
Triple/IL12 1:1
wt Vacc
challenge control
rechallenge control
Re-challenge
wt-BM185
↑

C
0
20
40
60
80
100
120
140
160
wtBM185 TripleVax Triple/IL12
10:1
Triple/IL12
3:1
Triple/IL12
1:1
Serum IL-12 [pg/ml]
D
0
10
20
30
40
50
100 30 10 3 1 0.3
E:T Ratio
Specific Target Cell Lysis [%]
Naïve
TripleVax
Triple/IL12 3:1
Triple/IL12 1:1

75
complexity of immune responses to multiple antigens in combination with multiple
immune modulatory molecules that can often explain inter- and intra-experimental
variations.

In the previous live leukemia challenge experiments (fig 3.3) we noted that the local
concentration of IL-12 might be crucial for the establishment of a memory response.
In our next prophylactic vaccination experiment we varied the doses of IL-12
secretion by varying the ratios of TripleVax/IL12Vax. Balb/c mice (Fig 3.4b) were
vaccinated twice with at total of 2x10
6
irradiated cells according to the following
scheme:

Table 5. Number of irradiated cells administered combined as a prophylactic vaccine
(Fig.3.4b)

Treatment Group wt-BM185 TripleVax IL12Vax
TripleVax 1 x 10
6
1 x 10
6
-
Triple/IL12 10:1 0.9x 10
6
1 x 10
6
0.1x 10
6

Triple/IL12 3:1 0.7x 10
6
1 x 10
6
0.3x 10
6

Triple/IL12 1:1 - 1 x 10
6
1 x 10
6

Wt Vacc 2 x 10
6
- -
Challenge Control - - -
Re-Challenge
Control
- - -

In this experiment, 20-40% survival of the primary challenge of 5000 live wt-BM185
cells was observed for vaccinated mice but not for unvaccinated or mice vaccinated
with wt-BM185 cells only. Survivors were re-challenged at day 60 with 5000 wt-
76
BM185, and one animal each of the TripleVax, Triple/IL12 3:1 and Triple/IL12 1:1
cohorts survived; in the cohort receiving the lowest dose of IL-12 (Triple/IL-12 10:1)
no survival after re-challenge was achieved.

To analyze whether the titration of IL12Vax combined with TripleVax treatment
resulted in a noticeable difference in serum IL-12 levels, peripheral blood was
collected from 3 mice of each cohort presented in fig 3.4b three days after each
vaccination and tested for IL-12 concentrations by IL12 +p40 ELISA (fig 3.4c). The
baseline levels of IL-12 were found to be around 70 pg/ml in mice vaccinated with
non-secretory wt-BM185 cells alone. Similar serum levels were found for the
TripleVax, Triple/IL12 10:1 and 3:1 vaccinated animals. Only in mice receiving the
highest dose of IL-12 secreting cells, Triple/IL12 1:1, showed an increased level of
serum IL-12, about 110 ng/ml. This shows that the irradiated IL12Vax cells were
able to secrete the cytokine in a vaccination setting, although IL-12 production in
mice receiving lower doses of IL12Vax cells was probably confined locally to the
vicinity of the secreting cells at levels too low to be detected in the blood stream by
ELISA.

Survivors of the secondary challenge were routinely sacrificed after a total
observation period of 120 days. To address whether immunologic memory
correlated with the presence of BM185-specific cytotoxic T lymphocytes (CTL), we
performed a CTL assay on lymphocytes harvested from the spleens of long-term
survivors and a naïve control. Splenocytes were stimulated in vitro with irradiated
wt-BM185 cells in the presence of IL-2 for 6 days and then co-cultured with
77
radiolabeled wt-BM185 target cells for 4h at the effector:target cells (E:T) ratios
indicated (fig3.4d). No lysis of BM-185 cells above background was observed for
lymphocytes from a naïve control mouse. Survivors initially vaccinated with
TripleVax or Triple/IL12 1:1 on days -14 and -7 prior to challenge showed moderate
capability to lyse BM185 target cells, about 5-15% at the highest E:T ratio. In
contrast, CTLs derived from the long-term survivor that received the Triple/IL12 3:1
treatment specifically lysed 40% of BM185 target cells. It is possible that moderate
doses of local IL-12 production are most efficient for the generation of long-term
memory, although more research needs to be done to support this hypothesis.

Therapeutic vaccination after the establishment of leukemia
The most rigid test of the efficacy of a whole cell cancer vaccine is to show that it
can eradicate pre-established tumors. Balb/c mice were challenged with 1000 live
wt-BM185 cells on day 0 and subsequently vaccinated with 1x10
6
irradiated cells on
days 1, 5 and 12 in the presence or absence of recombinant IL-12 administration on
days 0-4 and 14-18 (fig 3.5). All vaccinated mice rejected the pre-established
leukemia challenge, even when recombinant IL-12 was given with irradiated
wtBM185 cells, and only non-vaccinated control mice died of the challenge within 21
days. Mice were re-challenged on day 60 with 5000 live wt-BM185 cells to
determine which vaccination strategy resulted in immunological memory. The re-
challenge dose was fatal in a naïve control group receiving this dose as their
primary challenge (re-challenge control). Mice that previously received IL12Vax
alone were not able to maintain event-free survival after re-challenge, and
recombinant IL-12 treatment alone established low levels of survival (20%). A
78
number of combination treatments, however, resulted in 40% of long-term survival,
including TripleVax, TripleVax or IL12Vax in combination with systemic IL12
treatment and a combination of TripleVax and IL12Vax. The implementations of
these results are not completely clear, especially because we were never able to
show memory development after therapeutic vaccination with TripleVax before (not
shown). However, the data suggest that combinations of TripleVax plus systemic or
local IL12 delivery can enhance rejection of pre-established Ph
+
ALL and mount a
memory response capable of rejecting relapsing leukemia.

0
20
40
60
80
100
0 20406080 100 12
Days after Challenge
Survival [%]
0
challenge control
wt-BM185 + IL12
TripleVax + IL12
IL-12Vax + IL12
TripleVax
IL-12Vax
TripleVax + IL12Vax
Re-challenge control
↑
Re-Challenge
wt-BM185

Fig 3.5: Leukemia rejection after therapeutic vaccination
Mice were challenged i.v. with 1000 live wt-BM185 cells on day 0 and subsequently
vaccinated with 1x10
6
irradiated cells (wt-BM185, TripleVax or IL-12Vax) on days 1, 5
and 12 after challenge. Mice received additional s.c. rmIL-12 treatment (2.5 μg) on days
0-4 and 14-18 where indicated (wt-BM185 + IL12, TripleVax + IL12, IL-12-Vax + IL12).
Long term survivors and naïve control mice (re-challenge control) were injected with
5000 live wt-BM185 cells on day 60 after the initial challenge and observed for survival
for an additional 60 days.

79
3.4- Discussion

Ph
+
ALL continues to be associated with poor prognosis and new treatment
modalities are urgently needed. Current standard care is a combination drug
induction followed by blocks of chemotherapy intensification adding the Bcr-Abl
kinase inhibitor Imatinib to the regimen of chemotherapeutic agents (review: Jones
2005). Although the initial response rate is quite successful, over 25-30% of patients
develop Imatinib resistance and undergo allogeneic stem cell transplantation (allo-
SCT) to sustain remission. If all treatment options fail and relapse occurs, one could
take advantage of the unique immunogenicity of the BCR-ABL tumor antigen and
consider immunotherapy as an alternative treatment. Clinical trials have shown
moderate prognostic benefits of cell-based vaccines for a variety of human
malignancies, including cancers of the colon (Uyl-de-Groot 2005), prostate (Simons
2006), brain (Sloan 2000) and skin (Parmiani 2003; Dranoff 2003). While the anti-
tumor effects of IL-12 as a single agent for systemic cancer therapy have been
disappointing, IL-12 administered as an adjuvant for cancer vaccination has been
reported to reverse immune tolerance to tumor antigens (Portielje 2003). The
problem with IL-12 therapy is that systemic administration can have severe side
effects in humans and lead in rare cases to fatalities (Portielje 2003). Local delivery
of IL-12 by genetically modified cancer cells may reduce toxicity risks and has been
shown to be well-tolerated in patients. IL-12 transduced cancer cells given as tumor
vaccines have been investigated in a clinical trial for melanoma patients (Kang
2001) and in a large number of animal models for lung (Myers 1998, Popovic 1998),
breast (Carr-Brendel 1999), colon cancer (Lechanteur 2000) and acute myeloid
80
leukemia (Dunussi-Joannopoulos 1999). Our goal was to investigate whether IL-12
transduced Ph
+
ALL cells were efficient to induce rejection of this highly aggressive
form of leukemia and establish long-lasting immunological memory in mice.

The murine model of Ph
+
ALL established in our laboratory utilizes the BCR-ABL
transformed pre-B cell line BM185 that expands rapidly in vitro and in vivo, with
1000 cells given intravenously resulting in 100% mortality 3-4 weeks after challenge
of syngeneic mice. Although findings in animals do not necessarily translate into the
success of human studies, using this highly aggressive murine leukemia provides us
with a very stringent model to test the efficacy of various immunotherapies. Previous
work demonstrated that the modification of BM185 cells with immunomodulatory
genes could result in tumor rejection and prophylactic protection, and systemic
administration of IL-12 as an adjuvant increased the effectiveness of therapeutic
vaccines, although a lack of memory formation was noted. To extend our studies
and improve ALL cell-based vaccination with IL-12 adjuvant administration, we
engineered BM185 cells to express and secrete IL-12. BM185 cells were
successfully transduced with either one of our three retroviral constructs, p35/p40,
p40/p35 and the single chain scIL12 vector. The p35/p40 construct was more
effective than the other two vectors regarding secretion levels and biological activity.
Surprisingly, both bicistronic transgenes secreted more p40 subunits than p70
heterodimers regardless of which subunit was translated in an IRES-dependent
manner, but we did not address the intracellular expression levels or mRNA and
protein half lives that might play a role in the stochastic balance of both subunit
levels. Secreted monomeric p40 molecules and p40 homodimers have been
81
predicted to be antagonistic to IL-12 activity (Gillessen 1995), but biological activity
of dimeric p70 produced from both bicistronic vectors was superior to that of the
single chain IL-12 fusion protein even though p40 overexpression was prevented by
monocistronic translation. We concluded that bicistronic vector design in particular
in the order p35-IRES-p40 was the most useful of all three possible vector
configurations for the genetic modification of leukemia cells. It has to be noted that
the in vitro effects of the IL-12 p40 subunit may differ largely from its in vivo effects
as reported by Sun et al. (2004). The study shows that p40-transduced dendritic
cells activate NK cells possibly through enhanced production of IL-23, a cytokine
redundant to IL-12, rather than inhibiting immune responses. The usage of a
bicistronic IL-12 vector producing p40 subunits may therefore be even beneficial for
cell-based vaccinations, although the single chain vector has been successfully
used in several gene therapy models (Lode 1998; Jiang 1999).

IL-12 expressing live leukemia cells were rejected in mice at a cell dosage that was
100% lethal for untransduced BM185 cells. Prophylactic vaccines with IL12Vax
reduced the rate of tumor incidence and resulted in increased survival after primary
challenge, and therapeutic vaccines with IL12Vax protected mice from pre-
established leukemia. These findings are all in accordance with our previous
observations that systemic IL-12 alone exhibited similar anti-leukemic effects in this
mouse model. The mechanism of protection by systemic IL-12 administration was
found to be mediated by a combination of CD4
+
T helper cells, CD8
+
effector T cells
and natural killer (NK cells), and the lack of one cellular subset was compensated
for by the presence of the others (Gruber 2005). IL-12 is a pleiotropic cytokine that
82
directly promotes the proliferation and effector function of CTLs, activates NK cell
activity, induces dendritic cell maturation and indirectly stimulates CD4 helper cells
by skewing the cytokine milieu towards a T
H
1 response. Its ability to rapidly expand
cells of both the innate and adaptive immune system explains why IL-12 becomes
highly effective even against tumors with fast growth kinetics like ALL.

IL-12 seems to play a dual role in tumor immunology: it enhances tumor rejection in
primary challenge settings and promotes immunologic memory to secondary
challenges. Although the role of IL-12 in memory formation is well-established (Li
2006), the exact requirements are not quite understood. When mice were
challenged with 2000 IL12-BM185 cells, all survivors of the primary challenge were
protected from a secondary challenge, but the majority of mice surviving a primary
challenge of 5000 IL12-BM185 cells lacked immunological memory and succumbed
to the re-challenge. It is known that both the IL-12 concentration present at the time
of first antigen encounter and the duration of the stimulus are crucial for the
generation of a memory response. Naïve CD8
+
T cells require prolonged exposure
to antigen, costimulatory molecules and IL-12 for at least 36h in vitro to develop
their full effector function (Mescher 2006), and failure to provide those signals can
induce non-responsiveness. IL-12 abundance during chronic inflammation inhibits
clonal expansion of CD8
+
T cells and memory generation (Badovinac 2004). In our
study, serum levels of IL-12 were high after systemic administration but rose
moderately and temporarily over background with IL12Vax, and vaccinations with
IL12Vax led to slightly higher incidences of memory formation and survival after
secondary challenge compared to systemic IL-12 treatment. IL12Vax at the dosage
83
given (1x10
6
cells) produced a maximum amount of 15 ng IL-12 over 24 hours, and
the initial dosage of recombinant IL-12 was set to 2.5 μg per injection, more than
100 fold higher. In a murine sarcoma model, low dosage of systemic IL-12 (1
ng/day) given as an adjuvant with a p53 peptide vaccine resulted in CTL activity and
tumor rejection, whereas increasing the IL-12 dose to >10ng/ day failed to induce a
response (Noguchi 1995). Taken together, it appears that using IL-12 as a memory-
inducing adjuvant in ALL whole cell vaccinations requires a careful assessment of
dosage effects, and more work has to be done to determine the optimal dosage.
However, it appears that lower dosages are favorable and more likely to be
produced by local IL-12 secretion of genetically modified tumor cells (our data and
Colombo 1996).

One important component of proper CTL activation and memory induction is the co-
stimulation of naïve CD8
+
T cells through B7/CD28 interactions. Live or irradiated
BM185 cells release tumor-specific antigens when they are phagocytosed by
dendritic cells or macrophages, and those antigens are cross-presented by MHC
molecules on the surface of the APC to naïve CD4
+
and CD8
+
T cells. DCs express
IL-12 receptors and can be stimulated to upregulate surface MHC molecules, co-
stimulatory molecules (B7.1 and B7.2) and mature in the presence of IL-12. BM185
cells lacking co-stimulatory molecule expression were therefore capable of inducing
an immune response in conjunction with IL-12. As previous work has revealed, wt-
BM185 were not potent enough to induce immune responses but could be made
highly immunogenic when they were transduced to express the immunomodulatory
molecules CD80 (B7.1), GM-CSF and CD40L (CD154). Here we wanted to
84
investigate whether a combination of TripleVax and IL-12 would be beneficial to
enhance memory formation. We observed high cytolytic activity of BM185 specifc,
long-term surviving CD8
+
T cells if mice were exposed to low IL-12 levels provided
by IL12Vax and co-stimulation in the form of TripleVax after prophylactic vaccination.
In the therapeutic vaccine setting, adding either recombinant IL-12 or IL12Vax to the
TripleVax vaccine led not only to the rejection of pre-established leukemia but also
resulted in increased survival of a secondary challenge (40%) compared to IL12Vax
alone (0%) or wtVax with systemic IL-12 (20%). A combination of TripleVax along
with IL-12 is therefore the most promising treatment for immunotherapy of Ph
+
ALL;
IL-12 enhances early stimulation of innate and adaptive immune cells to rapidly
reject existing leukemia cells and, in conjunction with immunostimulatory molecules
present at the time of antigen encounter, aids in the generation of memory CD8
+
T
cells that clear relapsing leukemia or secondary challenges.

Like many biological processes, formation of immunological memory is very
complex and benefits from synergistic actions of many combined signals. To
translate therapeutic vaccination for Ph
+
ALL into the clinic, it can be hypothesized
that including a variety of immunomodulators will diversify the immunologic
response and improve clinical outcome. It has been shown that leukemia vaccine
manufacture is a feasible procedure because leukemia cells can be collected at the
time point of diagnosis or relapse and cryopreserved until further use (Haining 2005).
Transducing the cells with more than one transgene simultaneously, however, might
prove extremely difficult, especially if the secretion levels need to be tightly
controlled as found to be the case for IL-12. An alternative approach is to provide
85
GM-CSF and IL-12 in the form of an allogeneic, secretory “bystander” cell line that
can be screened and characterized in advance and used for multiple patients, so
that autologous patient cells would only have to be transduced with a single
bicistronic vector expressing CD80 and CD40L. A “bystander” platform for GM-CSF
secretion has already been tested in a clinical trial for non-small-cell lung cancer
(Nemunaitis 2006), albeit with little success, and more research needs to be
conducted to evaluate the feasibility of this approach. Another potential drawback of
therapeutic vaccines is that patients are usually relatively immunodeficient after
chemotherapy or allo-SCT, and attempts to define a window of immunologic
recovery in ALL patients in remission have failed so far (Haining 2006). However,
given the aggressive nature of Ph
+
ALL and the poor survival outcome, research
needs to be continued to define alternative therapies including immunotherapy that
can complement current treatment strategies and improve patients’ lives.

Acknowledgements:
This study was a collaborative effort and significant contributions have been made
by Dianne Skelton, Tanja Gruber, Duri Yun, Beth Pallian, and Daniel Cotoi.
86
Chapter 4- Leukemia Cell Vaccination after Bone
Marrow Transplantation

4.1- Introduction

Intensive chemotherapy and total body irradiation (TBI) followed by bone marrow
transplantation (BMT) is currently the treatment of choice for hematologic
malignancies and can offer prolonged disease-free survival. Patients with high-risk
leukemia, however, often relapse after transplantation and achieve 5-year event-
free survival in less than 30% of all cases, and alternative or synergistic therapies in
addition to current intensification regimens are highly warranted.

Immunotherapy is a promising adjunct approach to dose-intensive chemoradiation,
especially since immune-mediated tumor cell killing has can be achieved for chemo-
resistant tumor cell lines (Fuchs 1995; Shtil 1999). The efficacy of whole cell tumor
vaccinations is likely to be directly related to tumor burden and may therefore be
most efficient at a time period of minimal residual disease which is observed after
BMT. However, the functional recovery of the immune system after transplantation
can take up to 1-2 years in patients and antitumor effects of a cancer vaccine may
be compromised by this immunodeficiency. Defining the right time point for
vaccination in the post-transplantation period characterized by partial or full
restoration of the immune response but lack of tumor progression is therefore the
most challenging task. Yet the fact that immunizations against infectious diseases
like mumps, measles, tetanus (Henning 1997) or poliovirus (Parkkali 1997) can be
87
effective within the first 1-2 years after BMT makes the development of
immunotherapy for cancer in a post-BMT setting a feasible goal.

Leukemia patients undergoing allogeneic bone marrow transplantation (allo-BMT)
from a matched sibling donor (MSD), partially matched related donor (PMRD) or
matched unrelated donor (MUD) have a reduced relapse rate compared to
autologous BMT. This has been attributed to the Graft versus Leukemia (GVL)
effect conferred by donor T cells against tumor-specific antigens or minor
histocompatibility antigens (mHAgs) expressed on leukemia cells. Unfortunately, the
same mHAgs are widely expressed on normal tissues, and Graft versus Host
Disease (GVHD) is a commonly observed complication leading to a substantial
increase in morbidity and mortality. Strategies to minimize GVHD are to give
immunosuppressive drugs for a prolonged period after BMT (6 months or more) or
to deplete the graft of reactive T cells, measures that both minimize the beneficial
GVL effect.

The lack of a suitable donor, advanced age and other considerations may preclude
a patient from allogeneic transplantation. About 9,700 autologous BM
transplantations are performed annually in the U.S. compared to 7,500 allo-BMTs
(source: 2006 US Organ and Tissue Transplant Cost Estimates, www.milliman.com).
Autologous transplantations using the patient’s own bone marrow purged of
leukemia cells has a more favorable safety profile because graft-versus host
disease is not initiated. Reconstitution kinetics are generally similar between
autologous BMT and allo-BMT, although immune reconstitution can be significantly
88
delayed by the need to T-cell deplete allo-grafts, immunosuppressive therapy for the
control of GVHD and chronic GVHD itself, whereas immunosuppression is only brief
in the autologous BMT setting. Autologous antitumor effects are thought to be
mostly achieved by the intensive chemo and radiation therapy beyond doses
permissible without stem cell support. However, several animal studies support the
hypothesis that a tumor-specific response can be induced after autologous
transplants, proving that the mechanism of GVL effect observed after allo-BMT is
mechanistically separable from GVHD effects (Borello 2000; Mundhada 2005).

A successful whole cell leukemia vaccination approach for Philadelphia
chromosome-positive acute lymphoblastic leukemia (Ph
+
ALL) in a Balb/c model
was previously established in our laboratory using the syngeneic pre B leukemia cell
line BM185 modified to express the three immunomodulatory molecules GM-CSF,
CD80 and CD40L (Gruber 2002). This irradiated “TripleVax” vaccine rejected a
subsequent leukemia challenge with wild type BM185 cells in a majority of mice, but
failed to protect from pre-established leukemia progression when given as a
prophylactic vaccine. In this case, the test animals were overwhelmed by the highly
aggressive leukemia burden before the vaccine could show its effects, and we
hypothesize that vaccination may be more efficient at a time point of minimal
residual disease after bone marrow transplantation, presuming immune functions
are partially or fully restored. We implied TripleVax vaccinations either one or three
months after syngeneic BMT and observed successful induction of anti-leukemic
responses. Three months after transplantation, immune functions were expected to
be completely restored in the recipients, whereas one month post-BMT has been
89
shown to be a time point of impaired immune function in some animal models
(Bolotin 1996; Teshima 2001). To increase the stringency of our model, we
compared whole bone marrow transplants to lineage-depleted bone marrow which
is deficient of mature lymphoid cells including effector T cells. We report that a
syngeneic graft-versus-leukemia effect was achieved even under incomplete
reconstitution conditions.
90
4.2- Material and Methods

Mice
Male 6 to 8-week old Balb/c mice were purchased from Jackson Laboratories (Bar
Harbor, ME) and und housed at the Childrens Hospital Los Angeles (CHLA) Animal
Care Facility under standard conditions. All animal procedures were reviewed and
approved by the CHLA Institutional Animal Care and Use Committee (IACUC).

Cell lines
The murine pre-B cell leukemia cell line BM185 and modified BM185 cells
transduced with murine CD80, GM-CSF and CD40L (“TripleBM185”) have been
described elsewhere (Stripecke 1998; Gruber 2001). Cells were cultured in
RPMI1640 supplemented with 10% fetal bovine serum (FBS), 2 mM L-glutamine,
100 U/ml penicillin/streptomycin and 50 μM β-mercaptoethanol in a 5% CO
2

humidified atmosphere.
Yac-1 cells were obtained from the American Type Culture Collection (ATCC,
Manassas VA) and cultures in RPMI1640 supplemented with 10% FBS, L-glutamine
and pen/strep.

Bone marrow harvest , lineage depletion and transplant
Bone marrow was isolated from the tibiae and femurs of Balb/c donor mice, washed
with IMDM (Iscove’s modified Dulbecco’s medium; Cambrex, Walkersville MD) and
kept on ice until transplant or depleted of differentiated lineage cells by negative
selection using the StemSep
TM
Mouse Progenitor Enrichment Cocktail (StemCell
91
Technologies). In brief, mature T lymphocytes ( α-CD5), erythroid cells ( α-TER119),
B lymphocytes ( α-B220), granulocytes ( α-Gr-1), macrophages ( α-Mac-1) and
neutrophils ( α-7-4) were captured by biotinylated monoclonal antibodies linked to a
magnetic colloid, applied to a magnetic surface by peristaltic flow and retained while
unlabeled cells were recovered. Enrichment of hematopoietic progenitors in the
flow-through was about 100-fold according to the manufacturer, and the number of
recovered cells was typically 1% of the starting material. Whole bone marrow cells
and progenitor cells were washed and resuspended in Hank’s balanced salt solution
containing 50 U/ml Heparin. Mice were irradiated with 900 rads unless stated
otherwise and injected intravenously with 1x10
6
whole bone marrow cells (bone
marrow transplant, BMT) or 1x10
4
enriched stem and progenitor cells (stem cell
transplant, SCT) in 100 μl volume. Mice were treated with 0.5mg/ml Maxim-200
(Phoenix Pharmaceuticals, Burlingame CA) for 14 days after transplant to prevent
infections.

Immunophenotyping by FACS
Mice were sacrificed at the time points indicated (1 or 3 months after transplant)
either before or after receiving vaccination. Peripheral blood was collected by
cardiac puncture; organs were perfused with 10-20 ml PBS, spleen and thymus
were removed and single cell suspension were prepared by passing the tissue
through a 0.7 μm cell strainer screen. Red blood cells were lysed with Red Blood
Cell Lysing Buffer (Sigma-Aldrich, St. Louis MO) in peripheral blood and splenocyte
samples, and all cells were washed with PBS. Cells were stained with APC-labeled
antibodies for CD45 (lymphoid cells) CD8, CD3 (T cells), and PE-labeled antibodies
92
for CD4 (T cells), CD19 (B cells), M/G (macrophage granulocytes) and DX5 (NK
cells) and analyzed on a FACScan flow cytometry apparatus (Becton-Dickinson, ).
Unprocessed peripheral blood samples were also analyzed on a Hemavet System
(Drew Scientific, Oxford CT) for white blood cell (WBC) counts.

ConA proliferation assay
Splenocytes were seeded at 0.5, 1, 1.5 or 2.5x10
6
cells/ml in 200 μl R10 medium
(RPMI with 10% FBS, Pen/Strep and L-Glutamine) in 96 well plates in triplicates and
stimulated with 12 μg/ml Conconavalin A (ConA; Sigma-Aldrich, St. Louis MO) or
mock-stimulated ( medium only) for 72 hours. Cells were pulsed with 1 μCi [
3
H]-
Thymidine (50 μl) and harvested 6-8 hours after pulsing with a vacuum Cell
Harvester. Thymidine incorporation was measured on a Beta Counter, and the
stimulation index was calculated as the mean of experimental counts (+ConA) over
the mean of control counts (-ConA).

NK assay
Splenocytes were seeded in 96 well plates at 2.5 x 10
6
cells/ml in 100 μl R10
medium in triplicates. Yac-1 target cells were labeled with 20 μCi
51
Cr /10
6
cells
(Sodium chromate; Perkin-Elmer, Billerica MA) for 1 h at 37°C, washed and added
to splenocyte wells in 100 μl R10 at an effector: target ratio of 50:1 or to control
wells containing medium only (spontaneous lysis) or 0.1% Triton-X (total lyis). After
16 hours of incubation, supernatants were harvested and counted on a Gamma
Counter. Percent target cell lysis was calculated as the mean of experimental
93
counts minus the mean of background counts (spontaneous lysis) divided by the
mean of total lysis.

Leukemia vaccination and challenge
Triple-BM185 cells were irradiated with 3000 rad in a
137
Cs gamma irradiator
(Gammacell 1000; Norton International, Kanata, Ontorio, Canada), washed and
resuspended in sterile PBS with 50 U/ml Heparin. TripleVax vaccinations were
injected subcutaneously as 1x10
6
irradiated cells in 100 μl volume into the inguinal
region of BMT and SCT recipients 1 or 3 months after transplant or naïve control
mice. Vaccinations were given twice one week apart on days -14 and -7 before
leukemia challenge. On day 0, vaccinated mice and non-vaccinated control mice
were challenged with 5000 live wt-BM185 cells and observed for event-free survival
for up to 60 days after challenge. Moribund animals were euthanized by CO
2
narcosis, spleens were weighted and peripheral blood smears were stained for the
presence of leukemic blasts with Wright-Giemsa staining.

Cytotoxic T lymphocyte (CTL) assay
Splenocytes were harvested from vaccinated or control animals and stimulated with
irradiated BM185 cells in the presence of murine recombinant IL-2 (10U/ml; source)
for 6 days at 37°C, under 5% CO
2
and humidified conditions. Splenocyte effector
cells were then co-cultured with
51
Cr-labeled BM185 target cells at various
effector:target (E:T) ratios for 4 h and percent specific target cell lysis was measured
as the mean values of released chromium minus spontaneous release devided by
total lysis minus spontaneous release.
94
Transformation of B6 bone marrow
Bone marrow was harvested from 6-8 week old male C57Bl/6ByJ mice (Jackson
Laboratories, Bar Harbor ME) and pre-stimulated for 48 hours with murine
recombinant IL-3 (10 ng/ml; BioSource, Camarillo CA), IL-6 (25 ng/ml; R&D,
Minneapolis, MN) and Stem Cell Factor (SCF, 2.5 ng/ml; Biosource) suspended in
basal bone marrow medium (BBMM; IMDM medium with 30% FBS, 1% bovine
serum albumin (BSA), L-glutmanine, pen/strep and β-mecaptoethanol). Cells were
transduced twice with the retroviral vector MND-p185 expressing the p185 BcrAbl
fusion gene at an MOI of 1 and 0.1, respectively, for 12 hours on retronectin and in
the presence of IL3/IL6/SCF. Cells were cultured for 14 days in IMDM with growth
factors and suspension and adherent cells or suspension cells alone were
transferred to R10 medium (RPMI, 10% FBS, L-glutamine, pen/strep). A fraction of
freshly transduced cells or cells cultured for 14 days were injected i.v. at 0.5 x10
6
or
1x10
6
, respectively, into sublethally irradiated (450 rad) C57Bl/6 recipients treated
with acid water and Maxim-200 and observed for survival for 180 days. All cell
cultures were maintained for additional 60 days with weekly subcultering at 1x10
5
/ml.

Southern Blot
Genomic DNA was isolated form transduced B6 bone marrow cultures by phenol
chloroform extraction. DNA from BM185 cells known to have a single BcrAbl copy
per cell was used undiluted (1 copy/cell) or diluted into DNA of BcrAbl negative
murine 3T3 fibroblasts (0.3, 0.1 copies per cell) as a copy number standard. For
each lane, 15 μg of DNA were simultaneously digested with the restriction enzymes
Hind-III and Eco-R1, separated on a 0.8% agarose gel and blotted onto a nylon
95
membrane (Micron Separation Inc, Westborough MA) by 20x SSC wet transfer. The
blot was hybridized with a
32
P-dCTP labeled (Random Primer Labeling Kit,
Stratagene) Stu-I/Xho-I probe (542 bp) specific for the Brc region of the BcrAbl gene,
washed and exposed to Kodak X-Omat film.
96
4.3- Results

Transplantation and vaccination schedule
Balb/c mice were transplanted, vaccinated, challenged and analyzed according to
the schedule presented in Fig 4.1. In brief, mice were given a total body irradiation
(TBI) of 900 rad and rescued with a syngeneic bone marrow transplant (BMT) or a
stem cell transplant (SCT). Lineage-depleted bone marrow cells used for the SCT
were about 100-fold enriched by negative magnetic column selection and given at
1x10
4
cells per mouse, 100 fold less than the BMT dose of 1x10
6
cells. Mice were
allowed to recover for 30 days (1 month post-transplant) or 90 days (3 months post-
transplant) and two mice of each group plus untreated control mice (naïve) were
sacrificed and analyzed for numerical and functional immune reconstitution just
before the time point of vaccination (pre-vaccine). The remaining mice were
vaccinated twice with irradiated, CD80, GM-CSF and CD40L-expressing BM185
(TripleVax) cells at day -14 and day -7 before challenge. Age-matched control mice
that did not receive any transplants previously were vaccinated as a control (Vax).
On day 0, two additional mice were sacrificed per vaccination group, including
untreated control mice (naïve), and analyzed for immune reconstitution and function
(post-vaccine). The remaining mice were challenged with the pre-B leukemia cell
line BM185 and observed for event-free survival up to 60 days after challenge.

97

Fig 4.1: Schematic overview of experimental design
Two cohorts of mice received bone marrow transplants after irradiation: whole bone
marrow cells (BMT, group (1)) or lineage-depleted progenitor cells (SCT, group (2))
followed by two vaccinations with irradiated leukemia cells either one or three months
post transplant and a challenge with live leukemia cells. A non-transplanted vaccination
(Vax) control group (3) received vaccinations and the leukemia challenge according to
the same schedule; a naïve control group (4) received the leukemia challenge only. At
the time point of the first scheduled vaccination, subgroups of mice were sacrificed and
analyzed for immune reconstitution; at the time point of the challenge subgroups of mice
were sacrificed and analysed for specific responses to leukemia cells. Challenged
animals were monitored for event-free survival (EFS) for 60 days afer challenge.

Radiation dose and post-transplant survival
Based on previous publications, the TBI dose leading to maximal myloablation and
engraftment and minimal transplant-related deaths in Balb/c mice was set to 900
rads. When rescued with 1x10
6
whole bone marrow cells, 90% of mice survived the
irradiation dose that was lethal after 15 days in the absence of bone marrow rescue
(Fig 4.2a). However, only 40% of mice irradiated with the same dose could be
rescued with an SCT of 1x10
4
lineage-depleted cells. To minimize radiation-induced
98
toxicity, split doses of TBI were given 18 hours apart with the first dose consisting of
600 rads and the second dose given in increments as 150, 300, 450 and 600 rads.
Mice were rescued by BMT and 100% survived the first 30 days post-transplant
(data not shown); animals were sacrificed after 30 days and the peripheral blood
was analyzed for its white blood cell (WBC) count. Increasing the second radiation
dose to 600 rad did not decrease blood levels of lymphocytes, granulocytes or
macrophages and thus could not further compromise peripheral blood reconstitution
(Fig 4.2b). But survival was drastically improved when mice were set on acid water
supply for five days prior to transplantation and treated with antibiotics for 14 days in
the post-transplant period, even when rescued with SCT after a single TBI of 900
rads (not shown), and 900 rads with concomitant medication was used as the
standard protocol in subsequent experiments.

Immunophenotypic reconstitution in peripheral organs
Mice were sacrificed either before or after the scheduled vaccinations after a
recovery period of 1 or 3 months post-transplant, and thymus, spleen and peripheral
blood tissues were analyzed for the absolute and relative numbers of lymphoid and
myeloid populations. The percentage of CD4/CD8 double negative and double
positive T cell progenitors and mature single positive CD4
+
and CD8
+
in the thymus
was similar in all animals examined (Fig 4.3a) with only minor variations. The total
number of thymocytes was found to be higher in the 1 month post-transplant group
(pre- and post-vaccine) compared to the 3 month post-transplant group, and thymic
hypercellularity can probably be attributed to the younger age of the first group.
However, no major differences were detected when transplanted animals were
99
A
0
20
40
60
80
100
0 5 10 15 20 25 30
Days after Transplant
Survival [%]
no transplant
BMT
SCT

B
0
2
4
6
8
10
12
14
600 + 150 rad 600 + 300 rad 600 + 450 rad 600 + 600 rad
Radiation Dose
Total cell number [10
3
/ μl]
Granulocytes
Monocytes
Lymphocytes

Fig 4.2: Irradiation dose titration and post-transplant survival
A) Balb/c mice (n=20) were irradiated with 900 rad and immediately transplanted with 10
6

whole bone marrow cells (BMT, grey triangles) or 10
4
Iineage-depleted bone marrow
cells (SCT, white circles) or left untransplanted as a control (no transplant, dotted line;
n=1). Mice were observed for survival up to 30 days after transplant.
B) Balb/c mice were set on acid water supply for 5 days prior to irradiation with a split
dose of 600 rads followed by the indicated dose (150, 300, 450 or 600 rads) on the
following day (n=3). Mice were transplanted immediately after receiving the second dose
of radiation with 1.5x10
6
syngeneic whole bone marrow cells and treated with
tetracycline for 14 days after transplant. Mice were sacrificed 30 days post-transplant
and white blood cell count was measured.

100
compared to age-matched, non-transplanted controls (Naïve or Vax control),
indicating that numerical thymic reconstitution was completed as early as 4 weeks
after full (BMT) or T-cell depleted (SCT) transplantation.

In the spleen, severe lymphocyte deficiencies were detected at the early 1 month
post-transplant time point (Fig 4.3b). Before the vaccine, the total lymphocyte
numbers were reduced to 50% in BMT recipients compared to the numbers present
in naïve spleens, and to 10% in SCT recipients. Two weeks later after the second
vaccination, those numbers recovered to 70% in BMT recipients and to 50% in SCT
recipients and thus were still below normal. Although the impact of transplantation
on absolute counts was quite pronounced, the relative proportions of B cells and
CD4
+
or CD8
+
T cells remained mainly unaffected. Three months after transplant,
absolute numbers were normalized and there was no major difference between
absolute B cell count prior to vaccination (T cell count was not determined) and
absolute lymphocyte count after vaccination across transplant and non-transplant
recipients, demonstrating that complete numerical recovery could be achieved
within 3 months.

White blood cell (WBC) counts in the peripheral blood (Fig 4.3c) were consistently
found within a normal range of 8-12 x10
3
cells/ μl, and no leukopenia was detected
at any time point analyzed after transplant (1 month, 3 months). Stem cell transplant
recipients, however, showed a profound reduction in absolute and relative B cell
numbers accompanied by an increase in cells of the myeloid lineage, i.e.
macrophages and granulocytes, a distortion that was less apparent in BMT
101
recipients. After a recovery time of 3 months after transplantation, leukocyte counts
in all transplant recipients were found to be normalized to levels measured in non-
transplanted controls. Relative numbers of NK cells were reduced in all animals
analyzed in the pre-vaccine setting after 3 months regardless of the transplantation
status, but this apparent deficiency was not confirmed in the post-vaccine analysis
including naïve controls, and numbers were comparable to those found across the 1
month post-transplant analysis.

Overall, the numbers and distribution of various leukocyte subgroups was found to
be almost normal in the thymus and peripheral blood as early as 4 weeks after
whole bone marrow or lineage-depleted bone marrow transplantation, with the
exception of slow engraftment kinetics of lymphocytes detected in the spleen.

Functional recovery of splenic effector cell populations
Natural killer (NK) cell activity was measured by incubating splenocytes with the
murine lymphoma cell line Yac-1 which is sensitive to the lytic action of NK cells (Fig
4.4a). The 1 month post-transplant data set was only generated for animals
sacrificed after their second vaccination, and neither transplantation nor vaccination
treatment affected splenic NK cell activity which resulted in about 30% lysis of
added Yac-1 target cells. The 3 month post-transplant data set included both pre-
and post-vaccine sample groups. The pre-vaccine cohort showed very little lytic
activity of 10% or less, even in naïve mice which should not have impaired NK cell
activity compared to naïve mice analyzed at different time points. The reduced NK
function, however, correlates with the reduced number of NK cells found in the
102
peripheral blood of those animals (Fig 4.3c) and may therefore simply reflect a lower
abundance of effector cells in the spleen. NK-mediated target cell lysis after
vaccination was found to be in the range of 20-50%, similar to the post-vaccine
analysis after 1 month. In this case, a slight mitogen Concanavalin A (ConA) which
induces non- specific TCR signaling and proliferation in T lymphocyte populations
(fig 4.4b). One month after transplant, a 5-fold expansion of ConA- stimulated cells
over unstimulated cells was observed in the naïve, BMT and SCT group.
Vaccination induced an increase of T cell proliferation in non-transplant recipients
(up to 15-fold), but stimulation was not significantly altered in BMT or SCT recipients
compared increase in NK cell activity was observed in vaccinated animals
compared to the control group, and this increase was more pronounced in
untransplanted vaccinated mice compared to transplant recipients.

The proliferative capacity of T cells was determined by the stimulation of
splenocytes with the to naïve mice. This shows that in addition to reduced numbers
of splenic T cells 1 month after transplant, functional recovery of those cell may also
be slightly impaired. ConA stimulation was also tested 3 months after transplant, but
the results are questionable because no proliferation could be induced in any test
group (stimulation index ≤1) which may indicate that the ConA used in this
experiment had lost its activity.

103

Fig 4.3: Reconstitution of leukocyte subpopulations in various peripheral organs
Balb/c mice were given 900 rad TBI and 1x10
6
syngeneic whole (BMT) or lineage-depleted
(SCT) bone marrow cells. Mice were either sacrificed 30 or 90 days after transplantation
(pre-vaccine) or vaccinated twice with 1x10
6
irradiated TripleVax one week apart and
sacrificed 7 days after the last vaccination (post-vaccine). Untransplanted, non-vaccinated
(naive) and untransplanted, vaccinated mice (Vax) served as the appropriate controls.
A) Thymocytes were harvested, counted and analyzed by FACS for surface expression of
CD4 and CD8. Bars show the mean of total numbers of CD4
-
CD8
-
double negative (DN),
CD4
+
CD8
+
double positive (DP) and CD4
+
or CD8
+
single positive cells in the thymus (n=2;
except SCT, pre-vaccine, 1 month post-BMT: n=1).
B) Splenocytes were analyzed by FACS for the number of CD3
+
CD4
+
T cells, CD3
+
CD8
+
T
cells and CD45
+
CD19
+
B cells. Bars show the means of total numbers found in the spleen
(n=2; except SCT, pre-vaccine, 1 month post-BMT: n=1). The 3 months post-BMT, pre-
vaccine analysis includes data only for CD45
+
CD19
+
B cells; CD3, CD4 and CD8 surface
staining was not determined in this set.
C) Peripheral blood was analyzed for white blood cell count (WBC) and stained with
fluorescent antibodies for FACS. Bars represent the means of total numbers of T cells
(CD3
+
CD4
+
or CD3
+
CD8
+
), B cells (CD45
+
CD19
+
), macrophages and granulocytes
(CD45
+
M/P
+
), NK cells (CD3
-
NK
+
) and NKT cells (CD3
+
NK
+
); n=2; except SCT, pre-vaccine,
1 month post-BMT: n=1; SCT, post-vaccine, 1 month post-BMT: WBC only, FACS not
determined; 3 months post-transplant, pre-vaccine: CD3
+
NK
+
(NKT cells) not determined.
104
A 1 month post-transplant 3 months post-transplant

B 1 month post-transplant 3 months post-transplant

C 1 month post-transplant 3 months post-transplant

105
In addition to general effector function of NK and T cells, specific immunity of
cytotoxic T lymphocytes (CTL) against BM185 leukemia cells was assessed (Fig
4.4c). In vitro incubation of splenocytes with irradiated BM185 cells in the presence
if IL-2 allowed expansion of BM185-specific T cells that were activated during the in
vivo immunization with TripleVax. The percent of radioactive release from labeled
BM185 target cells indicated the activity of tumor-specific T cell present in each
spleen analyzed. Vaccination of untreated control mice (Vax) induced 15-20%
specific target cell lyis in both CTL assay (1 month and 3 months post-transplant) at
the highest E:T ratio, whereas splenocytes from non-vaccinated mice (Naïve) did
not show any significant activity. Interestingly, at both time points after
transplantation (1month and 3 months), SCT recipients showed the highest CTL
activity even compared to non-transplanted vaccination controls (Vax), whereas
CTL activity in the BMT group was reduced compared to the Vax control group. In
summary it can be concluded that the function of effector cells was not drastically
impaired after transplantation, even though the numerical recovery of lymphocytes
in the spleen was shown to take more than 4-6 weeks.

Survival of leukemia challenge after vaccination in the post-transplant setting
Vaccination of untreated mice with TripleVax on day -14 and -7 prior to challenge
led to 40% rejection of a BM185 leukemia dose that was otherwise fatal in non-
vaccinated, naive controls, a survival rate which is consistent with the data we have
previously observed in our laboratory (fig 4.5 and data not shown).

106

Fig 4.4: Functional analysis of splenic effector cell populations
Mice were transplanted with whole bone marrow (BMT) or lineage-depleted bone marrow
(SCT) and either sacrificed 30 or 90 days after transplantation (pre-vaccine) or vaccinated
twice with 1x10
6
irradiated TripleVax one week apart and sacrificed 7 days after the last
vaccination (post-vaccine). Untransplanted, non-vaccinated (naive) and untransplanted,
vaccinated mice (Vax) served as controls.
A) NK cell activity assay. Yac-1 cells were pulsed with 1mCi/ml
51
Cr for 1 h and co-cultivated
with splenocytes at an effector/ target ratio (E:T) of 50:1. Supernatants were harvested and
counted for
51
Cr release in a Gamma-counter. NK activity is defined as percentage of Yac-1
target cell lysis obtained by specific lysis divided by total lysis after subtraction of
background (spontaneous lysis).
B) ConA Stimulation of splenic T cells. Splenocytes were seeded at three different initial
concentrations (1.0, 1.5, and 2.5 x10
6
/ml), cultured in the presence or absence of 7.5 ng/ml
Conconavalin A (ConA) for 72 h, pulsed with 50 μl
3
H-Thymidine for 16 h and counted in a
Beta-counter. Stimulation indices show the ratio of
3
H-Thymidine incorporation between
ConA-stimulated cells and unstimulated cells.
C) Cytotoxic T lymphocyte assay. Splenocytes were stimulated with irradiated BM185 cells
and recombinant mIL-2 for 5 -6 days. Stimulated cells were incubated with
51
Cr-labeled
BM185 cells for 4 h at various effector/ target (E:T) ratios and supernatants were analyzed
for
51
Cr release in a Gamma-counter. Specific target cell lysis was determined as the ratio of
the mean of experimental counts over the mean of total lysis.
107
A
B 1 month post-transplant 3 months post-transplant
0
2
4
6
8
10
12
14
16
Naïve BMT SCT Naïve BMT SCT Vax
post-vaccine post-vaccine
ConA Stimulation Index
0
2
4
6
8
10
12
14
16
Naïve BMT SCT Naïve BMT SCT Vax
pre-vaccine post-vaccine
ConA Stimulation Index

C 1 month post-transplant 3 months post-transplant
0
5
10
15
20
25
100 30 10 3 1 0.3 0.1
E:T Ratio
Specific BM185 Cell Lysis [%]
Naïve
BMT
SCT
Vax
0
5
10
15
20
25
100 30 10 3 1 0.3 0.1
E:T Ratio
Specific BM185 Cell Lysis [%]
Naïve
BMT
SCT
Vax

108
Mice receiving vaccination 1 month after BMT or SCT were protected from leukemia
development similarly to the Vax control (Fig 4.5a). Survival was even improved with
vaccination given 3 months post-transplant (fig 4.5b), leading to a rejection of the
leukemia challenge in 60% of all transplanted mice (BMT and SCT). Despite initial
deficiencies in numerical or functional immune reconstitution, the post-
transplantation setting may be beneficial for the establishment of cancer-specific
immunity especially after a prolonged period of recovery.

Transduction of C57Bl/6 bone marrow with p185 Bcr-Abl
The C57Bl/6 (B6) mouse strain has several advantages over the Balb/c strain that
was used in this study because most of the transgenic strains that can be found are
in the B6 background. We sought to transform B6 bone marrow with the p185 Bcr-
Abl oncogene, particularly because donor chimerism can easily be followed in
CD45.1/CD45.2 congeneic recipient/donor pairs. Bone marrow was harvested from
B6 mice, prestimulated with murine recombinant IL-3, IL-6 and SCF in basal bone
marrow medium (BBMM) and transduced twice with MND-p185 retroviral vector 12
h apart. Three sublethally irradiated B6 mice (400 rad) were injected i.v. with
0.5x10
6
freshly transduced cells or control BM and observed for survival for 180
days. The remaining BcrAbl-transduced B6 BM cells (BBB) were cultured in BBMM
for 14 days and either injected into sublethally irradiated B6 mice (1x10
6
) or
switched to R10 medium. Because it has been reported that stromal cells present in
primary bone marrow cultures support the growth of cells with a pre-B cell
phenotype, suspension cultures were maintained either on top of the primary
adherent cell layer or transferred to fresh wells without adherent cells.
109
A
0
20
40
60
80
100
0 10203040506
Days after Challenge
Survival [%]
0
Naïve
Vax
BMT
SCT

B
0
20
40
60
80
100
0 10203040506
Days after Challenge
Survival [%]
0
Naïve
Vax
BMT
SCT

Fig 4.5: Survival of challenge after vaccination post-BMT
Balb/c mice were transplanted with 1x10
6
whole bone marrow cells (BMT; n=), 1x10
4

lineage-depleted bone marrow cells (SCT) or left un-transplanted (Naïve, Vax). Mice were
vaccinated with 1x10
6
irradiated TripleVax cells 30 or 90 days after transplant 14 and 7
days prior to challenge with 1x10
3
live wt-BM185 leukemia cells. Untransplanted control
mice were either vaccinated (Vax) or left untreated and likewise challenged with leukemia.
Mice were observed for event-free survival up to 60 days after challenge.
A) Vaccinations and challenge were given 1 month after transplant. Naïve: n=11; BMT:
n=12; SCT: n=9; Vax: n=13.
B) Vaccinations and challenge were given 3 months after transplant. Naïve: n=2; BMT:
n=5; SCT: n=3; Vax: n=6.

110
Suspension cells were counted weekly and depopulated to 1x10
5
cells. Fig 4.6a
shows the expansion of BBB cells per week after transduction; non-transduced
bone marrow cells cultured in BBMM did not survive the first 14 days after harvest
and are thus not included in the figure. Initially, the expansion rate of transduced
cells increased and peeked at 5 weeks post-transduction. At this point, cells
expanded 3.5-fold within 7 days, indicating that doubling time of cells was 48 hours.
The expansion rate then steadily decreased over the following weeks in culture and
reached a plateau of about one doubling per week; there was no difference in
growth characteristics between BBB cells cultured on adherent cells vs. pure
suspension cultures. Taken together, BBB cells, although surviving long-term in
culture, never acquired the rapid growth kinetics characteristic for Ph
+
ALL cells.

To test whether the p185 transgene was detectable in BBB cells over time, cells
were harvested at different time points immediately after transduction (7, 10, 14
days) or after 60 days of long-term culture either on adherent layers (adh) or as pure
suspension cultures (sus; Fig 4.6b). At the early time points after transduction,
BcrAbl was clearly detectable as an intact proviral unit in BBB cells at a copy
number of 0.1-0.3 per cell, but after 60 days of long-term culture, multiple bands
were detected. It can be concluded that cells tranduced with BcrAbl did not have a
selective advantage because no accumulation of cells with an intact transgene
could be demonstrated. All animals challenged with BBB cells survived for at least
180 days post challenge and did not develop leukemia, indicating that transduced
cells did not develop a growth advantage in vivo either (data not shown).

111
A
0
1
2
3
4
5
6
7
3 456 78 9 10 11 12
Weeks after Transduction
Fold expansion

B

Fig 4.6: BcrAbl-transduced bone marrow
Bone marrow from C57Bl/6 mice was transduced with MND-p185 vector (B6-BcrAbl
Bone marrow, BBB), cultured in basal bone marrow medium with IL-3, IL-6 and SCF for
7, 10 or 14 days and maintained in R10 medium as a suspension culture with or without
support of adherent feeder cells for 60 days in vitro. (A) In vitro expansion of transduced
BM cells. Cultures were counted weekly and depopulated to 1x10
5
cells/ml. The graph
shows the fold expansion measured right before splitting, 7 days after the previous split,
as an average of 5 wells (24 well plates) grown with or without feeder cell support.(B)
Southern Blot for BcrAbl proviral integrants. BBB cells were harvested 7, 10 or 14 days
after transduction or cultured for additional 60 days in R10 medium in the presence (adh)
or absence (sus) of adherent feeder cells. Genomic DNA from BM185 cells (copy
number standard), NIH-3T3 cells (Co) or transduced BBB cells was digested in a double
digest reaction with the restriction endonucleases EcoRI and HindIII , electrophorezed
and hybridized with a radioactively labeled probe specific for the Bcr gene. The 1.8 kb
band represents an endogenous Bcr fragment.
112
4.4- Discussion

Allogeneic transplants have clinically been proven to be more effective for the cure
of hematologic malignancies than autologous transplants because of the graft-
versus-leukemia (GVL) effect provided be allo-reactive T cells in the graft which at
the same time can provoke graft-versus-host-disease (GVHD). In our murine model
we chose a syngeneic transplantation setting that provided greater stringency
because any antitumor activity observed is attributable to specific recognition of
tumor-associated antigens rather than minor histocompatibility antigens. T cell
depletion of autologous graft is usually not necessary because there is no risk of
GVHD toxicities, but we performed lineage-depletion of bone marrow grafts in
addition to whole cell bone marrow transplantation to delay immune reconstitution.

Patients undergoing chemoradiation therapy and bone marrow transplantation are
usually severely immunodeficient immediately after the procedure, and immune
function is gradually restored over the following 1-2 years. The regeneration of the T
cell compartment is particularly delayed, although specific immunity against some
infectious diseases can be observed within the first months of recovery (Li 1994;
Hata 2002; Gandhi 2003). Immune reconstitution in mice is considerably faster,
leading to an almost complete restoration within 4-8 weeks after transplantation,
depending on the model. We decided to design our murine model very stringently to
mimic clinical immunodeficiency as closely as possible. Only 1x10
6
unfractionated
bone marrow cells were given after TBI without addition of splenic T cells to adjust
for the relative lack of mature T cells in murine bone marrow compared to the T cell
113
content in human bone marrow aspirates (Mundhada 2005). The dose of lineage-
depleted bone marrow cells (1x10
4
) was the minimal dose we could successfully
administer to rescue mice from 900 rads of irradiation without losing too many
animals to transplant-related mortality. Despite our efforts, cellular reconstitution of
the thymus and blood was completely achieved after 4 weeks with the exception of
profound hypocellularity of the spleen, and complete numerical and functional
recovery was observed after 3 months in all organs examined.

In humans, NK cells are among the first cell to recover, and levels normalize within
the first 1 month after transplantation (Guillaume 1998). NK cells may play an
important defense against infections or tumor relapse, but no correlations between
NK cell recovery and prolonged event-free survival have been established yet. In
our study, full NK cell activity was detectable in transplanted mice at the early and
late time points analyzed, and vaccination did not have any major effect on the
ability to lyze the NK responsive Yac-1 target cell line. This is in agreement with
previous reports that post-BMT anti-tumor activity was not detected in nude mice
which lack mature T and B cells but retain NK cell function (Borello 2000), and many
experiments in thymectomized mice indicate that anti-tumor responses are mainly
dependent on the T cell compartment in mice (Borello 2000, Mundhada 2005). T cell
numbers in the spleen were extremely low 1 month post-transplant in our study, but
proliferation in response to ConA stimulation and specific CTL activity was
detectable and not significantly impaired compared to controls.

114
To address whether leukemia-specific immune responses could be initiated after
bone marrow transplantation, mice were vaccinated with irradiated BM185 TripleVax
cells and subsequently challenged with live BM185 leukemia cells. Normalization of
splenic T cell numbers was apparently not a prerequisite for anti-tumor activity
because specific in vitro tumor cell lysis was not only detected in transplanted
animals but was increased in splenocyte cultures obtained from SCT recipients
compared to non-transplanted vaccination controls. Consequently, the survival rate
of transplanted animals was either comparable to non-transplant recipients in the 1
month post-transplant vaccination setting or even improved after vaccinations
scheduled at 3 months post-transplant.

Similar results were reported in a GM-CSF-based tumor vaccine model for murine
lymphoma: vaccinations administered to Balb/c mice 3-6 weeks after BMT
increased survival rates compared to vaccinated no-BMT controls, even though T
cell recovery in peripheral lymph nodes was not fully achieved at that time (Borello
2000). The mechanism for this observation is not clear, but it may be that the
regeneration of a new T cell population allows for increased expansion of antigen
specific T cells without competition for survival signals. Inhibitory T cells may also be
absent at that time point, especially after T cell depletion of the graft, so that host
tolerance to the tumor cells does not have to be broken before the vaccine becomes
effective. GM-CSF secreted by TripleVax cells may in addition enhance immune
recovery and therapeutic outcome by facilitating T cell engraftment and homeostatic
expansion (Zhao 2006).

115
The enhancement of cellular immune function in a post-BMT setting to induce anti-
tumor immunity seems to be well feasible in the murine model organism. Whether
these findings are suitable for the clinical setting remains to be tested, and murine
findings should be translated with caution due to the profound differences between
mice and humans. One caveat for instance is that thymic function is well preserved
in mice at the time point of transplantation, whereas patients show drastic age-
related decreases in thymic output leading to impairment of T cell recovery. It may
be speculated that vaccinations will be more efficient in pediatric patients in which
thymic involution has not proceeded to the same degree as in older patients.

Nevertheless, the need to find alternative therapies for high-risk leukemia patients
continues. In an attempt to refine our murine model for autologous BMT in Ph
+
ALL,
we sought to develop Bcr-Abl transformed cell lines syngeneic to the C57B/6
background to enable more thorough chimerism studies. It appears that B6 bone
marrow is not as easily transformed as Balb/c marrow because transduced cells
were merely immortalized but did not acquire the rapid growth kinetics that
characterize aggressive Ph
+
ALL. Balb/c mice may have a genetic background that
makes their cells more susceptible to transformation, and it has been shown for B6
mice that Bcr-Abl mediated transformation is directly correlated with heterozygous
or homozygous loss of the tumor suppressor gene p14
Arf
(Williams 2006). However,
when we repeated our transduction studies with Arf
+/-
B6 mice, ALL development
was not accelerated in our hands (data not shown) and we were not able to
establish B6 syngeneic ALL cell lines so far.

116
Despite several hurdles that still have to be overcome in pre-clinical and clinical
settings, immunotherapy for cancer remains a promising option that needs to be
further optimized. The conclusion drawn from our and other animal studies is that a
post-BMT might be a beneficial setting for tumor vaccination not only because of a
permissive environment for clonal T cell expansion but also because of the minimal
residual tumor burden present immediately after bone marrow transplantation.
117
Chapter 5- Concluding Remarks

5.1- Gene therapy for congenital hematological diseases

The great appeal of gene therapy for monogenic disorders is the promise to treat
the disease at its origin by correcting the underlying genetic defect. Genetic
correction of SCID was the first reported success of a clinical gene therapy trial, and
sustained restoration of immunity was achieved in the majority of pediatric patients.
Gene therapy shortly received a bad reputation because of the severe adverse
events reported that put all similar trials temporarily on hold. However, every
disease treatment has a risk-benefit profile, and the success and failure of gene
therapy has to be evaluated in the close context of other available treatment options
for SCID that were inadequate for the patients enrolled in the gene therapy trials at
the time, and that are often associated with severe side effects themselves.

The risk-benefit assessment of gene therapy for hematological disorders is an
important consideration because of the severity of the side effect, cancer
development. Quantitative pre-clinical models need to be developed in order to
predict the minimal number of viral insertions required for a clinical effect and the
associated frequency of oncogenic transformation (reviewed in: Porteus 2006).
Murine cancer susceptibility models seem to be the most stringent and appropriate
models for this type of evaluation, and it seems that the risk/benefit ratio has to be
carefully addressed for every single disease treated because it is suspected that the
118
therapeutic transgene may play a co-transforming role in the cancer development
process.

A variety of different approaches are currently investigated in an effort to reduce the
risk of insertional oncogenesis. Some aim at the optimization of vector design, for
instance minimizing viral LTR activity by switching to lentiviral vectors with internal
transgene promoters or including chromosomal insulator elements. Others opt to
reduce transgene activity by tailoring expression to the target cell type with lineage-
specific promoters, although the significance of this approach is uncertain because
insertional mutagenesis in the few clinical cases reported has been limited to the
cell lineages with the underlying genetic defects. A third line of optimization efforts
focuses on reducing the rate of random integration by transgene delivery through
transposon elements in combination with site-specific integrases or by direct
replacement of the endogenous gene using zinc finger endonucleases.

Despite all attempts to reduce the rate of insertional activation of cellular proto-
oncogenes, random integration will never be completely eliminated in any gene
delivery system. Cancer development is a multi-step process that progresses with
the cumulative acquisition of genetic lesions, and any time the chromosomal
integrity becomes interrupted it can represent a potential hit. The most careful
shielding of the chromosomal environment from proviral transactivation will not
prevent interruption of tumor suppressor genes which will increase cancer
susceptibility in the targeted progenitor and its lineage. And double strand breaks
induced by site-specific integration may lead to chromosomal translocations. Gene
119
therapy will therefore always bear a minimum risk of tumorigenesis that
counterbalances the clinical success of this form of treatment, and additional safety
features that reverse potential tumor development are highly warranted. The
inclusion of a suicide gene in the therapeutic vector is a fail-safe mechanism that
can be exploited for the eradication of vector-induced transformation as detailed in
this report. Suicide genes should therefore be viewed as synergistic method to
improve gene therapy in addition to all the improvements mentioned above.
The greatest drawback of suicide gene therapy is the potential loss of the entire
bone marrow graft upon treatment with the drug substrate. Transduction efficiencies
of hematopoietic stem cells are steadily increasing during the evolution of improved
vector generations. If a transduction rate of 100% of modified and re-infused CD34
+

cells is achieved, it can be hypothesized that Ganciclovir treatment of a gene
therapy patient diagnosed with a transgene-related malignancy has the power to
completely abrogate the entire bone marrow compartment without leaving a stem
cell reservoir for immune reconstitution. Although the transduction rate achieved in
previous clinical trials never surpassed 60%, clinical protocols for suicide gene
therapy will have to be carefully developed to account for this hypothetical event, for
example by storing a reserve of un-modified CD34
+
cells. More animal studies have
evidently to be conducted to address the feasibility of a secondary autologous bone
marrow transplantation following suicide gene-mediated depletion of a primary
transplant before clinical protocols can be implemented. The potential of suicide
genes to tip the balance of the risk-benefit profile of gene therapy for monogenic
disorders in favor of the clinical benefits is a powerful promise to help paving the
rocky road of the future of gene therapy.
120
5.2- Immunotherapy with modified cancer cell-based vaccines

The goal of immunotherapy is to stimulate the patient’s immune system to recognize
and destroy cancer cells. The greatest challenge in specifically targeting cancer
cells while leaving healthy tissue unharmed is that tumor cells derive from self, and
differences between the tumor and the tissue of origin are often subtle and not well
defined. The ongoing trends in immunotherapy for cancer therefore fall into two
different categories: therapies for cancers with known tumor-specific antigens and
for cancers with specific and highly immunogenic markers yet to be identified.

A number of tumors exist that are known to express unique immunodominant
surface antigens, such as MART-1 for melanoma or Her2/Neu for breast cancer.
These markers can be targeted by universal cell-based therapies that avoid the
development of costly individualized tumor vaccines derived from autologous tumor
cells. Instead, cells of the immune system can be modified, like peptide-pulsed or
antigen-transduced dendritic cells, or transgenic T cells transduced with a cloned T
cell receptor that shows a high avidity and specificity for antigens like MART-1
(reviewed in: Biagi 2007). Many of these cells can be maintained as stable cell lines
and applied to many patients as a cancer treatment.

For most cancers, however, tumor-specific antigens capable of inducing a strong
immune response are currently unknown, and whole cell vaccines with irradiated
autologous cancer cells will remain the immunotherapy method of choice. Even for
acute lymphoblastic leukemia which expresses the novel p185 BCR-ABL fusion
121
epitope, no immunodominant peptide has been found to bind to the most prevalent
human MHC class I allele (HLA-A2.1), although such immunopeptides have been
identified for the p210 variant of BCR-ABL found in chronic myeloid leukemia.
Whole cell vaccines have the major advantage of delivering a diverse repertoire of
multiple potential tumor antigens, limiting the risk of immunoediting and therapeutic
escape. The immune response in general is tightly balanced, and therapies
intervening at only a single point of the complex mechanisms are not very likely to
succeed. This holds true for gene transfer of immunostimulatory genes to modified
cellular vaccines as well, and the clinical trend will most likely go towards
combination of various cytokines and co-stimulatory molecules in a vaccine modality,
just like we tried to address with a combination of GM-CSF, CD40L, CD80 and IL-12.

One aspect of cellular immunity that is easily overlooked when developing a vaccine
is central and peripheral tolerance. Since tumor antigens are often slightly modified
versions of self-peptides, high avidity T cell undergoing positive selection during
thymic education may have differentiated into regulatory T
reg
cells of the CD4
+

CD25
+
Foxp3
+
phenotype that circulate in the blood stream and suppress specific T
cell activation. Even if the tumor-specific epitope was not presented on thymic
epithelial cells, tumor-specific activated T cells encounter a high and persistent
antigen load that may induce peripheral tolerance and render the T cells
unresponsive. More effective cancer vaccines have therefore to be developed that
augment specific immunity to cancer by breaking immune tolerance. Anti CTLA-4
antibody therapies depleting T
reg
populations have been a promising approach, and
combining tolerance suppressing agents with DC-stimulating adjuvants like IL-12
122
should be evaluated for gene therapy combination modalities in the future to further
improve the efficacy of immunotherapy.

Gene therapy will not only benefit from the combination of various transgenes to
maximize efficiency, but also from combinatorial trials rationally integrating cellular
immunotherapy with standard chemotherapies (reviewed in Emens 2006). The
persistent high antigen load derived from self molecules as presented on cancer
cells is likely to induce tolerance and allow tumor cells to escape from
immunosurveillance. If this antigen load is temporarily reduced by chemotherapeutic
and radiation treatment and existing tolerized T cells are eradicated, it is expected
that the immune system will eliminate recurring cancer more efficiently once it is
functionally reconstituted. Our vaccination studies in a post-BMT setting suggested
an advantage of T cells developing in the ablated post-BMT environment compared
to non-transplant recipients. Furthermore, antigen persistence at high
concentrations may inhibit differentiation of activated CD8
+
T cells into memory
effector cells, and effects of BMT after myeloablation on the memory formation in
response to cancer vaccines can be further studied in appropriate animal models
such as our post-BMT model.

Ultimately, the trend in gene therapy will be to target truly tumorigenic cancer stem
cells. Mounting evidence suggests that cancer development is a recapitulation of
normal tissue development gotten out of control, and that a hierarchy exists among
cancer cells that is comparable to the tissue of origin. Cancer stem cells either
maintained or re-acquired self-renewal capacities and give rise to more
123
differentiated cells that comprise the tumor mass but have a limited life span and
lost their tumorigenic potential. Conventional therapies target only the rapidly
dividing progenitor cells without affecting the primarily quiescent stem cell
population, which accounts for the high relapse rates experienced for many cancer
types. Immunotherapy has the great advantage that it can target any cell that
presents specific antigens independently of its mitotic state. Cellular vaccines with
autologous stem cells would clearly not be feasible because the limited stem cell
number is not suitable for isolation, but targeting specific epitopes with modified
cells of the immune system may become reality if appropriate targets are identified.
Stem cell markers expressed on functionally defined cancer stem cells are currently
not known or shared with normal tissue stem cells, but the search for antigens
restricted to the stem cell population of a tumor mass and not expressed on healthy
tissue may lead to the development of very potent and highly tailored therapies.
Again, immunotherapy will most likely be most efficient as an adjuvant treatment in
conjunction with radio/chemotherapy or surgery for solid tumors because the bulk of
more differentiated tumor cells might mask the small number of cancer stem cells
that might be further sequestered by specialized stem cell niches. Novel genes may
also be evaluated for cancer gene therapy that inhibit stem cell pathways, block self-
renewal and exhaust the stem cell population through induction of mitotic
proliferation (Sell 2006).

Immunotherapy for cancer is a powerful treatment because it is highly specific for
malignant cells while exerting a mild side effect profile. Although efficacies need to
be improved for immunotherapy to be developed as a single treatment for human
124
cancers, it shows promising results as adjunct therapy together with current cancer
treatments. Thus, both cancer antigen-specific and whole cell cancer cell specific
treatment modalities warrant more research in the field of gene therapy to ensure
that aberrant cancer cells can be selectively targeted and eliminated in patients
suffering from a broad spectrum of human cancers.
125
Bibliography

Adris S, et al. (2000). Mice vaccination with interleukin 12-transduced colon cancer
cells potentiates rejection of syngeneic non-organ-related tumor cells. Cancer Res
6: 6696-703.

Aiuti A, et al. (2002). Correction of ADA-SCID by stem cell gene therapy combined
with nonmyeloablative conditioning. Science 296: 2410-3.

Badovinac VP, Porter BB, Harty JT (2004). CD8+ T cell contraction is controlled by
early inflammation. Nat Immunol 5:809-17.

Baum C, et al. (2003). Side effects of retroviral gene transfer into hematopoietic
stem cells. Blood 101: 2099-2114.

Baum C, Kustikova O, Modlich U, Li Z, Fehse B (2006).Mutagenesis and
oncogenesis by chromosomal insertion of gene transfer vectors. Hum Gene Ther
17: 253-63.

Biagi E, Marin V, Giordano Attianese GM, Dander E, D'Amico G, Biondi A (2007).
Chimeric T-cell receptors: new challenges for targeted immunotherapy in
hematologic malignancies. Haematologica 92:381-8.

Black ME, Kokoris MS, Sabo P (2001). Herpes simplex virus-1 thymidine kinase
mutants created by semi-random sequence mutagenesis improve prodrug-mediated
tumor cell killing. Cancer Res 61: 3022-6.

Blumenthal M, Skelton D, Pepper KA, Methangkool E, Kohn DB (2007). Effective
suicide gene therapy for leukemia in a novel model of insertional oncogenesis in
mice. Mol Ther 15: 183-92.

Bolotin E, Smogorzewska M, Smith S, Widmer M, Weinberg K (1996).
Enhancement of thymopoiesis after bone marrow transplant by in vivo interleukin-7.
Blood 88: 1887-94.

Bonini C, et al. (1997). HSV-TK gene transfer into donor lymphocytes for control of
allogeneic graft-versus-leukemia. Science 276: 1719-24.

Bordignon C, et al. (1995). Gene therapy in peripheral blood lymphocytes and bone
marrow for ADA
-
immunodeficient patients. Science 270: 470-5.

Borrello I, Sotomayor EM, Rattis FM, Cooke SK, Gu L, Levitsky HI (2000).
Sustaining the graft-versus-tumor effect through posttransplant immunization with
granulocyte-macrophage colony-stimulating factor (GM-CSF)-producing tumor
vaccines. Blood 95: 3011-9.
126
Cathcart K, et al. (2004). A multivalent bcr-abl fusion peptide vaccination trial in
patients with chronic myeloid leukemia. Blood 103: 1037-42.

Cavazzano-Calvo M, et al. (2000). Gene therapy of severe combined
immunodeficiency (SCID)-X1 disease. Science 288: 669-72.

Challita PM, Kohn DB (1994). Lack of expression from a retroviral vector after
transduction of murine hematopoietic stem cells is associated with methylation in
vivo. Proc Natl Acad Sci USA 91: 2567-71.

Challita PM, Skelton D, El-Khoueiry A, Yu XJ, Weinberg K, Kohn DB (1995).
Multiple modifications in cis elements of the long terminal repeat of retroviral vectors
lead to increased expression and decreased DNA methylation in embryonic
carcinoma cells. J Virol 69: 748-55.

Chalmers D, et al. (2001). Elimination of the truncated message from the herpes
simplex virus thymidine kinase suicide gene. Mol Ther 4: 146-8.

Chang AH, Sadelain M (2007). The genetic engineering of hematopoietic stem cells:
the rise of lentiviral vectors, the conundrum of the LTR, and the promise of lineage-
restricted vectors. Mol Ther 15: 445-56.

Check E (2005). Gene therapy put on hold as third child develops cancer. Nature
433: 561.

Chen Y, Lin SM, Lai HS, Tseng SH, Chen WJ (2002). Effects of tumor vaccine and
continuous localized infusion of granulocyte-macrophage colony-stimulating factor
on neuroblastoma in mice. J Pediatr Surg 37: 1298-304.

Cohen JL, et al. (1997). Prevention of graft-versus-host disease in mice using a
suicide gene expressed in T lymphocytes. Blood 89: 4636-45.

Colombo MP, et al. (1996). Amount of interleukin 12 available at tumor site is critical
for tumor regression. Cancer Res 56: 2531-4.

Davé UP, Jenkins,NA, Copeland NG (2004). Gene therapy insertional mutagenesis
insights. Science 303: 333.

De Giovanni C, et al. (2004). Immunoprevention of HER-2/neu transgenic mammary
carcinoma through an interleukin 12-engineered allogeneic cell vaccine. Cancer Res
64: 4001-9.

Dranoff G (2003). GM-CSF secreting melanoma vaccines. Oncogene 22: 3188-92.

Dunussi-Joannopoulos K, Runyon K, Erickson J, Schaub RG, Hawley RG, Leonard
JP (1999). Vaccines with interleukin-12-transduced acute myeloid leukemia cells
elicit very potent therapeutic and long-lasting protective immunity. Blood 94: 4263-
73.
127
Emens LA (2006). Roadmap to a better therapeutic tumor vaccine.
Int Rev Immunol 25:415-43.

Frank O, et al. (2004). Tumor cells escape suicide gene therapy by genetic and
epigenetic instability. Blood 104: 3543-9.

Fuchs EJ, Bedi A, Jones RJ, Hess AD (1995). Cytotoxic T cells overcome BCR-
ABL-mediated resistance to apoptosis. Cancer Res 55: 463-6.

Fuji N, et al. (1999). Augmentation of local antitumor immunity in the liver by tumor
vaccine modified to secrete murine interleukin 12. Gene Ther 6:1120-7.

Gandhi MK, Wills MR, Sissons JG, Carmichael AJ (2003). Human cytomegalovirus-
specific immunity following haemopoietic stem cell transplantation. Blood Rev 17:
259-64.

Garin MI, et al. (2001). Molecular mechanism for ganciclovir resistance in human T
lymphocytes transduced with retroviral vectors carrying the herpes simplex virus
thymidine kinase gene. Blood 97: 122-9.

Gaspar HB, et al. (2004). Gene therapy of X-linked severe combined
immunodeficiency by use of a pseudotyped gammaretroviral vector. Lancet 364:
2181-7.

Gillessen SD et al. (1995). Mouse interleukin-12 (IL-12) p40 homodimer: a potent IL-
12 antagonist. Eur J Immunol 25: 200-6.

Gruber TA, Skelten DC, Kohn DB (2002). Requirement for NK cells in CD40 ligand-
mediated rejection of Philadelphia chromosome-positive acute lymphoblastic
leukemia cells. J Immunol 168: 73-80.

Gruber TA, Skelten DC, Kohn DB (2005). Recombinant murine interleukin-12 elicits
potent antileukemic immune responses in a murine model of Philadelphia
chromosome-positive acute lymphoblastic leukemia. Cancer Gene Ther 12: 818-24.

Guillaume T, Rubinstein DB, Symann M (1998). Immune reconstitution and
immunotherapy after autologous hematopoietic stem cell transplantation. Blood 92:
1471-90.

Hacein-Bey-Abina S, et al. (2002). Sustained correction of X-linked severe
combined immunodeficiency by ex vivo gene therapy. N Engl J Med 346: 1185-93.

Hacein-Bey-Abina S, et al. (2003). LMO2-associated clonal T cell proliferation in two
patients after gene therapy for SCID-X1. Science 302: 415-19.

Haining NW, et al. (2005). Failure to define window of time for autologous tumor
vaccination in patients with newly diagnosed or relapsed acute lymphoblastic
leukemia. Exp Hematol 33: 286-94.
128
Halene S, Wang L, Cooper RM, Bockstoce DC, Robbins PB, Kohn DB (1999).
Improved expression in hematopoietic and lymphoid cells in mice after
transplantation of bone marrow transduced with a modified retroviral vector. Blood
94: 3349-57.

Hata A, Asanuma H, Rinki M, Sharp M, Wong RM, Blume K, Arvin AM (2002). Use
of an inactivated varicella vaccine in recipients of hematopoietic-cell transplants. N
Engl J Med 347: 26-34.

Henning KJ, White MH, Sepkowitz KA, Armstrong D (1997). A national survey of
immunization practices following allogeneic bone marrow transplantation.
JAMA 277: 1148-51.

Hollis RP, et al (2006). Stable gene transfer to human CD34(+) hematopoietic cells
using the Sleeping Beauty transposon. Exp Hematol 34: 1333-43.

Jahn T, Seipel P, Urschel S, Peschel C, Duyster J (2002). Role for the Adaptor
Protein Grb10 in the Activation of Akt. Mol Cell Biol 22: 979-91.

Jiang C, Magee M, Cox RA (1999). Construction of a single-chain interleukin-12-
expressing retroviral vector and its application in cytokine gene therapy against
experimental coccidiomycosis. Infection Immunity 67: 2996-3001.

Jones LK, Saha V (2005). Philadelphia chromosome acute lymphoblastic leukemia
of childhood. Brit J Haematol 130: 489-500.

Kang WK, et al. (2001). Interleukin 12 gene therapy of cancer by peritumoral
injection of transduced autologous fibroblasts: outcome of a phase I study. Hum
Gene Ther 12: 671-84.

Kitayama H, et al. (1995). Constitutively activating mutations of c-kit receptor
tyrosine kinase confer factor-independent growth and tumorigenicity of factor-
dependent hematopoietic cell lines. Blood 85: 790-8.

Kohn DB, et al. (1995). Engraftment of gene-modified umbilical cord blood cells in
neonates with adenosine deaminase deficiency. Nat Med 1: 1017-23.

Kohn DB, et al. (1998). T lymphocytes with a normal ADA gene accumulate after
transplantation of transduced autologous umbilical cord blood CD34
+
cells in ADA-
deficient SCID neonates. Nat Med 4: 775-80.

Kohn DB, Sadelain M, Glorioso JC (2003). Occurance of leukaemia following gene
therapy of X-linked SCID. Nat Rev Cancer 3: 477-88.

Kwak LW (2000). Approaches for immunotherapies of lymphomas. Immunol Invest
29: 93-5.

129
Lechanteur C, et al. (2000). Antitumoral vaccination with granulocyte-macrophage
colony-stimulating factor or interleukin-12-expressing DHD/K12 colon
adenocarcinoma cells. Cancer Gene Ther 7: 676-82.

Lee P, et al. (2001). Effects of interleukin-12 on the immune response to a
multipeptide vaccine for resected metastatic melanoma. J Clin Oncol 19: 3836–47.

Leonard LP, et al. (1997). Effects of single-dose interleukin-12 exposure on
interleukin-12-associated toxicity and interferon-gamma production. Blood 90: 2541-
8.

Levine, BL et al. (2006). Gene transfer in humans using a conditionally replicating
lentiviral vector. Proc Natl Acad Sci USA 103: 17372–7.

Li Q, Eppolito C, Odunsi K, Shrikant PA (2006). IL-12-programmed long-term CD8
+

T cells responses requires STAT4. J Immunol 177: 7618-25.

Li Z, et al. (2002). Murine leukemia induced by retroviral gene marking. Science
296: 497.

Lieschke GJ, Rao PK, Gately MK, Mulligan RC (1997). Bioactive murine and human
interleukin-fusion proteins which retain antitumor activity in vivo. Nat Biotech 15: 35-
40.

Limaye AP, Corey L, Koelle DM, Davis CL, Boeckh M (2000). Emergence of
ganciclovir-resistant cytomegalovirus disease among recipients of solid-organ
transplants. Lancet 356: 645-9.

Lode HN, et al. (1998). Gene therapy with a single chain interleukin 12 fusion
protein induces T cell-dependent protective immunity in a syngeneic model of
murine neuroblastoma. Proc Natl Acad Sci USA: 95: 2475-80.

Mescher MF, et al. (2006). Signals required for programming effector and memory
development by CD8
+
T cells. Immunologic Rev 211: 81-92.

Montini E, et al. (2006). Hematopoietic stem cell gene transfer in a tumor-prone
mouse model uncovers low genotoxicity of lentiviral vector integration. Nat
Biotechnol 24: 687-96.

Moolten, FL (1994). Drug sensitivity ("suicide") genes for selective cancer
chemotherapy. Cancer Gene Ther 1: 279-87.

Morfin F, Thouvenot D (2002). Herpes simplex virus resistance to antiviral drugs. J
Clin Virol 26: 29-37.

Mundhada S, Shaw J, Mori S, Savary CA, Mullen CA (2005). Cellular tumor
vaccines administered after T cell-depleted allogeneic bone marrow transplantation
induce effective anti-tumor immune responses. Leuk Lymphoma 46: 571-80.
130
Myers JN, et al. (1998). Interleukin-12 gene therapy prevents establishment of SCC
VII squamous cell carcinomas, inhibits tumor growth, and elicits long-term antitumor
immunity in syngeneic C3H mice. Laryngoscope 108: 261-8.

Nanni P et al. (2001). Combined allogeneic tumor cell vaccination and systemic
interleukin 12 prevents mammary carcinogenesis in HER-2/neu transgenic mice. J
Exp Med 194: 1195-1205.

Nemunaitis J, et al. (2006). Phase 1/2 trial of autologous tumor mixed with an
allogeneic GVAX vaccine in advanced-stage non-small-cell lung cancer. Cancer
Gene Ther 13: 555-62.

Noguchi Y, Richards EC, Chen YT, Old LJ (1995). Influence of interleukin 12 on p53
peptide vaccination against established Meth A sarcoma. Proc Natl Acad Sci USA
92: 2219-23.

Ott MG, et al. (2006). Correction of X-linked chronic granulomatous disease by gene
therapy, augmented by insertional activation of MDS1-EVI1, PRDM16 or SETBP1.
Nat Med 12:401-9.
Parkkali T, Stenvik M, Ruutu T, Hovi T, Volin L, Ruutu P (1997). Randomized
comparison of early and late vaccination with inactivated poliovirus vaccine after
allogeneic BMT. Bone Marrow Transplant 20: 663-8.

Parmiani G (2003). Immunotherapy of melanoma. Semin Cancer Biol 13:391-400.

Popovic D, El-Shami KM, Vadai E, Feldman M, Tzehoval E, Eisenbach L (1998).
Antimetastatic vaccination against Lewis lung carcinoma with autologous tumor cells
modified to express murine interleukin 12. Clin Exp Metastasis 16: 623-32.

Carr-Brendel V, Markovic D, Smith M, Taylor-Papadimitriou J, Cohen EP (1999).
Immunity to breast cancer in mice immunized with X-irradiated breast cancer cells
modified to secrete IL-12. J Immunother 22: 415-22.

Porteus MH, Connelly JP, Pruett SM (2006). A look to future directions in gene
therapy research for monogenic diseases. PLoS Genet 2:e133.

Portielje JE, Gratama JW, van Ojik HH, Stoter G, Kruit WH (2003). L-12: a
promising adjuvant for cancer vaccination. Cancer Immunol Immunother 52: 133-44.

Qasim, W, Thrasher AJ, Buddle J, Kinnon C, Black ME, Gaspar HB (2002). T cell
transduction and suicide with an enhanced mutant thymidine kinase. Gene Ther. 9:
824-7.

Ram Z, et al. (1997). Therapy of malignant brain tumors by intratumoral implantation
of retroviral vector-producing cells. Nat Med 3: 1354-61.

131
Rezvani K, et al. (2003). Functional leukemia-associated antigen-specific memory
CD8+ T cells exist in healthy individuals and in patients with chronic myelogenous
leukemia before and after stem cell transplantation. Blood 102: 2892-900.

Robbins PB, et al. (1997). Increased probability of expression from modified
retroviral vectors in embryonal stem cells and embryonal carcinoma cells. J Virol 71:
9466-74.

Robbins PB, Skelton DC, Yu X, Halene S, Leonard EH, Kohn DB (1998).
Consistent, persistent expression from modified retroviral vectors in murine
hematopoietic stem cells. Proc Natl Acad Sci USA 95: 10182-7.

Roche Laboratories (2000) - Cytovene IV/Cytovene Complete Product Information,
p.14. Available at: http://www.rocheusa.com/products/cytovene/pi.pdf.

Ryan OS, Gantt KR, Finn OJ (2007). Tumor antigen-based immunotherapy and
immunoprevention of cancer. Int Arch Allergy Immunol 142: 179-89.

Satoh T, Irie A, Egawa S, Baba S (2005). In situ gene therapy for prostate cancer.
Curr Gene Ther 5: 111-9.

Schröder ARW, Shinn P, Chen H, Berry C, Ecker JR, Bushman F (2002). HIV-1
integration in the human genome favors active genes and local hotspots. Cell 110:
521-9.

Sell S (2006). Potential Gene Therapy Strategies for Cancer Stem Cells. Curr Gene
Ther 6: 579-91.

Shaked R. et al. (2005). The splenic mircroenvironment is a source of
proangiogenesis/inflammatory mediators accelerating the expansion of murine
erythroleukemia cells. Blood 105: 4500-7.

Shou Y, Ma Z, Lu T, Sorrentino BP (2006). Unique risk factor for insertional
oncogenesis in a mouse model of XSCID gene therapy. Proc Natl Acad Sci USA
103: 11730-5.

Shtil AA, Turner JG, Durfee J, Dalton WS, Yu H (1999). Cytokine-based tumor cell
vaccine is equally effective against parental and isogenic multidrug-resistant
myeloma cells: the role of cytotoxic T lymphocytes. Blood 93: 1831-7.

Simons JW, Sacks N (2006). Granulocyte-macrophage colony-stimulating factor-
transduced allogeneic cancer cellular immunotherapy: the GVAX vaccine for
prostate cancer.

Sloan AE, et al. (2000). Adaptive immunotherapy in patients with recurrent
malignant glioma: preliminary results of using autologous whole-tumor vaccine plus
granulocyte-macrophage colony-stimulating factor and adoptive transfer of anti-
CD3-activated lymphocytes. Neurosurg Focus 9: e9.
132
Soneoka Y, et al. (1995). A transient three plasmid expression system for the
production of high titre retroviral vectors. Nucleic Acids Res 23: 626–33.

Stripecke R, et al. (1998). Immune response to Philadelphia chromosome-positive
acute lymphoblastic leukemia induced by the expression of CD80, interleukin-2, and
granulocyte-macrophage colony-stimulating factor. Hum Gene Ther 9: 2049-62.

Stripecke R, Skelton DC, Pattengale PK, Shimada H, Kohn DB (1999). Combination
of CD80 and granulocyte-macrophage colony-stimulating factor coexpression by a
leukemia cell vaccine: preclinical studies in a murine model recapitulating
Philadelphia chromosome-positive acute lymphoblastic leukemia. Hum Gene Ther
10: 2109-22.

Su Z, et al. (2003). Immunological and clinical responses in metastatic renal cancer
patients vaccinated with tumor RNA-infected dendritic cells. Cancer Res 63: 2127-
33.

Sun W, et al. (2004). IL-12p40-overexpressing immature dendritic cells induce T cell
hyporesponsiveness in vitro but accelerate allograft rejection in vivo: role of NK cell
activation and interferon-gamma production. Immunol Letter 94: 191-9.

Teshima T, Mach N, Hill GR, Pan L, Gillessen S, Dranoff G, Ferrara JL (2001).
Tumor cell vaccine elicits potent antitumor immunity after allogeneic T-cell-depleted
bone marrow transplantation. Cancer Res 61: 162-71.

Tiberghien P, et al. (2001). Administration of herpes simplex-thymidine kinase-
expressing donor T cells with a T-cell-depleted allogeneic marrow graft. Blood 97:
63-72.

Uyl-de-Groot CA, et al. (2005). Immunotherapy with autologous tumor cell-BCG
vaccine in patients with colon cancer: a prospective study of medical and economic
benefits. Vaccine 23: 2379-87.

Wildner O (1999). In situ use of suicide genes for therapy of brain tumours. Ann
Med 31: 421-29.

Williams RT, Roussel MF, Sherr CJ (2006). Arf gene loss enhances oncogenicity
and limits imatinib response in mouse models of Bcr-Abl-induced acute
lymphoblastic leukemia. Proc Natl Acad Sci USA 103: 6688-93.

Wong S, Witte ON (2004). The BCR-ABL story: bench to bedside and back. Annu
Rev Immunol 22: 247-306.

Woods NB, Bottero V, Schmidt M, von Kalle C, Verma IM (2006). Gene therapy:
therapeutic gene causing lymphoma. Nature 440: 1123.

Wu S, Yuan D, Liu JN, Tan XY (2007). Induction of anti-tumor immunity by
lyophilized myeloma cells secreting GM-CSF. Oncol Rep 17:129-33.
133
Wu X, Li Y, Crise B, Burgess SM (2003). Transcription start regions in the human
genome are favored targets for MLV integration Science 300: 1749-51.

Zhao P, Liu W, Cui Y (2006). Rapid immune reconstitution and dendritic cell
engraftment post-bone marrow transplantation with heterogeneous progenitors and
GM-CSF treatment. Exp Hematol 34: 951-64.

134

Abstract (if available)

Abstract Gene therapy is an approach to treat human diseases by stable transfer of exogenous genes. Stem cell-based gene therapy for congenital blood disorders is accompanied by the risk for leukemia development conferred by insertional oncogenesis. The safety of retroviral gene transfer may be increased by including a suicide gene in the therapeutic vector to eliminate adverse events. Mice were challenged with leukemia cells transformed by proviral integration of a murine oncogene that simultaneously expressed the HSV-TK suicide gene or hyperactive mutants thereof (SR39 and sc39). After treatment with the drug substrate Ganciclovir, leukemia developed in mice given clonal cells expressing HSV-TK, but not SR39 or sc39. In vitro Ganciclovir resistance was observed in heterogeneously transduced leukemia pools, and single nucleotide changes or partial loss of the suicide gene were identified as mechanisms of drug escape. However, Ganciclovir treatment resulted in 80-100% survival of mice challenged with partially resistant leukemia pools expressing modified HSV-TK variants with improved biological activity.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Non-viral and viral hematopoietic progenitor cell gene therapy

PDF

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

PDF

The role of survivin in drug resistant pediatric acute lymphoblastic leukemia

PDF

Cytotoxic effect of NEO212, a novel perillyl alcohol-temozolomide conjugate, on canine lymphoma

PDF

Genomic stability of transcriptionally targeted replication competent retroviral vectors

PDF

Cell and gene therapy in the murine model of adenosine deaminase deficiency

PDF

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

PDF

Development of a temozolomide-perillyl alcohol conjugate, NEO212, for the treatment of hematologic malignancies

PDF

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

PDF

Induction of hypersignaling as a therapeutic approach for treatment of BCR-ABL1 positive Acute Lymphoblastic Leukemia (ALL) cells

PDF

Identification of molecular mechanism for cell-fate decision in liver; &, SARS-CoV replicon inhibitor high throughput drug screening

PDF

Silencing of expression of the DRIP-80 gene correlates with aberrations in calcium ion distribution in nickel compound- and 3-methylcholanthrene-transformed C3H/10T1/2 mouse embryo fibroblast cel...

PDF

Role of integrin α4 in drug resistant acute lymphoblastic leukemia

PDF

Integrin mediated cellular adhesion may alter the cytokine profile in acute lymphoblastic leukemia

PDF

Differential role of two coactivators, CCAR1 and CARM1, for dysregulated beta-catenin activity in colorectal cancer cell growth and gene expression

PDF

Molecular targets for treatment of glioblastoma multiforme

PDF

Reactivation of the Epstein Barr virus lytic cycle in nasopharyngeal carcinoma by NEO212, a novel temozolomide-perillyl alcohol conjugate

PDF

Glioblastoma treatment with 2,5-dimethyl-celecoxib (DMC) in vitro: - effects of additional chemotherapeutic drugs and tumor microenvironment, - inconsistencies among commonly used in vitro cell g...

PDF

Enhanced Burkitt 's lymphoma cell killing by the combination treatments of bortezomib with celecoxib and 2,5-dimethyl-celecoxib (DMC)

PDF

Oncolytic adenovirus based vaccine in cancer immunotherapy

Asset Metadata

Creator Blumenthal, Martina (author)

Core Title Gene therapy for the prevention and treatment of leukemia

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Molecular Microbiology

Publication Date 07/03/2007

Defense Date 04/25/2007

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

Tag gene therapy,immunotherapy,leukemia,OAI-PMH Harvest,suicide genes

Language English

Advisor Kohn, Donald B. (committee chair), Cannon, Paula M. (committee member), Ou, James (committee member), Schonthal, Axel H. (committee member)

Creator Email mblument@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m580

Unique identifier UC1287937

Identifier etd-Blumenthal-20070703 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-508302 (legacy record id),usctheses-m580 (legacy record id)

Legacy Identifier etd-Blumenthal-20070703.pdf

Dmrecord 508302

Document Type Dissertation

Rights Blumenthal, Martina

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