University of Southern California Dissertations and Theses

Stem cell and gene transfer-based approaches to generate insulin-producing cells

(USC Thesis Other)

Stem cell and gene transfer-based approaches to generate insulin-producing cells

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

STEM CELL AND GENE TRANSFER-BASED APPROACHES TO GENERATE
INSULIN-PRODUCING CELLS

by

Eszter Pais

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
(SYSTEMS BIOLOGY AND DISEASE)

August 2009

Copyright 2009 Eszter Pais

ii
DEDICATION

I would like to dedicate this work to my parents: Kadosa Pais and Dr. Zsuzsanna
Horvath; to my sister, Dr. Eniko Pais and to my husband, Dr. Tamas Alexy for being
always there for me and for their continuous support throughout my PhD years.

iii
ACKNOWLEDGMENTS

Foremost I would like to acknowledge my mentor, Dr. Donald B. Kohn for his
continuous support and encouragement throughout my years as a PhD student. My three
years in his laboratory was an extremely exciting period during which I have grown both
scientifically and personally. Thank you for accepting me as a graduate student and for
having faith in me throughout these years.
I would like to thank to my PhD Committee members, Carolyn Lutzko, PhD; Martin
Pera, PhD; Robert Chow, MD, PhD and Herbert J. Meiselman, ScD for guiding me
through my graduate studies and for being exceptionally helpful and supportive. Also, I
would like to thank to Alicia McDonough, PhD for running this exceptional graduate
program where the interest of the students always comes first.
Special thanks to my husband, Dr. Tamas Alexy for his contribution to my PhD work
and also for sharing his life with me. It has been an intense 3 years for us with its ups and
downs; thanks for sharing all those moments with me.
I would also like to thank for all members of the Kohn-lab and the Division of Research
Immunology / Bone Marrow Transplantation at Children’s Hospital Los Angeles
(CHLA) for being such a friendly and supportive group. Special thanks to the members
of the Vector Core: Roger Hollis, Dianne Skelton, Shantha Senadheera

and Cinnamon L.
Hardee for providing me the lentiviral vectors. I am also very thankful for Vahagn
Nikolian and Jean Park for their technical assistance.

iv
I had numerous teachers throughout these years whom I would like to acknowledge:
Denise Carbonaro Sarracino, Kit Shaw, Shundi Ge, Dinithi Senadheera, Teiko
Sumiyoshi, Chris Choi, Ewa Zielinska, Lora Barsky and Xingchao Wang. From each of
you I have learned a technique or a shortcut that was extremely helpful for me to move
my projects forward.
It was my pleasure to be part of the “embryonic stem cell team” and to have endless
scientific discussions with Ken Sakurai about ES cells.
Last but not least I am extremely thankful for Dr. Herbert Meiselman, Dr. Karin
Meiselman and Ms. Rosalinda Wenby for helping us to settle in the United States and for
encouraging me to enter the PhD program.

v
TABLE OF CONTENTS

Dedication ii

Acknowledgments iii

List of Tables vi

List of Figures vii

Abstract viii

Chapter I: Introduction 1

Chapter II: Gene transfer-based approach to achieve the controlled 16
expansion of pancreatic beta cells

Chapter III: Lentiviral vectors as a tool to identify insulin-positive 45
cells derived from human embryonic stem cells

Chapter IV: Conclusions 66

vi
LIST OF TABLES

Table 2.1: Lentiviral vectors encoding the F36Vcmet fusion protein. 27

Table 2.2: Effect of incremental AP20187 concentrations on the percentage 35
of insulin and BrdU double positive cells.

Table 2.3: Proportion and absolute number of insulin positive cells 37
in human pancreatic islet cultures, 7 days post-transduction.

Table 3.1: Primers utilized for RT-PCR analysis. 54

vii
LIST OF FIGURES

Figure 2.1: Schematic drawing of the F36Vcmet fusion protein. 19

Figure 2.2: Characteristics of the MND-F36Vcmet-PGK-eGFP vector. 28

Figure 2.3: Transcripts from the MND-F36Vcmet-PGK-eGFP construct. 29

Figure 2.4: Protein expression in cell lines. 30

Figure 2.5: Proliferation of Nit-1 cells. 32

Figure 2.6: F36Vcmet protein expression in human pancreatic islets. 33

Figure 2.7: Proliferation of human pancreatic beta cells. 36

Figure 2.8: The absolute number of insulin positive cells over time. 38

Figure 2.9: Functional analysis of human pancreatic beta cells. 39

Figure 3.1: Schematic representation of the protocol used to differentiate 52
hES cells into the pancreatic beta cell lineage.

Figure 3.2: eGFP expression in human pancreatic beta cells. 56

Figure 3.3: Successful differentiation of H1 cells into the pancreatic 57
beta cell lineage as indicated by RT-PCR analysis.

Figure 3.4: The successful differentiation of H1 cells into the pancreatic 58
beta cell lineage is confirmed by immunohistochemistry.

Figure 3.5: Gene expression profile of transduced and 59
nontransduced H1 cells.

Figure 3.6: eGFP-expression at stage 3 of the differentiation protocol 60
as detected by immunohistochemistry.

Figure 3.7: eGFP-expression at stage 4 of the differentiation protocol 61
as detected by immunohistochemistry.

viii
ABSTRACT

The incidence of Diabetes Mellitus (DM) continues to increase with approximately
7.8% of the population affected in the US alone. Although pancreatic beta cell
transplantation has the potential to cure the disease, limited donor tissue availability
poses a major challenge. Aiming to overcome this critical shortage, we utilized two
distinct, lentiviral vector-based approaches to generate or expand insulin-positive cells ex
vivo.
The first set of experiments was aimed at inducing the controlled proliferation of human
pancreatic beta cells through the specific activation of the Hepatocyte Growth Factor
(HGF) signaling pathway in beta cells. This goal was achieved by the lentiviral delivery
of a novel fusion protein (F36Vcmet) into the target cell population. Notably, this
transmembrane receptor protein is active only in the presence of a synthetic ligand
(AP20187). Our data provide evidence that the selective expansion of human pancreatic
beta cells has been achieved utilizing this approach.
The second project described herein provides a novel tool to indicate the successful
differentiation of human embryonic stem (hES) cells into the pancreatic beta cell lineage.
For these purposes, we utilized an insulin promoter-driven lentiviral vector construct that
specifically identifies those cultured cells that became insulin-positive. Here we show
evidence that the vector designed in our laboratory is indeed specific and indicates the
appearance of insulin-positive cells in the cultures reliably. As a consequence, this vector

ix
construct may be used for screening purposes to identify the optimal culture conditions
that support the differentiation of hES cells into the pancreatic beta cell lineage.
Both approaches represent novel strategies aiming to increase pancreatic beta cells
supply available for transplantation procedures. Although future studies and
optimizations are required, if successful, these approaches have the potential to expand
therapeutic options available for individuals with DM thereby improving the quality of
life of millions of patients worldwide.

1
CHAPTER I
Introduction

1 Background

1.1 Type I diabetes mellitus
Diabetes mellitus (DM) is a chronic disorder of carbohydrate, protein and fat
metabolism caused by a combination of hereditary and environmental factors. The
prevalence of the disease continues to increase worldwide; it is estimated that more than
23 million children and adults (7.8% of the total population) are affected currently in the
United States alone [54,96]. The World Health Organization (WHO) projects that the
number of affected individuals will exceed 350 million by 2030 worldwide.
The most characteristic feature of DM is hyperglycemia caused either by deficient
insulin secretory response or defective insulin action on the periphery. Most commonly, it
develops as a result of primary beta cell dysfunction although rarely, it may be a
secondary disease as a consequence of other systemic pathology, such as chronic
pancreatitis, hemochromatosis, malignancies, toxins and infections. Primary DM might
be further categorized based on the pathophysiology and inheritance: 1) Type 1 that is
commonly referred to as insulin-dependent DM (T1D) and 2) Type 2, once known as
adult-onset or noninsulin-dependent DM (T2D). Although T2D is more frequent,
accounting for approximately 90% of all cases, the primary focus of the present work will
be directed at T1D. This subtype of DM is characterized by the near complete loss of

2
insulin-producing beta cells of the islets of Langerhans in the pancreas leading to
deficient insulin production. Affected patients depend on exogenous hormone
replacement for survival; without continuous insulin therapy they are prone to severe
ketoacidosis that may result in coma or death. Although some cases are considered
idiopathic, the majority of T1D is immune-mediated where a selective T-cell autoimmune
attack leads to the destruction of pancreatic beta cell. As a consequence, insulin secretion
in response to physiologic stimuli fades over time and hyperglycemia develops [89].
Genetic susceptibility, based on certain HLA genotypes (particularly HLA DR3 and
DR4), as well as environmental factors have been implicated in triggering this
uncontrolled autoimmune response. Although the precise mode of inheritance remains
obscure, family, monozygotic and dizygotic twin studies have shown that T1D runs in
certain families. The concordance rate among monozygotic twins is reported to be
between 21% and 70%, which is five times higher than that described for dizygotic twins
[75]. While these observations underline the importance of genetic predisposition, they
also indicate that other factors, such as environmental factors are important and are likely
necessary to trigger the disease process in susceptible individuals. Several trials are
currently ongoing and focus on identifying possibly modifiable environmental triggers,
such as viral infections and dietary ingredients.
Owing to the nature of the disease process, therapeutic options are limited in patients
with T1D. Insulin replacement, either delivered as multiple daily subcutaneous injections
or administered continuously via portable pumps needs to be continued indefinitely.
Although these approaches allow long-term survival and successfully delay the

3
development of severe diabetic complications in most cases, yet they are far from ideal.
Treatment is inconvenient and often burdensome for patients prompting reduced
compliance. In addition, insulin is replaced in a non-physiologic manner and thus blood
glucose levels are extremely sensitive to the delivered hormone dose, dietary food intake,
physical activity, or any other forms of stress. Therefore, maintaining normal blood
glucose levels may be challenging and subjects with T1D are vulnerable to hypoglycemia
as well as to hyperglycemia.
Aiming to overcome some of the disadvantages of conventional insulin replacement, an
inhaled form of the hormone was approved by the Food and Drug Administration (FDA)
in 2006, although it was discontinued a year later due to failed marketing and safety
concerns. Considering the pathophysiology of the disease; a logical, and probably the
most physiological approach would be to replace damaged insulin producing cells with
functional pancreatic beta cells. Researchers focus on islet transplantation and stem cell
based therapies and these are also the approaches discussed herein.

1.2 Pancreatic islet transplantation
Replacing damaged and non-functional pancreatic beta cells via transplantation would
potentially restore normal glucose homeostasis and represents a promising option for
patients with T1D. During transplantation, islets are initially obtained from multiple
donors then undergo a complicated and technically demanding purification procedure and
are finally infused into the portal vein of recipients. In a successful clinical study
published by Shapiro in 2000, none of the seven patients required exogenous insulin

4
replacement one year following the procedure [82]. Their clinical approach-known in the
scientific community as the Edmonton protocol- stressed that it is necessary to transplant

a critical islet cell mass in order to achieve insulin independence. In addition, this was the
first protocol to utilize an immunosuppressive regimen without corticosteroids. Over 500
patients with T1D at 50 institutions worldwide have undergone successful islet
transplantation to date based on this protocol [36,65,77,78].
Despite its clinical success, routine islet transplantation faces two major challenges:
1) Insulin independence is lost over time in most transplant recipients, as explained by
immunological rejection [45]; 2) Donor tissue availability is limited. Lechner and
colleagues estimated that merely 1% of T1D patients could potentially receive islet
transplantation each year owing to the critical shortage of donor islets [29]. As a
consequence, finding alternative sources of insulin-producing beta cells has become a
critical issue and, if successful, it would allow transplantation to become a widely
available therapeutic option.

1.3 Potential new sources of human pancreatic beta cells
Several approaches have been proposed in recent years to overcome the critical beta cell
shortage available for transplantation. These include: 1) In vivo regeneration of
autologous pancreatic beta cells thus completely eliminating the need for transplantation;
2) Ex vivo beta cell expansion prior to transplantation that would reduce islet
requirements significantly; 3) Utilizing adult stem/progenitor cell populations to generate

5
insulin positive cells; 4) using xenografts; 5) embryonic stem cell-based approaches and
6) transdifferentiating fully differentiated cells into the beta cell lineage.

1.3.1 In vivo regeneration of pancreatic beta cells
Early studies suggested that the number of pancreatic beta cells is stable after birth and
no further increase is possible in the postnatal period. These cells were considered
terminally differentiated and quiescent without the ability to proliferate. However,

recent
studies have challenged this hypothesis and suggests that beta cell mass can increase even
at adulthood and has the ability to respond to changes in the metabolic balance even later
in life. For example, pregnancy and obesity with insulin resistance have been shown to
promote a significant increase in the islet cell mass in adults [38,88]. It is still debated
whether beta cell neogenesis or mature beta cell replication is responsible for this
phenomenon [50,52]. A recent study by Dor and colleagues has shown that beta cell
proliferation and not neogenesis was the primary mode of islet regeneration in adult mice
following partial pancreatectomy [21]. The critical importance of cellular replication has
also been confirmed by other animal studies showing that any disturbance in the cell
cycle regulation has a profound effect on the total beta cell number [25,35,40,74]. As an
alternative, islet neogenesis has also been implicated in pancreas regeneration in various
rodent models [69]. Cells proposed to serve as the source of newly generated insulin
positive cells fall into two categories: 1) Cells residing within the pancreas and
2) Progenitors outside of the pancreas. Detailed description of these cells is presented in
chapters 1.3.3 and 1.3.4 of the dissertation.

6
Despite these observations in rodent models, only limited data are available for humans.
As noted above, some studies suggest that islet mass may increase in pregnancy and
obesity [38,79,88]. On the other hand and contradicting those results obtained in rodents,
no beta cell regeneration was observed following partial pancreatectomy in a study by
Menge, et al [58]. Other sources suggest that some beta cell regeneration might occur in
adults. For example, autopsy samples obtained from the pancreas of patients with
long-standing T1D revealed the presence of limited number of viable beta cells despite
the sustained apoptosis suggesting some degree of regenerative capacity [53,56]. Further
studies are required to better understand beta cell regeneration that might prompt the
development of novel strategies to treat T1D.

1.3.2 Ex vivo expansion of pancreatic beta cells
Ex vivo expansion of pancreatic beta cells represents another logical approach to
improve donor tissue availability. Not only would it allow the transplantation program to
be more widely accessible but would also avoid the need to pool islets from multiple
donors prior to the procedure. As a consequence, the rate of immunological rejection due
to incompatibility is likely to decrease.
In recent years, multiple molecules have been proposed to successfully induce beta cell
expansion. Many of these molecules are hormones, such as growth hormone and
prolactin, while others belong to the family of growth factors, including insulin-like
growth factor I and II (IGF-I and IGF-II) as well as Hepatocyte Growth Factor (HGF)
[85,92]. HGF has received special attention and has been studied most extensively

7
because, in addition to its proliferative effect on beta cells, it has been noted to improve
islet graft survival. Further details on HGF are found in Chapter 2.1 of this work. Other
molecules studied in detail include glucagon-like peptide 1 (GLP-1) and its receptor
agonists. Studies in animal models have shown that GLP-1 not only induces beta cell
proliferation, but also inhibits apoptosis, induces islet neogenesis from precursors [22,23]
and improves glucose homeostasis. GLP-1 agonists were approved by the FDA in 2005
as adjunctive therapy for selected patients with T2D. Long-term studies are under way
with the aim to follow beta cell mass in these subjects.

1.3.3 Precursor cells residing in the pancreas
Several cell types present in the adult pancreas have been proposed as potential sources
of new beta cells. Among these, ductal cells are the most widely accepted to have the
capability to transdifferentiate into insulin-positive cells both in vivo and in vitro. For
example, Cornelius, et al. showed that ductal tissue obtained from murine pancreas form
islet-like clusters when maintained in culture [17]. In addition, hyperglycemia was
ameliorated in non-obese, diabetic mice upon implantation of these structures [73]. As
first reported by Bonner-Weir and colleagues, insulin-producing islet-like clusters can
also be derived from human ductal tissue in vitro [9]. Since these initial studies, several
research groups have confirmed these findings [10,42]. However, the efficiency of this
approach is still debated and it is unclear how feasible it would be to use ductal cells to
generate significant amounts of beta cells for transplantation.

8
Pancreas acinar cells have also been proposed as a potential source of insulin-producing
cells. Studies have shown that these cells have the capability to transdifferentiate into
ductal epithelial cells both in vivo and under in vitro conditions [76]. Then, they can give
rise to insulin-producing cells upon stimulation with specific growth factors, although
with a very low efficiency [3,86]. The transdifferantiation of acinar cells directly into
insulin-positive cells has also been reported in a murine model by Okuno, et al [64]. The
newly generated cells indeed produced insulin in response to glucose stimulus although
again, their secretory capacity was extremely low compared to that of native pancreatic
beta cells.
Islet progenitors residing in the pancreas are another cell type suggested as a potential
source for beta cells [27,72,115]. While it is widely accepted that such progenitor
populations are present in the developing pancreas and are characterized by the
co-expression of Pdx-1 (human equivalent: Ipf-1) and neurogenin-3 (Ngn-3), their
existence in mature adult tissue is still debated [68]. Some evidence suggests that
progenitors are located in the ductal epithelium of the adult pancreas; when cultured in
vitro under special conditions, they had the potential to yield insulin-positive cells [80].
In addition, they are also believed to participate in the regenerative processes of the
pancreas [11,102].
In summary, various cell types of the adult pancreas have been proposed to serve as a
potential source of beta cells. However, to date, researchers were unable to efficiently
differentiate any of the above cells into functional beta cells.

9
1.3.4 Other adult stem/progenitor cell populations
Transdifferentiation of various progenitor cells residing outside of the pancreas offers a
possible alternative approach to generate insulin-producing cells [47]. For these purposes,
one of the most extensively studied cell populations are the hepatocytes. Utilizing liver
tissue to derive beta cells seems to be a logical approach since liver not only shares a
common embryonic origin with the pancreas, but hepatocytes also have glucose-sensing
mechanisms similar to beta cells. Despite these similarities, transdifferention of liver cells
into functional beta cells is challenging with only limited success to date [72]. In fact, the
most promising results stem from studies where hepatocytes were modified via gene
transfer using various vector constructs. For example, expression of Pdx-1 and Pax4
prompted the transdifferentiation of mouse hepatocytes into insulin-positive cells with the
ability to revert Streptozotocin-induced diabetes in mice [43]. Similar success has been
reported upon the overexpression of Ngn3 or NeuroD genes in liver cells [106]. Despite
these promising results, the exact origin of the newly derived insulin-positive cells
remains uncertain with some emerging evidence suggesting that hepatic oval cells, the
residing multipotent progenitor cells of the liver, are the targets of the above detailed
approaches [105].
Bone marrow-derived progenitors represent the other, widely-studied population
proposed to serve as the source for cell types such as skin, lung, liver and also to have the
potential to generate insulin-positive cells. In this regard, two cell populations have
received special attention: bone marrow-derived 1) mesenchymal and 2) hematopoietic
stem cells. Using animal models of pancreas injury, both populations have been shown to

10
migrate into the damaged area and to participate in the regenerative processes [39]. While
mesenchymal stem cells have been suggested to create the insulin-positive cells [87,101],
the role of hematopoetic stem cells in pancreas regeneration is thought to be minimal
with only indirect effects on the transdifferentiating population.
Despite the extensive work in this field, the real potential of bone marrow-derived cells
to generate functional insulin-positive cells remains obscure. Limitations of current
techniques do not allow full characterization of the newly-derived populations in these in
vivo animal models, often leading to data misinterpretation. For example, many of the
early transplantation studies did not consider cell fusion [48,95,108], a process whereby
donor-derived and autologous cells residing in the recipient’s pancreas may fuse. As a
consequence, donor-derived cells were often mischaracterized. In addition, it is unclear
from studies published thus far whether the newly derived insulin-positive cells are
indeed functional and to what extent they contribute to the normalization of
hyperglycemia in these animal models.

1.3.5 Xenografts
Obtaining pancreatic islets from species other than humans represents an alternative
approach to acquire the significant amount of beta cells required for the transplantation
therapy of diabetes. Most studies focused on utilizing pig islets, because porcine tissues
have long been utilized in human medicine. For example, it was a common practice to
replace damaged heart valves by porcine valves [61,114], prior to the appearance of
synthetic transplant materials. Also, porcine insulin was the standard of care to manage

11
patients with T1D before recombinant human insulin became widely available. As such,
utilizing pig islets for human islet transplantation procedures might seem to be logical.
Not only can high islet yield be guaranteed by this approach, but also, genetic
manipulation of the source tissue might allow the synthesis of human insulin by the pig
beta cells. Nevertheless, the current methodology faces two major challenges: 1)
Hyperacute immunological rejection may follow the xenotransplantation and 2) Porcine
endogenous retroviral (PERV) sequence reactivation may occur that represents a
significant health care risk for transplant recipients [93]. One would think that these
major drawbacks would restrain the use of xenotransplantation in humans, but some
extremely controversial human clinical trials are currently under way to explore their
safety and efficacy [31,100]. A company based in Australia (Living Cell Technologies
Ltd.) has reported a trial run in Moscow where encapsulated porcine islet cells
(DiabeCell) were transplanted into two patients with diabetes, leading to their reduced
insulin requirements. Almost a decade after the procedure, investigators reported
evidence of residual, viable, encapsulated pig beta cells in one of the transplant recipients
[24]. The approach by this company has strongly been criticized by the International
Xenotransplantation Association as being premature and potentially risky [31]. Similarly,
earlier clinical trials of porcine islet transplantation performed in Mexico have been the
subject of extensive criticism regarding the preclinical data used to justify the trials and
the reported efficacy of the studies [93]. More carefully designed and controlled clinical
trials are necessary to determine whether pig islets may indeed represent a functional
substitute for human islets in the management of T1D.

12
1.3.6 Human embryonic stem cells (hES) and induced pluripotent stem cells (iPS)
The use of human embryonic stem cells (hES) and induced pluripotent stem cells (iPS)
as a theoretically unlimited source of pancreatic beta cells represent the most novel line
of investigations. Both cell types are considered pluripotent with the ability to give rise to
all cell types of the human body, including beta cells. HES cells are derived from
blastocyst-stage embryos; thereby creating vivid ethical debates over their use [13,97].
iPS cells are generated by introducing a set of transcription factors into various mature
cell types, such as fibroblasts [57,63,94,109]. Under specific conditions, both hES and
iPS have the ability to differentiate into insulin-positive cells, albeit with an extremely
low efficiency. In addition, the newly generated cells do not represent fully mature,
functional beta cells but rather are reminiscent of immature endocrine cells. Thus there is
a great need to optimize currently available differentiation protocols and to better
understand the basic biology of these pluripotent cells prior considering them as a
potential new source of pancreatic beta cells. A more detailed description of hES cells is
found in the Background section of Chapter 3 of this dissertation.

1.3.7 Transdifferentiation of fully differentiated cells into beta cell lineage
Besides the use of pluripotent stem cells as a potentially unlimited cell source for
regenerative medicine, a recent study by Melton and colleagues suggests that terminally
differentiated adult cells may be converted directly into other mature cell types or
progenitors [112]. In the study, authors demonstrate that by introducing three
transcription factors (Ngn-3, Pdx-1 and Mafa) into pancreatic exocrine cells in vivo it is

13
possible to generate insulin-positive cells that are very similar to beta cells. This
reprogramming was achieved by the direct injection of an adenoviral vector encoding the
three transcription factors into the pancreata of the study animals. Interestingly,
newly-derived insulin-positive cells appeared as early as 3 days after the injection which
is significantly shorter than that described in iPS studies where reprogramming may take
up to one month. Authors suggest that the faster reprogramming is likely due to the fact
that exocrine cells and beta cells are closely-related cell types and thus the epigenetic
make-up of the cells is very similar. However, it is tempting to speculate that the route of
vector administration (direct pancreatic injection) may have contributed to the rapid
formation of insulin-positive cells by creating a local inflammatory environment.
Nevertheless, these findings indicate that it is possible to circumvent the need to convert
cells into the pluripotent stage initially and then differentiate them into the desired cell
line. Introducing key transcription factors characteristic for the later stages of
development (the progenitor stage) may allow direct cell reprogramming. As such, these
findings represent an important milestone in our understanding of the plasticity of
terminally differentiated cells and open new possibilities to find new sources for various
cell types.

1.4 Lineage-specific lentiviral vectors
Vectors are gene delivery vehicles that facilitate the transfer of genetic material into
various cells. Lentiviral vectors represent one of the most potent and versatile systems
available to date to achieve high efficiency gene delivery into target cells. These vectors

14
are based on the human immunodeficiency virus (HIV) that belongs to the retroviridae
family of viruses. HIV harbors a complex genome that distinguishes this virus from other
types of retroviruses. Besides its core genes: gag, pol and env that are common to all
retroviruses, the genome of HIV also encodes a series of accessory proteins, such as: vif,
vpu, vpr, nef, tat and rev. These structural proteins harbor an important role in the normal
life cycle of the virus and in its high virulence. Since HIV is a threatened human
pathogen, its original genome had to undergo a series of modifications in order to become
a safe gene delivery vehicle. All alterations served one single purpose: to avoid the
formation of replication-competent lentiviruses (RCL) in the host. These modifications
included the elimination of parts of the original HIV genome encoding some of the
accessory proteins and separating the remaining viral protein-encoding sequences from
the inserted transgene of interest. Subsequent modifications lead to the development of
“third generation” lentiviral vectors in which the gag and pol sequences are on a
completely separate plasmid from that expressing env. In addition, the lack of
overlapping homologous sequences makes the production of RCL impossible. In
addition, in latest vector generations, the long terminal repeat (LTR) region of the vector
has been modified such that it became transcriptionally inactive. In these, so called “self
inactivating (SIN)” lentiviral vectors; internal promoters are driving transgene expression
[59,113]. The use of internal promoters not only allowed choosing the promoter with the
most optimal transgene expression in a given cell type, but also permitted the
development of vectors with regulated transgene expression. In these latter constructs,
transgene expression is driven by a promoter of a specific gene that is normally expressed

15
in that cell. For example, protein expression from an insulin promoter-driven vector
construct will only occur in cells that naturally produce insulin thereby allowing gene
targeting specifically to pancreatic beta cells.

2 Overall goal

As discussed above, it is of critical importance to find new sources of pancreatic beta
cells for transplantation in order to provide better treatment for T1D than the current
treatment options. Thus my research focused on achieving this goal by two distinct
approaches: 1) the controlled expansion of mature human pancreatic beta cells ex vivo;
using a gene-transfer-based approach (detailed in Chapter 2) and 2) the characterization
of a lentiviral vector as a tool to identify insulin-positive cells derived from hES. The
upcoming parts of this dissertation will further detail these approaches.

16
CHAPTER II
Gene transfer-based approach to achieve the controlled expansion of
pancreatic beta cells

1 Background

One possible approach proposed to successfully generate insulin-producing cells is the
expansion of pancreatic beta cells ex vivo. Hepatocyte Growth Factor (HGF) has been
studied extensively in this regard and is known to be a potent mitogen eliciting its effects
through the HGF/cmet signaling pathway. Importantly, the effects of HGF are highly
non-specific because its membrane receptors are present on the surface of multiple cell
types. This feature of HGF prevents it from being utilized as a selective beta cell
mitogen. However, with the advancement of molecular biology and gene therapy,
modifications to the native receptors became feasible thus raising hopes for selective and
highly specific interventions using synthetic ligands. Herein presented is a novel,
lentiviral vector-based approach aimed at achieving controlled HGF/cmet signaling
pathway activation that is exclusive to pancreatic beta cells.

17
1.1 HGF/cmet signaling pathway
HGF has originally been described by two research groups, almost simultaneously. One
group identified HGF as a motility factor (scatter factor) for epithelial cells [90] while the
other characterized it as a mitogenic molecule for hepatocytes [111]. Later it was shown
that HGF serves as a ligand for a Tyrosine kinase receptor already described, called cmet
[12,16]. This membrane-spanning receptor consists of one α- and one β-chain linked
together by a disulfide bridge. While the α-chain is highly glycosylated and is entirely
extracellular, the transmembrane β subunit consists of three segments: 1) ectodomain,
2) transmembrane helix and 3) cytoplasmic portion. The binding of HGF to the
extracellular portion of the receptor prompts its homodimerization followed by
transphosphorylation of multiple Tyrosine residues in the cytoplasmic domain (Y1349,
Y1356, Y1234 &Y1235). As a consequence, various signal transduction molecules are
activated and are responsible to initiate the variety of biological effects of HGF [30,71].
Mutagenic studies have revealed that only the cytoplasmic section of cmet is essential
for its signaling activity. This observation allowed significant modifications to be made
to the receptor. As an example; the cytoplasmic subunit was successfully fused to a
synthetically-designed drug-binding domain (F36V; detailed in section 1.2.) thus
allowing ligand-dependent, but HGF-independent, activation of the signaling pathway.
This particular fusion protein is presented and utilized in this thesis work.
HGF/cmet signaling is involved in promoting cell growth, motility

and morphogenesis
in a wide variety of cell types; in particular in cells of epithelial origin. The pioneer work
by Dr. Hayek's group in San Diego was the first to demonstrate that HGF is indeed a

18
potent mitogen for fetal human pancreatic islets ex vivo [66]. They have shown that HGF
preferentially increases the number of replicating beta cells and thus the proportion of
insulin-positive cells in fetal islet preparations. Later, HGF was also shown to increase
beta cell proliferation in adult human islets when cultured on selected extracellular
matrices [7]. This study was the first to demonstrate that human adult beta cells are
indeed able to replicate ex vivo upon exposure to selected growth factors, such as HGF.
In addition, viral delivery of HGF into rodent or other non-human primate islets prior to
transplantation improved islet graft survival and function significantly, suggesting that
HGF gene transfer into islets might improve transplantation outcomes [26]. These studies
suggest that activation of the HGF/cmet pathway in pancreatic beta cells may not only
increase islet mass ex vivo but may also improve beta cell function and survival
[4,6,28,67]. Despite these advantages, the treatment of mature pancreatic beta cells ex
vivo by HGF has been shown to prompt their functional impairment and early senescence
[34]. The proposed mechanism accountable for these observations is epithelial-
mesenchymal transition. It is however unclear from the published studies whether this
de-differentiation process is a direct effect of HGF on pancreatic beta cells or is mediated
indirectly through another cell type.

1.2 Ligand-responsive signaling proteins
Ligand-responsive signaling proteins represent a novel class of fusion proteins designed
to activate specific signal transduction pathways. Dr. Anthony Blau at the University of
Washington has created a series of transgenes encoding ligand-responsive fusion proteins

19
signaling domain
of cmet
drug-binding
domain (F36V)
that combine the intra-cytoplasmic signaling domain of various cytokine receptors with a
specific drug-binding domain: F36V. Receptor homodimerization and thus activation of
the particular signal transduction pathway only occurs in the presence of the specific,
pre-selected family of synthetic ligands, often referred to as “chemical inducers of
dimerization (CID)”[8,15]. The feasibility of this approach has been proven by multiple
studies that achieved the controlled and specific activation of various target pathways.
Kobinger et al. has shown that constitutive expression of fusion proteins with signaling
domains from the gp130 or EGF receptor can lead to moderate mature pancreatic beta
cell expansion [44].
As part of these studies, a novel fusion protein was developed by Dr. Blau’s group
where the signal transduction domain from the cmet protein was fused to the specific
drug-binding subunit (F36V) (Figure 2.1). To date, the activity of this F36Vcmet fusion
protein has only been evaluated in hepatocytes and has been shown to efficiently
stimulate their proliferation while preserving their functionality [51].

Figure 2.1: Schematic drawing of the F36Vcmet fusion protein.

20
2 Hypothesis and Specific Aims

Based on the background detailed above we hypothesized that the expression and
ligand-specific activation of the F36Vcmet fusion protein specifically in pancreatic beta
cells represents an efficient way to expand this cell population. In addition, we proposed
that fusion protein expression under the control of a lineage-specific promoter (insulin
promoter) will better preserve the overall function of the expanded cell populations than
HGF treatment. Thus the following specific aims were investigated:

Specific Aim #1: Design and characterize lentiviral vectors with the potential to
initiate lineage-specific expansion of pancreatic beta cells.
The goal of this aim was to develop a lentiviral vector construct to express the
F36Vcmet fusion protein efficiently and specifically in pancreatic beta cells. We also
aimed to test the ability of the vector to achieve selective beta cell expansion.

Specific Aim #2: Assess the function of the newly-expanded pancreatic beta cells.
The goal of this aim was to test the hypothesis that the function of the pancreatic beta
cells is better preserved with F36Vcmet fusion protein than HGF treatment. We argued
that F36Vcmet expression from the insulin promoter requires active insulin gene
expression and thus will selectively expand those beta cells that retained their insulin
production.

21
3 Materials and Methods

3.1 Lentiviral vector preparations
The plasmid encoding the ligand-inducible signaling fusion protein, F36Vcmet, with an
influenza hemagglutinin (HA) tag at its C-terminal end, was a generous gift of Drs. Blau
and Lieber at University of Washington. Three sets of lentiviral vector constructs were
produced where F36Vcmet expression (Table 2.1.) was directed by the human insulin
gene promoter (1.4 kb Bcl I to Hind III fragment from pFOXCAT1.4, provided by
Michael German, University of California at San Francisco) or from the MND promoter.
Two vector constructs also expressed enhanced green fluorescent protein (eGFP; BD
Biosciences, Mountainview, CA) downstream of the fusion protein either under control
of the internal ribosomal entry site (IRES) or the murine phosphoglycerate kinase
(mPGK) promoter.
All vectors were based on the pCCL self-inactivating (SIN) vector backbone developed
by Dr. Naldini [59] and were packaged using the VSV-G envelope pseudotype and the
p8.9 plasmid [113]. Vector particles were generated via the transient transfection of 293T
cells and were concentrated by ultrafiltration and ultracentrifugation. Quantitative PCR
analysis was used to determine vector titers on human colorectal adenocarcinoma (HT29)
cell lines.

22
3.2 Cell lines
HT29 and murine pancreatic insulinoma cell lines, Nit-1 and BetaTC-6, were purchased
from the American Type Culture Collection (ATCC, Manassas, VA). The HT29 cell line
was maintained in Dulbecco’s modified Eagle medium (DMEM; Mediatech, Inc.,
Herndon, VA) containing 10% fetal bovine serum (FBS; Omega Scientific, Inc., Tarzana,
CA), 2 mM L-glutamine, 100 U/mL penicillin and 100 μg/mL streptomycin (Gemini
Bio-products, West Sacramento, CA). The NIT-1 cell line was cultured in F12K medium
(ATCC, Manassas, VA) supplemented with 10% FBS, L-glutamine, penicillin and
streptomycin. The BetaTC-6 line was cultured in DMEM modified to contain 1.5 g/L
sodium bicarbonate (ATCC), 15% FBS, L-glutamine, and penicillin-streptomycin. All
cell lines were maintained at 37ºC in a humidified carbon dioxide (5%) incubator.

3.3 Cell line transduction
Cell line transduction was performed according to the protocol utilized in the laboratory
of Dr. Kohn. Briefly, 10
5
NIT-1/ beta TC6 cells were plated and transduced the following
day with a pre-determined vector concentration ranging from 10
5
TU/ml (Multiplicity of
infection {MOI}=1) up to 5X10
6
TU/ml (MOI:50) in the presence of 8 μg/ml Polybrene
(Sigma, St. Louis, MO). Transduction efficiency was assessed 5 days after the procedure
either using flow cytometry to detect eGFP expression or immunohistochemistry staining
for HA-tag.

23
3.4 Human adult pancreatic islet culture and transduction
Human adult pancreatic islets were obtained from the National Disease Research
Interchange (NDRI) and the Islet Cell Resource Center Basic Science Human Islet
Distribution program. All studies were performed in accordance with the Declaration of
Helsinki. Islets from ten different donors were used for the experiments with the islet
purity ranging between 35-95%.
Upon their receipt, human pancreatic islets were mechanically dissociated into small
cell clusters and one aliquot was transduced with the Insp-F36Vcmet construct at a final
vector concentration of 1X10
7
TU/ml. Cells were then distributed into the wells of a
96-well plate pre-coated with HTB9 matrix. This matrix is derived from the human
bladder carcinoma cell line (HTB9) and has been shown to provide superior in vitro
culturing conditions for human pancreatic beta cells in previous studies [7].
Approximately 75 islets were dispersed into each well and were cultured for 10 days in
RPMI-1640 medium containing 5.5 mM glucose, 10% FBS, 2 mM L-glutamine,
100 U/mL penicillin and 100 μg/mL streptomycin with daily medium changes.

3.5 Drug treatment
Lyophilized AP20187 (ARIAD Pharmaceuticals, Cambridge, MA) was reconstituted
with ethanol according to the manufacturer’s instructions to prepare a stock solution of
100 μM and aliquots were stored at -20°C until use. Working solutions with a final
concentration of 3-100 nM were prepared by diluting the stock with cell culture medium.
Human recombinant Hepatocyte Growth Factor (hrHGF; R&D Systems, Minneapolis,

24
MN) was used at a final concentration of 50 ng/mL. Drug treatment either with AP20187
(CID) or hrHGF was started on the day following the transduction and was continued
until the cells were collected for analysis (5, 7 and 10 days post-treatment).

3.6 Flow cytometry
To assess eGFP expression, cells were harvested using trypsin, washed three times and
were re-suspended in phosphate buffered saline (PBS). Samples were analyzed on a
FACScan flow cytometer using CellQuest software (Becton Dickinson, San Diego, CA).

3.7 Northern Blot analysis
Nit-1 and BetaTC6 cells transduced with the MND-F36Vcmet-PGK-eGFP vector
construct were expanded until confluent in a 100 mm diameter tissue culture dish. Total
cellular RNA was extracted using an RNeasy Midi Kit (Qiagen Inc., Valencia, CA) as
suggested by the manufacturer. Two micrograms of RNA were separated by agarose gel
electrophoresis and then blotted onto Hybond-XL nylon membrane (Amersham,
Buckinghamshire, UK). The blot was hybridized with
32
P-labeled DNA probe spanning
the F36Vcmet or eGFP region of the vector.

3.8 Immunohistochemical analysis
For immunohistochemical analysis, cytospin slides were prepared. Cells were fixed with
5% paraformaldehyde for 15 minutes and were washed 3 times (5 minutes each) using
PBS with 0.05% Tween-20 (PBS-T; Sigma-Aldrich, St Louis, MO). Blocking and

25
permeabilization was performed in PBS-T containing 10% (vol/vol) normal donkey
serum (Jackson ImmunoResearch, West Grove, PA) and 0.1% Triton X-100
(Sigma-Aldrich, St Louis, MO). Slides were incubated overnight at 4°C with polyclonal
guinea pig anti-insulin antibody (1:200 dilution; Dako, Carpinteria, CA) and HA-probe, a
murine monoclonal antibody to HA (F-7; Santa Cruz Biotechnology Inc, Santa Cruz,
CA). Following the overnight incubation, all slides were washed three times with PBS-T
and incubated for 45 minutes at room temperature in the presence of the following
secondary antibodies: Cy3-conjugated donkey anti-guinea pig antibody (1:250 dilution;
Jackson ImmunoResearch, West Grove, PA) and Alexa-488-conjugated donkey
anti-mouse antibody in 1:500 dilution (Invitrogen, Carlsbad, CA). Slides were washed
multiple times with PBS-T and mounted using Vectashield mounting medium with DAPI
(Vector Laboratories, Burlingame, CA). Fluorescent images were obtained with a Leica
DM RXA upright fluorescent microscope utilizing EasyFISH (Applied Spectral Imaging,
Vista, CA) software. Image quantification was performed by Metamorph (Molecular
Devices, Sunnyvale, CA).

3.9 Cell proliferation assay
Pancreatic beta cell proliferation was assessed by double-staining the islets for insulin
and bromodeoxyuridine (BrdU), followed by fluorescent microscopic evaluation. BrDU
labeling and detection was performed with a commercially available kit (BrdU In-Situ
Detection Kit; BD Biosciences, NJ) with minor modifications to the manufacturer’s
suggested protocol. In brief, islet preparations were incubated with BrdU at 0.1mM for

26
24 hours. Cytospin slides were prepared, fixed and washed 3 times (5 minutes each).
Following a 30-minute permeabilization step, antigen retrieval was carried out by treating
the slides with 2N HCL for 30 minutes. After an extensive washing cycle, slides were
incubated with the following primary antibodies for 1 hour at room temperature:
polyclonal guinea pig anti-insulin antibody in 1:200 dilution and biotinylated anti-BrdU
antibody in 1:10 dilution (provided in the kit). Cells were washed 3 times using PBS-T
and were stained for 45 minutes at room temperature with the following secondary
antibodies: Cy3-conjugated donkey anti-guinea pig antibody in 1:250 dilution and
Alexa-488-conjugated Streptavidin in 1:500 dilution. Slides were washed multiple times
with PBS-T and mounted using Vectashield mounting medium with DAPI. Fluorescent
images were obtained with a Leica DM RXA upright fluorescent microscope and were
analyzed using EasyFISH. Image analysis was performed with Metamorph version
6.1 software (Molecular Devices, Sunnyvale, CA).

3.10 C-peptide ELISA assessing beta cell function
An ultrasensitive, human-specific C-peptide ELISA assay (Mercodia, Uppsala, Sweden)
was used for the quantitative evaluation of pancreatic beta cell function. After
determining basal C-peptide release (5.5 mM glucose exposure for 18 hours), cultures
were sequentially exposed to low (1.6 mM) and high (16.8 mM) glucose concentrations,
each for 1 hour at 37˚C. A stimulation index was calculated using the ratio of C-peptide
release into the high versus low glucose-containing media.

27
4 Results

4.1 Ligand-inducible signaling protein expression in cell lines
Initial studies were aimed to characterize a set of dual-promoter vector constructs
(set #1: MND-F36Vcmet-PGK-eGFP and hInsp-F36Vcmet-PGK-eGFP; Table 2.1) and
their ability to express the fusion protein of interest, F36Vcmet. Using these vectors,
successfully transduced cells can be identified easily by eGFP. The MND promoter is
based on the murine gamma-retrovirus myeloproliferative sarcoma virus and was selected
because of its strong, constitutive gene expression profile observed in multiple cell types.
While the MND-F36Vcmet-PGK-eGFP vector was expected to provide fusion protein
expression non-specifically in different cell types (MND being a constitutive promoter),
we only expected fusion protein expression in insulin-positive cells from the
hInsp-F36Vcmet-PGK-eGFP vector construct.

Table 2.1: Lentiviral vectors encoding the F36Vcmet fusion protein.

Vectors
Set #1
pCCL-hInsp-F36Vcmet-PGK-eGFP
pCCL-MND-F36Vcmet-PGK-eGFP
Set #2
pCCL-hInsp-F36Vcmet-IRES-eGFP
pCCL-MND-F36Vcmet-IRES-eGFP
Set #3
pCCL-hInsp-F36Vcmet
pCCL-MND-F36Vcmet

28
Initial studies with the MND-F36Vcmet-PGK-eGFP vector revealed extremely high
transduction efficiency even at vector concentrations as low as 10
5
TU/ml (MOI: 1) as
assessed by eGFP expression in the murine insulinoma cell line: Nit-1 (Figure 2.2A). In
addition, long-term gene marking was observed (Figure 2.2B).

Figure 2.2: Characteristics of the MND-F36Vcmet-PGK-eGFP vector. Nit-1 cells were
transduced with incremental multiplicity of infection (MOI: 1-10) and transduction
efficiency was analyzed by flow cytometry (A) while long-term gene marking was
assessed by documenting the percentage of eGFP-positive cells over time (B).
A)
0
20
40
60
80
100
120
7 1419 3063 132
Percent eGFP positive cells
Days
MOI: 1
MOI: 5
MOI:10
B)
50.94% 86.94% 96.79%
MOI: 1 MOI: 5 MOI: 10
eGFP expression

29
NT MOI:1 MOI:5
Fusion
protein
transcript
eGFP
transcript
Despite these favorable features, this original vector construct failed to express the fusion
protein efficiently as detected by both Western blot analysis and immunohistochemistry.
Aiming to identify the reason for the minimal fusion protein expression, Northern blot
analysis was performed on non-transduced and transduced Nit-1 cells. Surprisingly, we
found a significant discrepancy between the levels of eGFP and fusion protein expressed
from the MND-F36Vcmet-PGK-eGFP vector construct (Figure 2.3). While the eGFP
transcript was detected at high levels, fusion protein transcript was barely present. Based
on these findings, it was necessary to redesign the original construct and two new sets of
vectors were cloned. When compared to the original vectors, the second set expressed
both eGFP and the fusion protein, while the third set only expressed the fusion protein of
interest.

eGFP PGK MND F36V/cmet
ψ
RRE cPPT
eGFP PGK MND F36V/cmet
ψ
RRE cPPT
A)
B)
Figure 2.3: Transcripts from the
MND-F36Vcmet-PGK-eGFP
construct. A) Schematic
representation of the vector with the
arrows indicating the various
transcripts. ( Ψ: psi packaging signal;
RRE: Rev Response Element; cPPT:
Central Polypurine Track; MND:
MND promoter; PGK: PGK
promoter). B) Northern blot analysis
of Nit-1 cells that were or were not
(NT: Control) transduced with the
MND-F36Vcmet-PGK-eGFP vector at
MOI: 1 and MOI: 5.

30
Aiming to compare the performance of the various vector constructs, protein expression
in Nit-1 cells was assessed by immunohistochemistry. The fusion protein was detected by
taking advantage of the HA-tag located on the C-terminal end of the protein. Our results
revealed significant differences in vector performance (Figure 2.4).

Figure 2.4: Protein expression in cell lines. Nit-1 cells were transduced with the 3 sets of
vector constructs at MOI: 5 and analyzed for fusion protein (detected by HA-tag: red) and
for eGFP (green) expression. DAPI (blue) stains cell nuclei.

pCCL-MND-F36Vcmet-PGK-EGFP

pCCL-hInsp-F36Vcmet-PGK-EGFP

pCCL-MND-F36Vcmet-IRES-EGFP

pCCL-hInsp-F36Vcmet-IRES-EGFP

pCCL-MND-F36Vcmet

pCCL-hInsp-F36Vcmet
DAPI HA-tag eGFP Merged
Set #1
Set #3
Set #2

31
As indicated previously, eGFP was very efficiently expressed by the dual-promoter
constructs (original set of vectors, set#1), while fusion protein expression was barely
detectable by HA-tag staining. Vector set #2 only provided moderate expression of both
proteins. In contrast, we found the single promoter vectors (set #3) to be the most
efficient in expressing the protein of interest, F36Vcmet. In addition, we tested the ability
of the vectors to induce cell proliferation in the presence of AP20187. Results of four
consecutive “growth curve experiments” are shown on Figure 2.5 for the MND-driven
vectors (Figure 2.5A) as well as the Insp-driven constructs (Figure 2.5B). We expected
increased cell proliferation when cells were transduced and treated with AP20187 or were
treated with HGF. Consistent with our immunohistochemistry findings, the single
promoter vectors achieved superior cell proliferation when compared to the original
vector sets or to the IRES constructs. Based on these results and considering the limited
supply of human tissue for research purposes, the single promoter construct driven by the
human insulin promoter (hInsp) was selected for further studies in human pancreatic
islets.

32
0
1
2
3
4
5
6
Fold increase in cell number
pCCL‐MND‐F36Vcmet‐PGK‐eGFP pCCL‐MND‐F36Vcmet‐IRES‐eGFP pCCL‐MND‐F36Vcmet
Nontransduced Transduced
NT CID HGF NT CID HGF
0
1
2
3
4
Fold increase in cell number
pCCL‐Insp‐F36Vcmet‐PGK‐eGFP pCCL‐Insp‐F36Vcmet‐IRES‐eGFP pCCL‐Insp‐F36Vcmet
Nontransduced Transduced
NT CID HGF NT CID HGF
A)

B)

Figure 2.5: Proliferation of Nit-1 cells. Cells were transduced with the three sets of
vector constructs, driven by the A) MND promoter; B) Insulin promoter at the following
vector concentration: MOI: 5. The effects of AP20187 and HGF on cell proliferation
were assessed at day 7 post-treatment. Values shown represent the average for four
separate experiments.

33
0
10
20
30
40
50
60
70
80
90
100
HA-tag positive cells (%)
Insulin- Insulin+
4.2 Ligand-inducible signaling protein expression in human pancreatic islets
Based on the experiments performed in murine insulinoma cell lines, the
Insp-F36Vcmet vector was selected for the human pancreatic beta cell studies. As
evidenced by HA-tag staining (shown on Figure 2.6A), transduced cells were very
efficient in expressing the fusion protein, while no F36Vcmet was detected in the
untransduced, control islet population. Of critical importance, fusion protein expression
was confined to insulin positive cells (Figure 2.6B); on average, 65% of insulin-positive
cells stained with HA-tag antibody. In contrast, merely 1.5% of insulin negative cells
were positive for HA-tag. These data suggested the specificity of the vector for the
insulin-positive pancreatic beta cell population.

DAPI HA-tag Insulin Merged
nontransduced
Insp-F36Vcmet
A)
B)
Figure 2.6: F36Vcmet protein
expression in human pancreatic
islets.
A) Immunohistochemistry of
nontransduced and transduced
human pancreatic islets 7 days
post-transduction. (DAPI: blue;
insulin: red; HA-tag staining:
green). B) Proportion of insulin-
negative (yellow column) and
insulin-positive cells (red
column) that stained positive for
the F36Vcmet protein using
HA-tag staining.

34
4.3 Ex vivo expansion of human pancreatic beta cells
To achieve the ex vivo expansion of pancreatic beta cells, the optimal dose of the CID
drug (AP20187) was initially tested. Insp-F36Vcmet-transduced and control human islets
were maintained in medium for 7 days containing increasing concentrations of AP20187
and were pulsed with BrDU during the final 24 hours of the incubation period. The data
for these experiments are summarized in Table 2.2. The percentage of insulin and BrdU
double positive cells, reflecting the percentage of pancreatic beta cells that are
proliferating, and cell viability were used to determine the most optimal drug
concentration. As indicated by the data, the synthetic drug applied at a concentration of
30 nM prompted the biggest increase in the proportion of insulin and BrdU double
positive cells without major toxic side effects. Thus, AP20187 was utilized at the final
concentration of 30 nM in all consecutive experiments.

35
Table 2.2: Effect of incremental AP20187 concentrations on the percentage of insulin
and BrdU double positive cells. Cell viability for each experimental arm is shown.

AP20187
Dose (nM)
Percent of Insulin and
BrdU Positive Cells
Total Percent of
Viable Cells
Experiment #1
0 3.0 70
10 11.5 73
30 23.0 76
100 15.2 65
Experiment #2
0 4.5 77
10 13.9 86
30 15.0 90
100 13.7 66
Experiment #3
0 2.5 81
10 8.9 90
30 13.8 90
100 11.2 75

Next, the proliferation of beta cells was assessed by determining the BrdU labeling
index of pancreatic beta cells as shown in Figure 2.7. On average, transduced
insulin-positive cells showed over a 4-fold increase in BrdU labeling index in the
presence of AP20187, when compared to the nontransduced or the transduced but
untreated groups. In contrast, HGF treatment induced a 3-fold increase in BrdU labeling
index, independent of viral transduction. Of critical importance, BrdU labeling indices for
untreated controls and untransduced but AP20187 treated cells were similar suggesting
that fusion protein synthesis is required for the synthetic drug to induce cell proliferation.
Interestingly, we also documented a slight increase (1.18-fold) in BrdU labeling for beta
cells that were transduced but not treated.

36
0
100
200
300
400
500
600
BrdU labeling index of beta
cells (% of control)
Nontransduced Transduced
NT CID HGF NT CID HGF

Figure 2.7: Proliferation of human pancreatic beta cells. Relative BrdU-labeling index of
pancreatic beta cells for each experimental arm, expressed as percentage of the
nontransduced, nontreated control. The total percentage of insulin and BrdU
double-positive cells equaled 3%; values shown represent the average of 6 separate
experiments. 1000 cells were counted per experimental during each experiment. (NT:
nontreated; CID: treatment with AP20187 at 30 nM; HGF: treatment with Hepatocyte
Growth Factor at 50 ng/ml).

Next, the proportion and absolute number of insulin positive cells were determined for
the transduced experimental groups 7 days post-treatment. The results are summarized in
Table 2.3. The data indicate that both the percentage and absolute number of
insulin-positive cells increased upon AP20187 treatment compared to the nontreated
group, further confirming the selective expansion of beta cells by our approach. In
contrast, HGF treatment reduced the proportion of insulin positive cells in the cultures.
On average, our approach achieved a 1.8 fold increase in the absolute number of insulin-
positive cells that is comparable to that of HGF treatment (1.6 fold).
3%

37
Table 2.3: Proportion and absolute number of insulin positive cells in human pancreatic
islet cultures, 7 days post-transduction (Vector construct used: hInsp-F36Vcmet).

Experiment
Total Cell
Number
Percent
Insulin+
Cells
Absolute
Number of
Insulin+ Cells
Fold Increase in
Absolute Number of
Insulin+ Cells
Transduced,
Nontreated
1 100,000 30 30,000 1
2 100,000 31 31,000 1
3 60,000 33.4 20,040 1
4 60,000 23.2 13,920 1
5 160,000 33 52,800 1
Average 96,000 30.1 29,552 1
Transduced
with
AP20187
1 150,000 41.8 62,700 2
2 120,000 38.9 46,680 1.5
3 74,000 46.8 34,632 1.7
4 78,000 46.8 36,604 2.6
5 188,800 40 75,520 1.4
Average 242,160 42.9 51,227 1.8
Transduced
with
HGF
1 204,000 24.8 50,592 1.7
2 208,000 26.2 54,496 1.8
3 94,000 28.9 27,166 1.4
4 96,000 18.3 17,568 1.3
5 280,000 30 84,000 1.6
Average 176,400 25.6 46,764 1.6

4.4 Human pancreatic beta cell expansion over time
The absolute number of insulin-positive cells in the transduced cell populations was also
assessed at days 5, 7, and 10 post-treatment and was graphed as fold difference compared
to the transduced, nontreated control (Figure 2.8). Interestingly enough, we observed
different dynamics of events utilizing the F36Vcmet protein versus HGF treatment. With
our approach we experienced a 1.8 fold increase in the absolute number of insulin
positive cells compared to the control as early as 5 days post-treatment and this value

38
0.00
0.50
1.00
1.50
2.00
2.50
Fold increase in absolute number of
insulin+ cells (compared to control)
Transduced +CID Transduced +HGF
Day 0 Day 5 Day 7 Day 10
remained constant over the course of the experiments. On the other hand, HGF elicited its
effect somewhat later with its peak effect on day 7 with a drop in the absolute number of
insulin positive cells at day 10.

Figure 2.8: The absolute number of insulin positive cells over time. Fold increase in the
absolute number of transduced insulin positive cells in pancreatic beta cell cultures over a
10-day period. Values are compared to the transduced, but not treated cells, represented
by the dotted line.

4.5 Functional analysis of the newly-expanded cells
Basal C-peptide release and glucose-induced C-peptide release were analyzed to assess
the function of the expanded cells. Since C-peptide is released in a 1:1 ratio with insulin
from pancreatic beta cells and its level is less affected by factors such as ingredients
present in culture media we chose this measurement for the functional assays. The data
for both assays are presented on Figure 2.9.

39
0.00
0.05
0.10
0.15
0.20
Basal C-Peptide release
(pMol/L/cell/18h)
0.00
0.50
1.00
1.50
2.00
2.50
3.00
Stimulation Index
(fold increase)
NT CID HGF NT CID HGF
Nontransduced Transduced

Figure 2.9: Functional analysis of human pancreatic beta cells. Basal C-peptide release
(top figure) and glucose-induced C-peptide release (commonly referred to as stimulation
index; bottom figure) was analyzed 7 days post-treatment. (NT: nontreated; CID:
treatment with AP20187; HGF: treatment with Hepatocyte Growth Factor).

Initially, 18 hours basal C-peptide release was compared in non-transduced and
transduced β cell cultures both in the presence and absence of CID or HGF. As shown on
the top part on Figure 2.9, transduction itself had no significant effect on beta cell
functionality while CID treatment only affected transduced β cells prompting a moderate
decrease in their overall functionality. HGF therapy reduced C-peptide release in both
groups to values that were significantly below control.

40
C-peptide release upon glucose challenge was also assessed and is shown on the bottom
half of Figure 2.9. This measurement indicates the cells’ ability to respond to increased
concentrations of glucose; a critical function of pancreatic beta cells. Stimulation index
reflects the ratio of the amount of C-peptide released to high versus low glucose
containing medium. Interestingly enough, we observed a similar trend for this
measurement as seen for basal C-peptide release further suggesting that our approach
better preserves the function of the expanded cells; however this difference did not reach
statistical significance.

5 Discussion

Replacing pancreatic beta cells through pancreatic islet transplantation represents a
novel, curative approach to treat T1D where the main pathophysiology is the autoimmune
destruction of pancreatic beta cells. This technique however currently faces two major
challenges. Beside the immunological rejection of the newly transplanted cells, the
significant shortage of donor tissue also represents a big challenge. Thus significant
efforts are focusing on finding new ways to generate insulin-producing cells; among
these are approaches aiming to achieve the ex vivo expansion of pancreatic beta cells.
HGF has received much attention in this regard with several published reports
suggesting its ability not only expand pancreatic beta cells, but also to improve their
survival both ex vivo and in vivo. The utility of HGF is limited by its non-specific nature
and its negative effect on the overall functionality of pancreatic beta cells. It has been

41
suggested that this functional decline is primarily attributable to epithelial-mesenchymal
transition (EMT) induced upon HGF treatment [81]. However, it is currently unclear
whether EMT results from the direct effects of HGF on pancreatic beta cells or is caused
by indirect mechanisms via other cell types always present in islet preparations.
In order to address this question and to achieve the beta cell-specific activation of the
HGF/cmet signaling pathway, a series of lentiviral vectors were designed. These vectors
express a novel, ligand-inducible fusion protein (F36Vcmet) that is only functional in the
presence of a synthetic ligand: AP20187 (CID). Utilizing this approach we aimed to
achieve specific pancreatic beta cell expansion while preserving cellular functionality
better than with HGF treatment. We argued that fusion protein expression from the
insulin promoter will allow the selective expansion of those insulin positive cells that
retained their functionality, since insulin promoter activity requires active insulin gene
expression.
We have utilized 10 human pancreatic islet preparations for our studies with cell
viability exceeding 80%, as determined by Trypan blue exclusion. The purity of the islet
preparations was determined by Dithiazone staining and it ranged between 35% and 95%.
Utilizing our approach we were able to achieve the specific expansion of pancreatic beta
cells with an average of 4-fold increase in their proliferation and with an overall of
1.8 fold increase in the absolute number of insulin-positive cells compared to control
cells over a 7 day period. This magnitude of cell proliferation was comparable to that
achieved with HGF. Importantly, our data suggested that the function of the cells

42
expanded via our approach was somewhat better preserved than of those cells that were
expanded by HGF.
Among the series of lentiviral vectors studied, the single promoter vector constructs
provided superior protein expression over the other constructs tested. The dual promoter
vectors (Table 2.1, set #1) did not perform well in our studies. One explanation might be
a phenomenon called “promoter interference” that reflects the promoters’ ability to
interfere with the function of each other and thereby modifying gene expression levels.
Similarly, we could not exclude the possibility that the transcript encoding the fusion
protein was unstable in this system resulting in extremely low levels of fusion protein.
Vectors utilizing the IRES segment have been described to occasionally provide low
protein levels which are consistent with our observation.
Utilizing insulin promoter to drive the expression of F36Vcmet allowed us to
specifically express the F36Vcmet fusion in pancreatic beta cells. The specificity of our
approach was supported by immunohistochemical data indicating that the vast majority
of the cells expressing the fusion protein were indeed also insulin-positive. Interestingly
however, we found that approximately 1% of the non-insulin positive cells also expressed
the fusion protein. We speculated that this observation can be due to the “leakiness” of
the vector or alternatively can be explained by the de-differentiation of previously
insulin-positive cells.
Importantly, increased proliferation of pancreatic beta cells was achieved when the cells
were transduced with the novel vector construct and treated with the synthetic ligand:
AP20187. No proliferation advantage was observed when the cells were nontransduced

43
or transduced, but not treated with AP20187. Consistent with these findings, the selective
expansion of beta cells via this novel approach was further indicated by the increase in
the proportion of insulin-positive cells over time. Interestingly, we consistently detected
an approximately 4-fold increase in the proportion of proliferating beta cells as assessed
by BrDU staining (compared to the nontransduced and nontreated control at day 7), while
the absolute number of insulin-positive cells only increased by 1.8-fold (vs. control at
day 7). We hypothesize that the inherent differences between the two techniques may be
responsible for this discrepancy. While BrDU labeling accurately reflects the proportion
of cells in the S-phase of the cell cycle, it does not necessarily mean that these cells
complete the cycle with successful cell division. In addition, BrDU may also label those
cells that undergo DNA synthesis for any other reasons, such as DNA repair. Despite
these limitations, BrdU labeling is a widely used technique to assess the proliferation of
various cell types, including pancreatic beta cells. In contrast to our approach, we found
that, despite the increased beta cell proliferation observed with HGF treatment, the
overall proportion of insulin-positive cells decreased in this experimental arm, quite
likely due to the non-specific nature of this growth factor. These data are consistent with
the findings of Lefebre et al [49] indicating that HGF has a significant mitogenic effect
on pancreatic ductal cells, but somewhat contradict of those reported by Beattie, et al.
indicating the preference of HGF for the beta cell population with negligible effect on
ductal cells [5]. However, it is important to note that the study conducted by Beattie et al.
utilized islet preparations pre-purified to maximize pancreatic beta cell number with
minimal contamination from other cell types. An alternative explanation to the

44
observations is the de-differentiation of the previously insulin-positive cells. We
speculate that our approach preferentially expands those cells that retained insulin
positivity, since the fusion protein is expressed from the insulin promoter.
Finally, the data indicate that the function of the cells expanded via our approach is
somewhat better preserved than those expanded by HGF. It is tempting to speculate that
by using the insulin promoter to drive the expression of the fusion protein, we selected
for those cells in the culture that retained their function.
A major limitation of this study was the low number of human islets available for each
experiment, possibly limiting the variety of assays used for the studies. In addition, the
use of lentiviral vectors to deliver the fusion protein may represent a safety concern
[32,33]. Lentiviral vectors integrate into the host genome and have the potential to
activate oncogenes or to interfere with the activity of tumor suppressor genes. Thus, they
might have some oncogenic potential. However, they are considered safer in this regard
when compared to their retroviral counterparts because they do not have a preference to
integrate into gene promoter regions.
In summary, this study represents a novel gene therapy-based approach to specifically
expand human pancreatic beta cells. As such, it may provide an alternative way to
generate insulin-producing cells for islet transplantation that represents a potential cure
for many millions of patients with T1D.

45
CHAPTER III
Lentiviral vectors as a tool to identify insulin-positive cells derived from
human embryonic stem cells

1 Background

1.1 Overview of human embryonic stem cell (hESC) research
Human embryonic stem cells (hESC) were first described in 1998 by a research group
directed by Dr. James Thomson [97]. Their paper published in Science marked the
beginning of a new and exciting, but at the same time, controversial research era related
to human embryonic stem cells. Since their discovery, several hESC lines have been
generated from the inner cell mass of non-implanted, blastocyst-stage embryos originally
derived for in vitro fertilization procedures. These cell lines are now maintained and
utilized for research in laboratories worldwide.
The two main characteristics that are hallmarks of ES cells are their self-renewal
capacity and pluripotency. Self-renewal capacity describes the cells’ ability to proliferate
and grow indefinitely in culture without undergoing senescence or any karyotypic
changes. On the other hand, pluripotency reflects the cells’ potential to differentiate into a
wide range of somatic and extra-embryonic tissues both in vitro and in vivo. These
tissues are of endodermal, mesodermal and ectodermal origin. Of interest, pluripotency is

46
not a unique characteristic of ES cells. Cells of the epiblast, primordial germ cells,
embryonal carcinoma cells and the most recently discovered “induced pluripotent stem
cells (iPS)” all possess this characteristic [13,57,63,94,99,109].
Currently hES cells are identified and characterized based on information gathered from
three complimentary research methods [60]. Characteristic cell surface antigen
expression (such as: GCTM-2, TRA-1-60, TRA-1-81, SSEA-4) is one of the technique
that is used. When combined with the analysis of pluripotency associated
gene-expression (such as Oct-4, Nanog, Sox-2) along with a biological assay testing the
differentiation potential of the cells (often by testing the ability to form teratomas when
injected into immune-deficient murine host) the combined data are currently sufficient to
identify pluripotent ES cells. It is of great importance to note that none of these methods
themselves are satisfactory to characterize cells as pluripotent stem cells.
The remarkable plasticity of ES cells and their potential to multiply indefinitely provide
us with the opportunity to utilize hES cells as a potentially unlimited, renewable source
of cells, tissues and organs, including neurons [20,91], cardiomyocytes [83,104],
hepatocytes [62,110] and insulin-producing pancreatic beta cells [19,41,46]. These
incredible features of ES cells give us hope that someday many of the currently
devastating diseases can be treated via a stem cell-based approach. But, before hES-based
therapies can be widely utilized to treat humans, many challenges have to be overcome.
It is a long-lasting desire to find an optimal, inexpensive, chemically-defined culture
condition that will support the long-term maintenance of karyotypically normal,
pluripotent hES cells. The cost associated with current culture conditions along with the

47
undefined nature of some of the culture ingredients are among the biggest limitations
hindering the clinical utility of these cells. In addition, most of the cell lines that are
currently available were derived from low quality tissues and were exposed to various
animal products further inhibiting their utility for clinical applications. Thus it is
absolutely necessary to derive new cell lines and maintain them in a way that makes them
appropriate for human use. An additional big obstacle is our limited knowledge to drive
the differentiation of hES into various cell lineages with high efficiency. Although there
are several reports suggesting that recapitulating the key steps of embryonic development
of various organs in vitro may be a feasible approach to derive different cell types current
protocols fail to provide cells with high purity.

1.2 Human embryonic stem cells as the source of insulin-positive cells
Initial reports suggested that hES cells have the ability to spontaneously differentiate
into insulin-producing cells, albeit at an extremely low frequency [1]. Using selective
culture conditions that were initially described in studies utilizing murine ES cells [55]
did not improve the low derivation efficiency highlighting the need to develop new
differentiation protocols.
The most recent protocols highly rely on the developmental biology knowledge gained
from mouse models and aim to utilize this information to drive the differentiation of
hESC into the pancreatic beta cell lineage. Great examples of this approach have been
published by Novocell Inc. (La Jolla, CA) and Geron Corp. (Menlo Park, CA),
suggesting that recapitulating pancreas development in vitro may represent a feasible

48
approach to generate insulin-producing cells [18,19,41,46]. However, even with the most
successful protocols available to date (D’Amour et al, Kroon et al.), merely 8-12% of
hES cells become insulin-positive by the end of the differentiation process. Even more
concerning is that none of these protocols yielded fully functional beta cells. Thus, there
is a great interest in developing novel or improved differentiation protocols.

1.3 Lentiviral vectors as tools to study hES differentiation
Current technology allows temporary as well as permanent gene marking to be achieved
using lentiviral vector systems. In addition, expression from these vectors can be
restricted to specific cell types depending on the env protein and the promoter sequence
utilized. Lentiviral vectors are uniquely able to transduce quiescent cells, a feature that
makes them especially attractive for hES cell research. As a consequence, they are
increasingly often used in hES cell studies with various aims, ranging from reporter gene
expression through the delivery of active genes [2,37,70,98,103]. Cell-specific targeting
of various genes represents an especially useful tool to identify, enrich, select and modify
certain cell populations, both in vitro and in vivo [14,107]. HES cell differentiation
studies can especially gain from this feature considering the extremely heterogeneous cell
populations that appear during the currently available differentiation protocols.
Certain criteria need to be met for the vector to be used as a tool to identify a specific
cell type derived from hES following in vitro differentiation. First, it has to be expressed
specifically in the cell type of interest with minimal “background” expression only.
Second, it has to reliably identify the appearance of the cell type of interest in the culture.

49
For instance, reporter gene expression from a vector that is designed to identify
insulin-positive cells should coincide with the appearance of insulin positive cells during
the differentiation protocol. Third, the vector needs to be protected from epigenetic
effects, a goal that is currently extremely challenging, if at all possible, to achieve. The
most frequent epigenetic modification observed in hES vector studies is called gene
silencing. Gene silencing refers to the absence of gene expression in the target cells
despite the presence of the vector provirus in the host’s genome. It is a frequently
observed phenomenon in cases when vectors are introduced into undifferentiated hES
cells that are then driven through a differentiation process. Differentiation itself is
accompanied by dramatic changes in the cellular epigenetic status often leading to the
inactivation of a previously active genomic region. Thus it is possible that the initially
active genomic region of a transgene becomes inactivated diminishing expression from
the transgene sequence. Insulators incorporated into the vector constructs can
theoretically provide some protection against these epigenetic effects.
In summary, lentiviral vectors with the appropriate gene expression profile can serve as
an extremely useful tool to monitor the differentiation of hES into various cell types. This
thesis work focuses on utilizing an insulin promoter-driven vector expressing eGFP to
monitor the differentiation of hES into the pancreatic beta cell lineage.

50
2 Hypothesis and Specific Aims

Our hypothesis was that a lentiviral vector, in which insulin promoter drives the eGFP
marker gene expression, will reliably indicate the appearance of insulin positive cells in
hES cultures that were differentiated into the pancreatic beta cell lineage.

Specific Aim #1: Establish a successful and efficient differentiation protocol in the
laboratory that drives the differentiation of hES cells into the pancreatic beta cell
lineage.

Specific Aim #2: Test the hypothesis that hInsp-eGFP vector can be used as a tool to
identify insulin-positive cells derived from hES.

3 Materials and Methods

3.1 Culturing hES cells
H1 and H9 human embryonic stem cell lines provided by the Stem Cell Core of
Childrens Hospital Los Angeles were utilized in our studies. Cells were cultured on
irradiated murine embryonic fibroblast (MEF) feeder layer. In brief, feeder layers were
prepared by irradiating MEF cells with 55 Gy and plating them onto gelatin pre-coated
6-well plates with a density of 10
6
cells/well. Human embryonic stem cell lines were kept
in DMEM/F12 culture medium supplemented with 20% knock-out serum replacement,

51
1mM L-glutamine, 1% nonessential amino acids, 4ng/ml basic fibroblast growth factor
(all products obtained from Gibco, Carlsbad, CA) and 0.1mM beta-mercaptoethanol
(Sigma, St Louis, MO). Media over the cells were changed on a daily basis and cells
were mechanically passaged 1:3 every 5 days using a commercially available passaging
tool (Stem-Pro EZ Passage Disposable Stem Cell Tool, Invitrogen, Carlsbad, CA).

3.2 Lentiviral vectors
All HIV-1-based lentiviral vectors used in the present study were based on the pCCL
self-inactivating (SIN) vector backbone developed by Dr. Naldini [113]. Two internal
promoters were used in these studies: the 570 basepair (bp) murine phosphoglycerate
kinase promoter (PGK) and the 1.4 kb human insulin promoter (hInsp). These promoters
were driving the expression of the enhanced green fluorescent protein (eGFP; BD
Biosciences, Mountainview, CA).
Lentiviral vector supernatants were produced by triple transfection of 293T cells via
DNA/calcium phosphate co-precipitation. Vector supernatants were harvested,
ultrafiltered and then ultracentrifuged to achieve high vector concentrations. Vector
concentrations were determined by transducing HT29 cells with serial dilutions of the
vector preparations and vector titers were calculated based on FACS analysis. The titer of
the vectors in the present study was 10
8
TU/ml for hInps-eGFP and 5X 10
9
TU/ml for the
PGK-eGFP vector.

52
Oct-4
Sox-2
Nanog
SOX-17
FOX A2
CXCR4
SOX-17
FOX A2
HNF1b
SOX-17
FOX A2
HNF1b
PDX-1
SOX-17
FOX A2
HNF1b
PDX-1
Nkx2.2
Ngn-3
insulin
Stage 1. Stage 2. Stage 3. Stage 4.
Oct-4
Sox-2
Nanog
SOX-17
FOX A2
CXCR4
SOX-17
FOX A2
HNF1b
SOX-17
FOX A2
HNF1b
PDX-1
SOX-17
FOX A2
HNF1b
PDX-1
Nkx2.2
Ngn-3
insulin
Oct-4
Sox-2
Nanog
SOX-17
FOX A2
CXCR4
SOX-17
FOX A2
HNF1b
SOX-17
FOX A2
HNF1b
PDX-1
SOX-17
FOX A2
HNF1b
PDX-1
Nkx2.2
Ngn-3
insulin
Stage 1. Stage 2. Stage 3. Stage 4.
3.3 Transducing hESC with lentiviral vectors
HESC were transduced on the last day of stage 2 of the differentiation protocol using a
final vector concentration of 5X10
6
. 1 ml was the total volume of the transduction media
and it was supplemented with 8 μg/mL polybrene (Sigma, St. Louis, MO).

3.4 Differentiating hESC into the pancreatic beta cell lineage
HES were differentiated into the pancreatic beta cell lineage utilizing a protocol
published by Kroon, et al. in Nature Biotechnology [46]. This particular protocol drives
hES cells through 4 stages of differentiation as indicated on Figure 3.1. The figure also
provides a brief summary for each stage listing the types of media and growth factors
used along with the markers that are characteristic of each stage. It is important to note
that the protocol omits the use of serum at later stages of the differentiation with
insulin-free B27 replacing serum components.

Figure 3.1: Schematic representation of the protocol used to differentiate hES cells into
the pancreatic beta cell lineage. All stages of the protocol are shown with stage-specific
culture conditions and characteristic gene-expression profile. (ES: Undifferentiated hES
cells; ME: Mesendoderm; DE: Definite Endoderm; PG: Primitive Gut Tube PF: Posterior
Foregut; PE: Pancreatic Endocrine Cells). Figure modified from Kroon, et. al, [46].

53
3.5 Monitoring hESC differentiation
To assess whether hES cells have reached a certain stage of the differentiation, the
expressions of multiple stage-specific genes were assessed using RT-PCR. In addition,
we characterized the cultures for insulin, C-peptide and eGFP expression via
immunohistochemistry.

a) RT-PCR
Total cellular RNA was extracted at each stage of the differentiation utilizing RNeasy
Mini Kit (Qiagen, Valencia, CA). Reverse transcription was performed using the
Omniscript kit (Qiagen, Valencia, CA) and oligoDT primers (Amersham Biosciences,
Piscataway, NJ). PCR reactions were performed utilizing HotStar Taq Master Mix kit
with the primer sets listed in Table 3.1. The following PCR conditions were utilized: One
initial cycle of denaturation at 95°C for 15 minutes was followed by 35 cycles of
denaturation at 94°C for 1 minute each and annealing at the appropriate annealing
temperature for 1 minute with extension at 72°C for 2 minutes. The final incubation was
performed for 10 minutes at 72°C. The annealing temperatures were the following: for
Oct-4, Nanog, CXCR4, FOXA2 and ngn-3: 60°C; for Sox-2, HNF-1b: 58°C; for PDX-1,
insulin: 53°C and for β2-microglubulin: 55°C. PCR products were visualized on 2%
agarose gels.

54
Table 3.1: Primers utilized for RT-PCR analysis.

Gene Primers Product Size (bp)
Oct-4
5’-AGCTGGAGAAGGAGAAGCTGG-3’
5’-TCGGACCACATCCTTCTCGAG-3’
234
Sox2
5’-CCGGCGGCAACCAGAAAAACAG-3’
5’-ACCCCGCTCGCCATGCTATTG-3’
340
Nanog
5’-GCAAACAACCCACTTCTGC-3’
5’-AGGCCTTCTGCGTCACAC-3’
155
FOXA2
5’-CTACGCCAACATGAACTCCA-3’
5’-AAGGGGAAGAGGTCCATGAT-3’
207
SOX-17
5’-GGGATACGCCAGTGACGACC-3’
5’-GAAGCACCTCCTCCGTCTCG-3’
350
HNF-1b
5’-GGTCTTCCATACTCTCACCAACGG-3’
5’-ACGCTTCTGGGTCTTCATAGGG-3’
316
PDX-1
5’-CCCATGGATGAAGTCTACC-3’
5’-GTCCTCCTCCTTTTTCCAC-3’
262
NKX-2.2
5’-TCTACGACAGCAGCGACAAC-3’
5’-GCGTCACCTCCATACCTTTC-3’
393
Ngn-3
5’-AAAGGATGACGCCTCAACCC-3’
5’-TCAGTGCCAACTCGCTCTTAGG-3’
237
Insulin
5’-TGTGAACCAACACCTGTG-3’
5’-CGTCTAGTTGCAGTAGT-3’
260
Beta2-microglobulin
5’-CTCGCGCTACTCTCTCTTTC-3’
5’-CATGTCTCGATCCCACTTAAC-3’
330

b) Immunohistochemistry
Immunohistochemical analysis was performed according to the protocol described in
section II.3.9. of this dissertation. Briefly, cytospin slides were prepared from cells at
each stage of the differentiation and were stained with the following primary antibodies:
polyclonal guinea pig anti-insulin antibody (1:200 dilution; Dako, Carpinteria, CA),
rabbit anti human C-peptide antibody (1:400 dilution; Millipore, Billerica, MA), goat
polyclonal PDX-1 antibody (1:5,000 dilution; Abcam, Cambridge, MA) and polyclonal

55
rabbit anti-GFP antibody (1:1,000 dilution; Invitrogen, Carlsbad, CA). The following
secondary antibodies were used: Cy3-conjugated donkey anti-guinea pig antibody (1:250
dilution; Jackson ImmunoResearch, West Grove, PA), Alexa-488-conjugated donkey
anti-goat antibody in 1:500 dilution (Invitrogen, Carlsbad, CA), Cy3-conjugated donkey
anti-rabbit antibody (1:250 dilution; Jackson ImmunoResearch, West Grove, PA) and
Alexa-488-conjugated donkey anti-rabbit antibody (1:500 dilution; Invitrogen, Carlsbad,
CA ). Slides were mounted using Vectashield mounting medium with DAPI (Vector
Laboratories, Burlingame, CA). Fluorescent images were obtained with a Leica DM
RXA upright fluorescent microscope utilizing EasyFISH software. Image analysis was
performed with Metamorph (Molecular Devices, Sunnyvale, CA).

4 Results

4.1 Lineage-specific protein expression from hInsp-eGFP
Initially, we tested the hInsp-eGFP vector in a cell lines with and without insulin
expression. From this particular construct, eGFP expression was only observed in those
cell lines that were positive for insulin, thus suggesting the specificity of this particular
vector for the insulin producing cells (data not shown; see Reference 84 ). Next, we
tested the construct in human pancreatic islets and a representative dataset is presented on
Figure 3.2. We experienced relatively high (~18%) eGFP-expression in pancreatic islets
transduced with hInps-eGFP at a final vector concentration as low as 3X10
6
(Figure 3.2A). In addition, consistent with the cell line data, eGFP expression from the

56
hInsp-eGFP vector was confined to the insulin positive human pancreatic beta cells,
providing further evidence that our approach is highly specific (Figure 3.2B). In contrast
and as expected, global gene expression in multiple cell types was documented from the
PGK-eGFP vector that we utilized as positive control in our experiments.

Figure 3.2: eGFP expression in human pancreatic beta cells. Islets were transduced with
the hInsp-eGFP construct at a final vector concentration of 3X10
6
TU/ml. A) eGFP
expression as determined by flow cytometry and B) Immunohistochemical analysis of
nontransduced and transduced human pancreatic islets.

4.2 Differentiating hES cells into the pancreatic beta cell lineage
Differentiating hES cells into the pancreatic beta cell lineage turned out to be an
extremely challenging task despite our efforts trying to reproduce established and
published protocols in our laboratory. Initially, we attempted to adapt a differentiation
protocol published by Jiang, et al. in the journal Stem Cells [41]; however, our efforts
proved unsuccessful. Significant cell death was consistently encountered at stage 2 of this
Nontransduced Transduced
18.4%
A)
B)

57
B2M
Insulin
PDX-1
MEF S1 S2 S3
NT Insp PGK
Panc H
2
0 UD
RT+
RT -
RT+
RT+
RT -
RT -
NT Insp PGK D
S4
protocol with essentially no viable cells remaining by stage 3 [data not shown]. After
multiple failures, we decided to test another protocol published by Kroon, et al. in Nature
Biotechnology in 2008 (please see summary on Figure 3.1.). Utilizing this approach, we
were successful in detecting insulin positive cells by RT-PCR in both H1 and H9 cultures
at the final stage of differentiation. Data for the H1 cell line is shown on Figure 3.3. Since
the protocol was performed on hES cells cultured on MEF feeder layers; human-specific
primer sets were used for the RT-PCR reactions. RNA from MEF cells served as the
negative control while RNA derived from human pancreas was used as positive control
throughout these experiments. Our results were reproducible and insulin expression was
consistently observed with this approach.

Figure 3.3: Successful differentiation of H1 cells into the pancreatic beta cell lineage as
indicated by RT-PCR analysis. (MEF: Murine Embryonic Fibroblast; UD:
Undifferentiated hES cells; S1: Stage 1; S2: Stage 2; S3: Stage 3; S4: Stage 4; Panc:
Pancreas; NT: nontransduced; Insp: hInsp-eGFP; PGK: PGK-eGFP transduced cells).
PCR amplification was carried out for 35 cycles.

58
DAPI Insulin Merged
DAPI C-peptide Merged Merged DAPI C-peptide Merged Merged
Furthermore, as shown on Figure 3.4, not only insulin RNA was present, but also insulin
as well as C-peptide protein expression was detected by immunohistochemistry. The
proportion of insulin positive cells varied among individual experiments and tested cell
lines, with an average of 5% of the cells being insulin positive. According to our data, H1
cells were more likely to generate insulin-positive cells than H9 cells.

Figure 3.4: The successful differentiation of H1 cells into the pancreatic beta cell lineage
is confirmed by immunohistochemistry.

4.3 Transduction does not interfere with the differentiation process
After establishing the differentiation protocol, it was critical to test whether transduction
with our vector constructs (hInps-eGFP and PGK-eGFP) would interfere with the
differentiation process. If transduction hinders hES cell differentiation into the beta cell
lineage then, evidently, these vectors cannot be utilized as tools to monitor the
differentiation process. To evaluate this possibility, we differentiated non-transduced H1

59
NT
INSp-
eGFP
PGK-
eGFP
NT
INSp-
eGFP
PGK-
eGFP
MEF S1 S2 Panc H
2
0 UD
Insulin RT+
PDX-1
CXCR4
Oct4
SOX2
Nanog
FOX A2
HNF1B
NGN3
Insulin RT -
Β2-microglobulin RT+
Β2-microglobulin RT-
S3 S4
cells and those that were transduced with either hInsp-eGFP or PGK-eGFP vector
construct parallel into the beta cell lineage. Representative results are shown in
Figure 3.5. As evidenced by RT-PCR analysis, we did not found significant differences in
the gene expression profile of the transduced and non-transduced cell populations.

Figure 3.5: Gene expression profile of transduced and nontransduced H1 cells.
Transduction had no significant effect on the relative gene expression profile of H1 cells.
(MEF: Murine Embryonic Fibroblast; hES: Undifferentiated H1 Human Embryonic Stem
Cell; S 1-4: Differentiation Stages 1-4; NT: nontransduced; Panc: Human Pancreas; H
2
O:
Water). PCR amplification was performed for 35 cycles.

60
nontransduced
hInsp-eGFP-WPRE
PGK-eGFP
DAPI Insulin eGFP Merged
nontransduced
hInsp-eGFP-WPRE
PGK-eGFP
DAPI Insulin eGFP Merged
4.4 Using hInsp-eGFP as a tool to identify insulin positive cells derived from hES
cells
In the following experiment we tested the hypothesis that the Insp-eGFP vector can be
used as a novel tool to identify cells that become insulin positive during the
differentiation process. For these purposes, H1 and H9 cells were transduced on the final
day of stage 2 of the differentiation protocol and the cultures were monitored for
eGFP-expression. E-GFP-expression was not detectable at stage 3 if the cells were not
transduced or were transduced with the hInsp-eGFP vector as shown on Figure 3.6. In
contrast and as expected, eGFP expression from the PGK-eGFP construct was detectable
as early as stage 3.

Figure 3.6: eGFP-expression at stage 3 of the differentiation protocol, as detected by
immunohistochemistry.

To confirm that the vector was indeed present even in the non-expressing cell
populations, vector copy analysis was performed via quantitative PCR. According to

61
DAPI Insulin eGFP
nontransduced
hInsp‐eGFP‐WPRE
these data, the average vector copy number ranged between 0.7-1 copy/cell for both the
PGK-eGFP and hInsp-eGFP transduced cell populations indicating the presence of the
construct in both groups. At stage 4 of the differentiation, both hInsp-eGFP and
PGK-eGFP transduced cells showed eGFP expression as illustrated on Figure 3.7. As
expected, eGFP-expression from the constitutive promoter-driven vector PGK-eGFP was
robust, while merely 1% of cells transduced with hInsp-eGFP expressed the protein.
Nevertheless, all cells that expressed eGFP in the hInsp-eGFP transduced cell population
were insulin positive suggesting the unique ability of the vector to specifically detect
cells that are insulin positive.

Figure 3.7: eGFP-expression at stage 4 of the differentiation protocol as detected by
immunohistochemistry. eGFP expression reliably identified cells that became insulin
positive at the end of the differentiation protocol.

62
To further confirm that the hInsp-eGFP vector reliably identifies insulin positive cells,
we compared eGFP expression from this vector in cells that completed the differentiation
protocol versus those that were stopped at stage 3 of the protocol [data not shown]. No
eGFP-expression was detected from cells that remained in stage 3 and, consistent with
this observation, no insulin expression was detected in these cultures. In contrast, eGFP
expression could be detected in cultures that completed the entire differentiation process.
Importantly, all cells that expressed eGFP were indeed insulin positive indicating the
specificity of our approach to identify this particular cell population.

4 Discussion

HES cells represent a potentially abundant source for all cell types of the human body
and are thereby considered an invaluable asset for the purposes of regenerative medicine.
Several difficulties need to be addressed, however, before this precious cell source can be
utilized to manage or treat human diseases. Among the obstacles is our current inability
to drive the differentiation of these cells into a specific target cell type with high
efficiency, in clinical-grade quality and in sufficient quantities. As an example, it is
estimated that approximately 1 million pancreatic beta cells need to be transplanted per
patient in order to sufficiently alleviate the symptoms of diabetes. Current protocols
aiming to drive the differentiation of hES cells into the pancreatic beta cell lineage not
only fail to provide pure cells in such high quantities, but are also unable to generate
mature, fully functional beta cells. Thus, there is a critical need to improve current

63
procedures and to develop novel differentiation protocols that meet all these stringent
requirements.
Lentiviral vectors have the potential to indicate the successful differentiation of hES
cells into a certain target cell lineage. In order to succeed, these vectors need to contain
lineage-specific promoters that drive the expression of a marker gene (e.g., eGFP in our
studies). For example, gene expression from the lineage-specific insulin promoter
requires that insulin to be expressed in the target cells. As such, insulin promoter-driven
vector constructs can theoretically be utilized to specifically identify cells in a hES
culture that successfully differentiated into the pancreatic beta cell lineage and, consistent
with their physiological function, started insulin expression. To test the hypothesis that
insulin-positive cells can indeed be identified by this approach, we tested an insulin
promoter-driven vector construct expressing the eGFP marker protein (hInsp-eGFP).
Upon testing in various cell lines and human pancreatic islets, our studies confirmed that
the hInsp-eGFP vector construct designed in our laboratory is indeed highly specific for
the insulin-positive cell population.
To test the performance of the vector in a hES cell-based system, we have adapted a
previously published protocol developed to drive the differentiation of hES cells into the
pancreatic beta cell lineage. Of critical significance, our initial studies revealed no
difference in the gene expression profile of the transduced and nontransduced cells
indicating that transduction per se had no effect on the differentiation process.
Interestingly, we consistently found the persistence of stem cell markers in the cultures,
even at later stages of the differentiation when tested by RT-PCR. This observation may

64
indicate the persistence of undifferentiated cells under these culture conditions which
would also explain why these cells promoted the growth of teratomas when transplanted
into immune deficient mice []. Our experiments further revealed that eGFP expression
from the hInsp-eGFP construct was limited to those cells that become insulin-positive as
a result of the differentiation process. We found that approximately 3-5 % of the cells
became positive for C-peptide and insulin by the end of the differentiation if H1 cells
were used. These were more likely to generate insulin-positive cells, when compared to
H9 cells. Approximately 1% of the total cell population transduced with hInps-GFP was
eGFP-positive at the final day of the differentiation protocol. Although this might appear
to be a low number indicating limited success, but the cells that became eGFP positive
were easily detectable under an inverted fluorescent microscope clearly identifying
cultures containing insulin-positive cells.
One of the limitations of the present study is the relatively low titer of the hInsp-eGFP
vector available for our studies (vector titer: 10
8
TU/ml) that necessitated transduction of
the differentiating cells at a later stage of the protocol. Based on previous observations
(Dr. Lutzko, unpublished data), a minimum final vector concentration of 3X10
7
TU/ml is
required to effectively transduce undifferentiated hES cells. Unfortunately, this was not
achievable with the particular vector preparation we used. In addition, only a low
proportion of the cells became eGFP-positive with the hInsp-eGFP vector by the end of
the differentiation process making it challenging, if not impossible, to further characterize
the eGFP-positive cells. Future studies focusing on improving the vector titer and thus the
transduction efficiency will allow utilizing this vector at its full potential.

65
In summary our group has successfully designed a novel vector construct (hInsp-eGFP)
that reliably indicates the appearance of insulin-positive cells derived from hES cells
during an in vitro differentiation process. This approach may provide an invaluable asset
to researchers trying to improve currently available protocols and may also promote the
development of novel methodologies that aim to drive the differentiation of hES cells
into the pancreatic beta cell lineage.

66
CHAPTER IV
Conclusions

Diabetes mellitus represents a global health care problem; it is commonly referred to as
the epidemic of the 20
th
century. According to a recent statistical report, the incidence of
DM is increasing dramatically and continuously worldwide. The US represents no
exception with the number of affected individuals exceeding 23 million (7.8% of the total
population) and is predicted to reach 30 million by 2030.
Approximately 5-10% of the diabetic population suffers from type 1 DM, which is
caused by the autoimmune destruction of insulin-producing beta cells at a relatively early
age. As the beta cell mass of these patients is not enough to maintain glucose homeostasis
critical for survival, continuous insulin-replacement represents the only therapeutic
option currently available at their disposal. Although this approach represented a
milestone in the management of T1D, it still fails to protect individuals from the
secondary complications of the disease (e.g., microvascular, macrovascular changes,
neuropathy). As a consequence, the average life expectancy of an individual with T1D is
approximately 15 years below the average. Thus, it is of crucial importance to find new
treatment options that not only provide temporary glucose control but also improve
long-term outcome. One extensively investigated approach is replacement of the
non-functional beta cell mass with insulin producing cells via pancreatic islet
transplantation. However, currently it is only offered as an experimental therapeutic

67
intervention for those T1D patients who frequently experience life-threatening
hypoglycemic episodes. Short-term data from post-transplantation studies are
encouraging and indicate that insulin-independence can be achieved in selected T1D
patients. Follow-up studies are less promising showing that insulin-independence can
only be maintained for a short period (average: 15 months) owing to the immunological
rejection of the transplanted cells. In addition, limited donor tissue availability poses a
significant challenge. As a consequence, there is a critical interest in finding novel
strategies and developing new approaches to increase allogeneic beta cell supply
available for transplantation.
The present PhD work focused primarily on exploring alternative approaches to
generate insulin-producing beta cells and utilized novel lentiviral vector constructs to
achieve this goal. The first project tested the hypothesis that selective expansion of
human pancreatic beta cells can be achieved ex vivo using a gene transfer-based
approach. By specifically introducing a novel fusion protein, F36Vcmet into pancreatic
beta cells, we have achieved a 4-fold increase in the percentage of proliferating beta cells
on average when cells were co-treated with the fusion protein ligand: AP20187. This
extensive cell proliferation was comparable to that achieved with HGF. An important
difference between our approach and HGF treatment is that HGF is a nonselective
mitogen promoting the proliferation of various cell types while our approach is specific
and only targets the pancreatic beta cell population. In addition, HGF has been shown to
induce beta cell dedifferentiation thus leading to a significant loss of functionality.
Consistent with these studies, our experiments have also shown that HGF treatment

68
impairs the functionality of the expanded cell population significantly. In contrast, our
novel approach only prompts a moderate reduction in functionality. We hypothesize that
this difference may be explained by the positive selection of cells that retained their
functionality when using our approach.
The limitations of the present study include the limited number of human tissue
available for the experiments. In addition, the use of lentiviral vectors to deliver the
fusion protein may raise safety concerns. Future animal studies will provide ultimate
evidence whether the controlled proliferation of pancreatic beta cells may be achieved
safely in vivo by utilizing our novel, lentivirus-based approach.
The second project of the present thesis work focused on the differentiation of human
embryonic stem cells into the pancreatic beta cell lineage, as an alternative to beta cell
expansion. Recognizing the critical need to improve current differentiation protocols, our
group designed a lentiviral vector construct (hInsp-eGFP) with the ability to consistently
and specifically identify insulin-positive cells derived from hES cells. In our studies, we
have shown that 1) the vector is expressed specifically in insulin-positive cell
populations; 2) transduction of the differentiating hES cells does not interfere with the
differentiation process; 3) eGFP-expression from the vector reliably indicates the
appearance of insulin-positive cells in the culture. If successfully adapted to a 96-well
format, this technique has the potential to be used as a high-throughput screening assay
with the aim to optimize current and future differentiation protocols.

69
In summary, our approaches represent new strategies aiming to generate insulin-positive
cells suitable for the transplantation therapy of DM. In addition, they represent an
invaluable asset for the fields of gene therapy and diabetes research.

70
BIBLIOGRAPHY

1. Assady S, Maor G, Amit M, et al. Insulin production by human embryonic stem
cells. Diabetes 2001;50:1691-1697.
2. Aubert J, Stavridis MP, Tweedie S, et al. Screening for mammalian neural genes via
fluorescence-activated cell sorter purification of neural precursors from Sox1-gfp
knock-in mice. Proc Natl Acad Sci USA 2003;100(Suppl 1):11836-11841.
3. Baeyens L, De Breuck S, Lardon J, et al. In vitro generation of insulin-producing
beta cells from adult exocrine pancreatic cells. Diabetologia 2005;48:49-57.
4. Beattie GM, Cirulli V, Lopez AD, et al. Ex vivo expansion of human pancreatic
endocrine cells. J Clin Endocrinol Metab 1997;82:1852-1856.
5. Beattie GM, Itkin-Ansari P, Cirulli V, et al. Sustained proliferation of PDX-1+ cells
derived from human islets. Diabetes 1999;48:1013-1019.
6. Beattie GM, Montgomery AM, Lopez AD, et al. A novel approach to increase
human islet cell mass while preserving beta-cell function. Diabetes
2002;51:3435-3439.
7. Beattie GM, Rubin JS, Mally MI, et al. Regulation of proliferation and
differentiation of human fetal pancreatic islet cells by extracellular matrix,
hepatocyte growth factor, and cell-cell contact. Diabetes 1996;45:1223-1228.
8. Blau CA, Peterson KR, Drachman JG, et al. A proliferation switch for genetically
modified cells. Proc Natl Acad Sci USA 1997;94:3076-3081.
9. Bonner-Weir S, Taneja M, Weir GC, et al. In vitro cultivation of human islets from
expanded ductal tissue. Proc Natl Acad Sci USA 2000;97:7999-8004.
10. Bonner-Weir S, Toschi E, Inada A, et al. The pancreatic ductal epithelium serves as
a potential pool of progenitor cells. Pediatr Diabetes 2004;5(Suppl 2):16-22.
11. Bonner-Weir S, Baxter LA, Schuppin GT, et al. A second pathway for regeneration
of adult exocrine and endocrine pancreas: A possible recapitulation of embryonic
development. Diabetes 1993;42:1715-1720.
12. Bottaro DP, Rubin JS, Faletto DL, et al. Identification of the hepatocyte growth
factor receptor as the c-met proto-oncogene product. Science 1991;251:802-804.
13. Brons IG, Smithers LE, Trotter MW, et al. Derivation of pluripotent epiblast stem
cells from mammalian embryos. Nature 2007;448:191-195.

71
14. Chiao E, Elazar M, Xing Y, et al. Isolation and transcriptional profiling of purified
hepatic cells derived from human embryonic stem cells. Stem Cells
2008;26:2032-2041.
15. Clackson T, Yang W, Rozamus L, et al. Redesigning an FKBP-ligand interface to
generate chemical dimerizers with novel specificity. Proc Natl Acad Sci USA
1998;95:10437-10442.
16. Cooper CS, Park M, Blair DG, et al. Molecular cloning of a new transforming gene
from a chemically transformed human cell line. Nature 1984;311:29-33.
17. Cornelius JG, Tchernev V, Kao KJ, et al. In vitro-generation of islets in long-term
cultures of pluripotent stem cells from adult mouse pancreas. Horm Metab Res
1997;29:271-277.
18. D’Amour KA, Agulnick AD, Eliazer S, et al. Efficient differentiation of human
embryonic stem cells to definitive endoderm. Nat Biotechnol 2005;23:1534-1541.
19. D’Amour KA, Bang AG, Eliazer S, et al. Production of pancreatic hormone-
expressing endocrine cells from human embryonic stem cells. Nat Biotechnol
2006;24:1392-1401.
20. Daadi MM, Steinberg GK. Manufacturing neurons from human embryonic stem
cells: biological and regulatory aspects to develop a safe cellular product for stroke
cell therapy. Regen Med 2009;4:251-263.
21. Dor Y, Brown J, Martinez OI, et al. Adult pancreatic β-cells are formed by
self-duplication rather than stem-cell differentiation. Nature 2004;429:41-46.
22. Drucker DJ. Glucagon-like peptides: Regulators of cell proliferation, differentiation,
and apoptosis. Mol Endocrinol 2003;17:161-171.
23. Egan JM, Bulotta A, Hui H, et al. GLP-1 receptor agonists are growth and
differentiation factors for pancreatic islet beta cells. Diabetes Metab Res Rev
2003;19:115-123.
24. Elliott RB, Escobar L, Tan PL, et al. Live encapsulated porcine islets from a type 1
diabetic patient 9.5 yr after xenotransplantation. Xenotransplantation
2007;14:157-161.
25. Fajas L, Annicotte JS, Miard S, et al. Impaired pancreatic growth, beta cell mass,
and beta cell function in E2F1 ( −/ −) mice. J Clin Invest 2004;113:1288-1295.

72
26. Fiaschi-Taesch NM, Berman DM, Sicari BM, et al. Hepatocyte growth factor
enhances engraftment and function of nonhuman primate islets. Diabetes
2008;57:2745-2754.
27. Gao R, Ustinov J, Pulkkinen M, et al. Characterization of endocrine progenitor cells
and critical factors for their differentiation in human adult pancreatic cell culture.
Diabetes 2003;52:2007-2015.
28. Garcia-Ocana A, Takane KK, Syed MA, et al. Hepatocyte growth factor over-
expression in the islet of transgenic mice increases beta cell proliferation, enhances
islet mass, and induces mild hypoglycemia. J Biol Chem 2000;275:1226-1232.
29. Georgia S, Bhushan A. Beta cell replication is the primary mechanism for
maintaining postnatal beta cell mass. J Clin Invest 2004;114:963-968.
30. Gherardi E, Youles ME, Miguel RN, et al. Functional map and domain structure of
the MET (HGF/SF) receptor. Proc Natl Acad Sci USA 1993;100:12039-12044.
31. Grose S. Critics slam Russian trial to test pig pancreas for diabetics. Nat Med
2007;13:390-391.
32. Hacein-Bey-Abina S, von Kalle C, Schmidt M, et al. A serious adverse event after
successful gene therapy for X-linked severe combined immunodeficiency. N Engl J
Med 2003;348:255-256.
33. Hacein-Bey-Abina S, von Kalle C, Schmidt M, et al. LMO2-associated clonal T cell
proliferation in two patients after gene therapy for SCID-X1. Science
2003;302:415-419.
34. Halvorsen TL, Beattie GM, Lopez AD, et al. Accelerated telomere shortening and
senescence in human pancreatic islet cells stimulated to divide in vitro. J Endocrinol
2000;66:103-109.
35. Harvey M, Vogel H, Lee E, et al. Mice deficient in both p53 and Rb develop tumors
primarily of endocrine origin. Cancer Research 1995;55:1146-1151.
36. Hirshberg B, Rother KI, Digon BJ, et al. Benefits and risks of solitary islet
transplantation for type 1 diabetes using steroid-sparing immunosuppression.
Diabetes Care 2003;26:3288-3295.
37. Hotta A, Cheung AY, Farra N, et al. Isolation of human iPS cells using EOS
lentiviral vectors to select for pluripotency. Nat Methods 2009;6:370-376.

73
38. Huang C, Snider F, Cross JC. Prolactin receptor is required for normal glucose
homeostasis and modulation of beta cell mass during pregnancy. Endocrinology
2009;150:1618-1626.
39. Ianus A, Holz GG, Theise ND, et al. In vivo derivation of glucose-competent
pancreatic endocrine cells from bone marrow without evidence of cell fusion. J Clin
Invest 2003;111:843-850.
40. Iglesias A, Murga M, Laresgoiti U, et al. Diabetes and exocrine pancreatic
insufficiency in E2F1/E2F2 double-mutant mice. J Clin Invest 2004;113:1398-1407.
41. Jiang J, Au M, Lu K, et al. Generation of insulin-producing islet-like clusters from
human embryonic stem cells. Stem Cells 2007;25:1940-1953.
42. Katdare MR, Bhonde RR, Parab PB. Analysis of morphological and functional
maturation of neoislets generated in vitro from pancreatic ductal cells and their
suitability for islet banking and transplantation. J Endocrinol 2004;182:105-112.
43. Kawasaki H, Mizuguchi T, Kikkawa Y, et al. In vitro transformation of adult rat
hepatic progenitor cells into pancreatic endocrine hormone-producing cells.
Hepatobiliary Pancreat Surg 2008;15:310-317.
44. Kobinger GP, Deng S, Louboutin JP, et al. Pharmacologically regulated
regeneration of functional human pancreatic islets. Mol Ther 2005;11:105-111.
45. Korsgren O, Nilsson B, Berne C, et al. Current status of clinical islet transplantation.
Transplantation 2005;79:1289-1293.
46. Kroon E, Martinson LA, Kadoya K, et al. Pancreatic endoderm derived from human
embryonic stem cells generates glucose-responsive insulin-secreting cells in vivo.
Nat Biotechnol 2008;26:443-452.
47. Lechner A, Habener JF. Stem/progenitor cells derived from adult tissues: Potential
for the treatment of diabetes mellitus. Am J Physiol Endocrinol Metab
2003;284:E259-266.
48. Lechner A, Yang YG, Blacken RA, et al. No evidence for significant
transdifferentiation of bone marrow into pancreatic beta-cells in vivo. Diabetes
2004;53:616-623.
49. Lefebvre VH, Otonkoski T, Ustinov J, et al. Culture of adult human islet
preparations with hepatocyte growth factor and 804G matrix is mitogenic for duct
cells but not for beta-cells. Diabetes 1998;47:134-137.

74
50. Levine F, Itkin-Ansari P. Beta-cell Regeneration: Neogenesis, replication or both? J
Mol Med 2008;86:247-258.
51. Li ZY, Otto K, Richard RE, et al. Dimerizer-induced proliferation of genetically
modified hepatocytes. Mol Ther 2002;5:420-426.
52. Lipsett M, Aikin R, Castellarin M, et al. Islet neogenesis: A potential therapeutic
tool in type 1 diabetes. Int J Biochem Cell Biol 2006;38:498-503.
53. Lohr M, Kloppel G. Residual insulin positivity and pancreatic atrophy in relation to
duration of chronic type 1 (insulin-dependent) diabetes mellitus and
microangiopathy. Diabetologia 1987;30:757-762.
54. Long-term Trends in Diabetes. National Center for Health Statistics, Centers for
Disease Control and Prevention, Division of Diabetes Translation. May, 2008.
www.cdc.gov/diabetes/statistics
55. Lumelsky N, Blondel O, Laeng P, et al. Differentiation of embryonic stem cells to
insulin-secreting structures similar to pancreatic islets. Science 2001;292:1389-1394.
56. Meier JJ, Bhushan A, Butler AE, et al. Sustained beta cell apoptosis in patients with
long-standing type 1 diabetes: Indirect evidence for islet regeneration? Diabetologia
2005;48:2221-2228.
57. Meissner A, Wernig M, Jaenisch R. Direct reprogramming of genetically
unmodified fibroblasts into pluripotent stem cells. Nat Biotechnol
2007;25:1177-1181.
58. Menge BA, Tannapfel A, Belyaev O, et al. Partial pancreatectomy in adult humans
does not provoke beta-cell regeneration. Diabetes 2008;57:142-149.
59. Miyoshi H, Blomer U, Takahashi M, et al. Development of a self-inactivating
lentivirus vector. J Virol 1998;72:8150-8157.
60. Mountford JC. Human embryonic stem cells: origins, characteristics and potential
for regenerative therapy. Transfus Med 2008;18:1-12.
61. Myken PS, Bech-Hansen O. A 20-year experience of 1712 patients with the Biocor
porcine bioprosthesis. J Thorac Cardiovasc Surg 2009;137:76-81.
62. Navarro-Alvarez N, Soto-Gutierrez A, Kobayashi N. Stem cell research and therapy
for liver disease. Curr Stem Cell Res Ther 2009;4:141-146.
63. Okita K, Ichisaka T, Yamanaka S. Generation of germline-competent induced
pluripotent stem cells. Nature 2007;448:313-317.

75
64. Okuno M, Minami K, Okumachi A, et al. Generation of insulin-secreting cells from
pancreatic acinar cells of animal models of type 1 diabetes. Am J Physiol Endocrinol
Metab 2007;292:E158-165.
65. Onaca N, Klintmalm GB, Levy MF. Pancreatic islet cell transplantation: a treatment
strategy for type I diabetes mellitus. Nutr Clin Pract 2004;19:154-164.
66. Otonkoski T, Beattie GM, Rubin JS, et al. Hepatocyte growth factor / scatter factor
has insulinotropic activity in human fetal pancreatic cells. Diabetes
1994;43:947-953.
67. Otonkoski T, Cirulli V, Beattie M, et al. A role for hepatocyte growth factor/scatter
factor in fetal mesenchyme-induced pancreatic beta cell growth. Endocrinology
1996;137:3131-3139.
68. Piper K, Ball SG, Turnpenny LW, et al. Beta-cell differentiation during human
development does not rely on nestin-positive precursors: implications for stem
cell-derived replacement therapy. Diabetologia 2002;45:1045-1047.
69. Pittenger GL, Taylor-Fishwick D, Vinik AI. The role of islet neogenesis-associated
protein (INGAP) in pancreatic islet neogenesis. Curr Protein Pept Sci
2009;101:37-45.
70. Pomper MG, Hammond H, Yu X, et al. Serial imaging of human embryonic
stem-cell engraftment and teratoma formation in live mouse models. Cell Res
2009;19:370-379.
71. Ponzetto C, Bardelli A, Zhen Z, et al. A multifunctional docking site mediates
signaling and transformation by the hepatocyte growth factor/scatter factor receptor
family. Cell 1994;77:261-271.
72. Porat S, Dor Y. New sources of pancreatic bets cells. Curr Diab Rep
2007;7:304-308.
73. Ramiya VK, Maraist M, Arfors KE, et al. Reversal of insulin-dependent diabetes
using islets generated in vitro from pancreatic stem cells. Nat Med 2000;6:278-282.
74. Rane SG, Reddy EP. Cell cycle control of pancreatic beta cell proliferation. Front
Biosci 2000;5:D1-19.
75. Redondo MJ, Fain PR, Eisenbarth GS. Genetics of type 1A diabetes. Recent Prog
Horm Res 2001;56:69-89.

76
76. Rooman I, Heremans Y, Heimberg H, et al. Modulation of rat pancreatic
acinoductal transdifferentiation and expression of PDX-1 in vitro. Diabetologia
2000;43: 907-914.
77. Ryan EA, Lakey JR, Paty BW, et al. Successful islet transplantation: Continued
insulin reserve provides long-term glycemic control. Diabetes 2002;51:2148-2157.
78. Ryan EA, Paty BW, Senior PA, et al. Five-year follow-up after clinical islet
transplantation. Diabetes 2005;54:2060-2069.
79. Saisho Y, Butler AE, Meier JJ, et al. Pancreas volumes in humans from birth to age
one hundred taking into account sex, obesity, and presence of type-2 diabetes. Clin
Anat 2007;20:933-42.
80. Seaberg RM, Smukler SR, Kieffer TJ, et al. Clonal identification of multipotent
precursors from adult mouse pancreas that generate neural and pancreatic lineages.
Nat Biotechnol 2004;22:1115-1124.
81. Seeberger KL, Eshpeter A, Rajotte RV, et al. Epithelial cells within the human
pancreas do not coexpress mesenchymal antigens: epithelial-mesenchymal transition
is an artifact of cell culture. Laboratory Investigation 2009;89:110-121.
82. Shapiro AM, Lakey JR, Ryan EA, et al. Islet transplantation in seven patients with
type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N
Engl J Med 2000;343:230-238.
83. Sharma R, Raghubir R. Stem cell therapy: a hope for dying hearts. Stem Cells Dev
2007;16:517-536.
84. Shaw KL, Pais E, Ge S, et al. Lentiviral vectors with amplified beta cell-specific
gene expression. Gene Ther 2009 (Epub).
85. Sieradzki, H. Fleck H, Chatterjee AK, Schatz H. Stimulatory effect of insulin-like
growth factor-I on [3H]thymidine incorporation, DNA content and insulin
biosynthesis and secretion of isolated pancreatic rat islets. Journal of Endocrinology
1988; 117:59-62.
86. Song KH, Ko SH, Ahn YB, et al. In vitro transdifferentiation of adult pancreatic
acinar cells into insulin-expressing cells. Biochem Biophys Res Commun
2004;316:1094-1100.
87. Sordi V. Mesenchymal stem cell homing capacity. Transplantation 2009;87:S42-S45.

77
88. Sorenson RL, Brelje TC. Adaptation of islets of Langerhans to pregnancy: beta-cell
growth, enhanced insulin secretion and the role of lactogenic hormones. Horm
Metab Res 1997;29:301-307.
89. Srikanta S, Ganda OP, Jackson RA, Type I diabetes mellitus in monozygotic twins:
Chronic progressive beta cell dysfunction. Ann Intern Med 1983;99:320-326.
90. Stoker M, Gherardi E, Perryman M, et al. Scatter factor is a fibroblast-derived
modulator of epithelial cell mobility. Nature 1987;327:239-242.
91. Svendsen CN, Langston JW. Stem cells for Parkinson disease and ALS: replacement
or protection? Nat Med 2004;10:224-225.
92. Swenne D, Hill DJ, Strain AJ, Milner RD. Growth hormone regulation of
somatomedin C/insulin-like growth factor I production and DNA replication in fetal
rat islets in tissue culture, Diabetes 1987; 36: 288-294.
93. Sykes M, Cozzi E. Xenotransplantation of pig islets into Mexican children: Were the
fundamental ethical requirements to proceed with such a study really met? Eur J
Endocrinol 2006;154:921-922.
94. Takahashi K, Tanabe K, Ohnuki M, et al. Induction of pluripotent stem cells from
adult human fibroblasts by defined factors. Cell 2007;131:861-872.
95. Terada N, Hamazaki T, Oka M, et al. Bone marrow cells adopt the phenotype of
other cells by spontaneous cell fusion. Nature 2002;416:542-545.
96. The DIAMOND Project Group, Incidence and trends of childhood Type 1 diabetes
worldwide 1990–1999. Diabet. Med. (2006), 23: 857–866.
97. Thomson JA, Itskovitz-Eldor J, Shapiro SS, et al. Embryonic stem cell lines derived
from human blastocysts. Science 1998;282:1145-1147.
98. Tulpule A, Daley GQ. Efficient gene knockdowns in human embryonic stem cells
using lentiviral-based RNAi. Methods Mol Biol 2009;482:35-42.
99. Tweedell KS. New paths to pluripotent stem cells. Curr Stem Cell Res Ther
2008;3:151-162.
100. Valdes-Gonzalez RA, White DJ, Dorantes LM, et al. Three-yr follow-up of a type 1
diabetes mellitus patient with an islet xenotransplant. Clin Transplant
2007;21:352-357.
101. Vija L, Farge D, Gautier JF, et al. Mesenchymal stem cells: Stem cell therapy
perspectives for type 1 diabetes. Diabetes Metab 2009;35:85-93.

78
102. Vinik A, Pittenger G, Rafaeloff R, et al. Determinants of pancreatic islet cell mass:
A balance between neogenesis and senescence/apoptosis. Diab Rev 1996;4:235-263.
103. Wang S, Roy NS, Benraiss A, et al. Promoter-based isolation and fluorescence-
activated sorting of mitotic neuronal progenitor cells from the adult mammalian
ependymal/subependymal zone. Dev Neurosci 2000;22:167-176.
104. Xu C, Lebkowski J, Gold JD. Growth and differentiation of human embryonic stem
cells for cardiac cell replacement therapy. Curr Stem Cell Res Ther 2006;1:173-187.
105. Yang L, Li S, Hatch H, et al. In vitro trans-differentiation of adult hepatic stem cells
into pancreatic endocrine hormone-producing cells. Proc Natl Acad Sci USA
2002;99:8078-8083.
106. Yatoh S, Akashi T, Chan PP, et al. NeuroD and reaggregation induce beta-cell
specific gene expression in cultured hepatocytes. Diabetes Metab Res Rev
2007;23:239-249.
107. Ye Z, Yu X, Cheng L. Lentiviral gene transduction of mouse and human stem cells.
Methods Mol Biol 2008;430:243-253.
108. Ying QL, Nichols J, Evans EP, et al. Changing potency by spontaneous fusion.
Nature 2002;416:545-548.
109. Yu J, Vodyanik MA, Smuga-Otto K, et al. Induced pluripotent stem cell lines
derived from human somatic cells. Science 2007;318:1917-1920.
110. Zaret KS, Grompe M. Generation and regeneration of cells of the liver and pancreas.
Science 2008;322:1490-1494.
111. Zarnegar R, Michalopoulos G. Purification and biological characterization of human
hepatopoietin A, a polypeptide growth factor for hepatocytes. Cancer Res
1989;49:3314-3320.
112. Zhou Q, Brown J, Kanarek A, et al. In vivo reprogramming of adult exocrine cells to
beta-cells. Nature 2008;455:627-632.
113. Zufferey R, Dull T, Mandel RJ, et al. Self-inactivating lentivirus vector for safe and
efficient in vivo gene delivery. J Virol 1998;72:9873-9880.
114. Zuhdi N, Hancock W, Hawley W, et al. Porcine aortic valves for replacement of
human heart valves. J Cardiovasc Surg (Torino) 1973;Spec No:425-429.

79
115. Zulewski H, Abraham EJ, Gerlach MJ, et al. Multipotential nestin-positive stem
cells isolated from adult pancreatic islets differentiate ex vivo into pancreatic
endocrine, exocrine, and hepatic phenotypes. Diabetes 2001;50:521-533.

Abstract (if available)

Abstract The incidence of Diabetes Mellitus (DM) continues to increase with approximately 7.8% of the population affected in the US alone. Although pancreatic beta cell transplantation has the potential to cure the disease, limited donor tissue availability poses a major challenge. Aiming to overcome this critical shortage, we utilized two distinct, lentiviral vector-based approaches to generate or expand insulin-positive cells ex vivo.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Investigating the role of STAT3 in mouse and rat embryonic stem cell self-renewal and differentiation

PDF

Non-viral and viral hematopoietic progenitor cell gene therapy

PDF

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

PDF

Role of beta-catenin in mouse epiblast stem cell, embryonic stem cell self-renewal and differentiation

PDF

A novel construct to study the pulsatility of insulin secretion in single cells, islets and whole pancreas

PDF

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

PDF

Derivation and characterization of human embryonic stem (hES) cells and human induced pluripotent stem (hiPS) cells in clinical grade conditions

PDF

Modeling biochemical components and systems through data integration and digital arts approaches

PDF

Mechanisms that dictate beta cells’ response to stress in the context of genetic mutation, pregnancy, and infection

PDF

Pten regulates beta-cell regeneration intrinsically and independently of development

PDF

Epigenetic plasticity of cultured female human embryonic stem cells and regulation of gene expression and chromatin by PR-SET7 mediated H4K20me1

PDF

Role of cancer-associated fibroblast secreted annexin A1 in generation and maintenance of prostate cancer stem cells

PDF

A functional genomic approach based on shRNA-mediated gene silencing to delineate the role of NF-κB and cell death proteins in the survival and proliferation of KSHV associated primary effusion l...

Asset Metadata

Creator Pais, Eszter (author)

Core Title Stem cell and gene transfer-based approaches to generate insulin-producing cells

Contributor Electronically uploaded by the author (provenance)

School Keck School of Medicine

Degree Doctor of Philosophy

Degree Program Systems Biology

Degree Conferral Date 2009-08

Publication Date 07/18/2009

Defense Date 06/10/2009

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

Tag Diabetes,differentiation,human embryonic stem cell,human pancreatic beta cell,lentiviral vector,OAI-PMH Harvest

Language English

Advisor Lutzko, Carolyn (committee chair), Chow, Robert HP. (committee member), Kohn, Donald B. (committee member), Meiselman, Herbert J. (committee member), Pera, Martin F. (committee member)

Creator Email epais@usc.edu,eszterpais@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2374

Unique identifier UC1188933

Identifier etd-Pais-3128 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-405795 (legacy record id),usctheses-m2374 (legacy record id)

Legacy Identifier etd-Pais-3128.pdf

Dmrecord 405795

Document Type Dissertation

Rights Pais, Eszter

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