The derivation, propagation, storage and gene expression of human embryonic stem cells on human feeders

GENERAL INTRODUCTION
Several types of stem cells have been discovered from germ cells, the embryo, fetus and
adult. Each of these has promised to revolutionize the future of regenerative medicine
through the provision of cell replacement therapies to treat a variety of debilitating
diseases. The tremendous versatility of embryonic stem cells versus the unprecedented
reports describing adult stem cell plasticity have ignited debates as to the choice of one
cell type over the other for future applications. However, the biology of these mysterious
cells have yet to be understood through a lot more basic research before new therapies
using stem cell differentiated derivatives can be applied.
Everyday, we read and listen to news reports about how stem cells promise to
revolutionize medicine and change our lives with panaceas for every imaginable disease
including rhetoric that stem cell therapy will some day delay the process of ageing.
Embroiled in the hype and media frenzy are also political agendas, numerous religious
and genuine ethical concerns. To further fuel the debate, embryonic stem cell research is
often unjustly associated with reproductive cloning. Stem cell research is politically
charged, receives considerable media coverage, raises many ethical and religious debates
and generates a great deal of public interest.
Stem cell research also opens the new field of ‘cell based therapies’ and as such
several safety measures have also to be evaluated. The hope that someday many
debilitating human diseases may be treated with stem cell therapy is inspired by
remarkable examples of whole organ and limb regeneration in animals as well as the
historical success of bone marrow transplants that have improved the lives of many
patients suffering from leukemia, immunological and other blood disorders. Clearly,
stem cell research leading to prospective therapies in reparative medicine has the
potential to affect the lives of millions of people around the world for the better and there
1


is good reason to be optimistic. However, the road towards the development of an
effective cell-based therapy for widespread use is long and involves overcoming
numerous technical, legislative, ethical and safety issues.

Embryos of most mammals are comprised of a special group of cells that have
the potential to give rise to all the tissues and organs of the fetus and future adult. This
group of cells called the inner cell mass (ICM) cells evolves into embryonic stem cells
(ESCs) in vitro. Unlike other cell types that can only divide a maximum of 50 times or so
in tissue culture dishes (Hayflick & Moorhead 1961), ESCs can divide indefinitely
without losing their ability to form different cell types. Human embryonic stem cells
(hESCs) derived from isolation and serial sub-culture of ICMs from 5-day old human
blastocysts hold the promise of revolutionizing the future of medicine by the creation of
early developmental models for a multitude of human genetic diseases and through the
development of cell and tissue replacement therapies. Immense commercial interest as
well as ethical controversy surrounds hESC research.
Several improvements in blastocyst culture techniques were a prerequisite for
culturing and harvesting good quality blastocysts with large ICMs. These breakthroughs
not only led to increased pregnancy rates with blastocyst transfer in patients undergoing
in vitro fertlilization (IVF) cycles but also enabled scientists to derive hESC cell lines
from human blastocysts. hESCs cells were first isolated in 1994 (Bongso et al 1994)
while the first continuous immortal hESC lines were established only in 1998 (Thomson
et al 1998).
hESCs are colony forming social cells that are unspecialized. This means that if
they are coaxed properly, ESCs have in theory the ability to turn into any of the cell
types in the human body. In contrast, adult stem cells, which are found in adult tissues
and organs, have the ability to transform into only a limited variety of cell types. Adult
2


stem cells are also difficult to isolate and very challenging to grow in culture. This
coupled with their restricted developmental potential are the main reasons why many
scientists believe that embryonic stem cells are more promising and better alternatives
for developing a wider range of cell based therapies. In order for hESCs to retain their
ability to form different cell types, they need to be grown on feeder cell supports. The

feeder layer produces growth factors and extracellular matrix components that may help
to keep the hESCs from differentiating into other specialized cells. Without the support
of a feeder layer, hESCs spontaneously and uncontrollably differentiate into a milieu of
mixed cell types. This is often undesirable for the researcher as specific cell types are
often difficult or near impossible to isolate from this mixed milieu.
Embryonic stem cells are a unique class of cell type for various reasons. Most
significantly, they can undergo self-renewal for extended periods of in vitro cultivation,
have the ability to form teratomas when injected into severely combined
immunodeficient (SCID) mice and can differentiate into a variety of cell types from all 3
primitive germ layers in vitro and in vivo, thus distinguishing them from other adult stem
cells.
hESCs differ in many ways from mouse embryonic stem cells (mESCs). Several
lines of evidence suggest that hESCs and mESCs do not represent equivalent embryonic
cell types. In vitro differentiation of hESCs leads to the expression of AFP and HCG,
which are typically produced by trophoblast cells in the developing human embryo,
while mESCs are generally believed not to differentiate along this lineage. In addition,
hESCs express SSEA-3, SSEA-4, TRA-1-60 and TRA-1-81 surface antigens prior to
differentiation but only SSEA-1 upon differentiation, while mESCs only express SSEA1 prior to differentiation (Thomson et al 1998, Reubinoff et al 2000, Henderson et al
2002). The cytokine leukemia inhibitory factor (LIF) has an established facultative role
3


in keeping mESCs undifferentiated and an exogenous supply of LIF in the culture
medium is sufficient to keep mESCs undifferentiated for prolonged culture periods
(Williams et al 1988, Smith et al 1988). hESCs on the other hand do not appear to have a
perceivable LIF response (Thomson et al 1998, Reubinoff et al 2000). The molecule or
group of molecules involved in autocrine or paracrine signaling in keeping hESCs
undifferentiated has also not been identified making the culture of undifferentiated
hESCs heavily reliant on feeder layer support.
It has been over five years since the first hESC lines were established but our

understanding of hESC biology is still limited for it to be exploited for clinical
application. This hopefully would change once hESC lines become widely and routinely
available to all researchers. Nevertheless, hESC research needs to be pursued
aggressively if we are to quickly realize the full therapeutic potential of reparative cell
therapy and several areas in particular warrant immediate attention.
Specifically, advances must be made to improve hESC culture techniques. Purer
and safer populations of functionally normal undifferentiated hESCs and differentiated
hESC progenitor cell types need to be derived. All current 78 NIH listed hESC lines
approved for US government federal research funding have been derived and propagated
on mouse embryonic fibroblast (MEFs) and in the presence of culture medium
containing animal based ingredients. The use of a feeder layer of animal origin and
animal components in the culture media substantially elevates the risk of the crosstransfer of viruses and other pathogens to the hESCs.
Many studies have focused on the differentiation of MEF supported hESCs into a
range of clinically useful cell types, while this is important, the development and
refinement of a xeno-free culture system that decreases the risk of hESC contamination
with adventitious agents while maintaining pure undifferentiated hESC populations
4


amenable to expansion of cell numbers is critical before any clinical exploitation of
hESC technology can occur. New hESC lines need to be derived and bulk-cultured in
current good manufacture practice (cGMP) conditions according to a xeno-free gold
standard.
The establishment of new hESC lines in cGMP conditions necessitates the
development of an effective cryopreservation protocol that minimizes or restricts the
possibility of early passage hESC seed stock contamination with adventitious agents
such as viruses and other pathogens during long-term liquid nitrogen storage.
Interestingly, hESC lines have heterogeneous genetic backgrounds unlike mESC
lines that are from inbred mouse strains and appear to behave differently in culture. For
example, not all hESC lines are amenable to bulk and feeder-free culture protocols,

doubling times differ considerably between different lines and the degree of spontaneous
differentiation in vitro also appears to show much variation (Vogel 2002).
Functional genomics data need to be gathered for a better understanding of the
genetic pathways that regulate pluripotency, self-renewal and differentiation. Identifying
“master regulators” and as yet undiscovered genes that control hESC self-renewal and
immortality will shed light on cancer genetics as well as have implications in ageing
research. In an effort to better understand the molecular cascades controlling the
pluripotent phenotype in hESCs, transcriptional profiling using, microarray, serial
analysis of gene expression (SAGE) and massively parallel signature sequencing
(MPSS) technology are being undertaken on undifferentiated and differentiated hESCs
(Sato et al 2003, Sperger et al 2003, Brandenberger et al 2004, Ginis et al 2004, Rao
2004a). Comparative analysis of the transcriptome profiles using these techniques will
reveal several interesting candidate genes that are potentially important to the embryonic
stem cell phenotype.
5


In principle, hESCs are capable of differentiating into all cell types in the adult
human, therefore they have the potential to provide a source of tissues for replacement in
diseases in which native cell types are inactivated or destroyed. However, the wideranging diverse nature of human diseases and technical shortcomings will in most cases
limit the promise of hESC replacement therapy to a few common human diseases. In
contrast, the value of hESCs as a developmental model in helping us gain insight to
virtually all human diseases with a genetic basis appears limitless. hESCs also hold
promise in screening and toxicity testing assays in the pharmaceutical industry.
Established hESC lines provide a convenient tool for investigating cell
differentiation in a way that is pertinent to human embryonic development, providing
insights into the causes of birth defects and diseases such as cancer that involve aberrant
cell proliferation and differentiation. Perhaps even more powerful than generating
healthy tissues from existing hESC lines for cell replacement therapy, would be the
ability to generate diseased hESC lines with genetic defects through somatic cell nuclear

transfer (SCNT). These diseased and genetically abnormal hESC lines could produce an
unlimited quantity of diseased cell types that will be an invaluable resource and model to
study the disease phenotype and its genetic basis. Several technical hurdles need to be
bypassed before the enormous implications of ES cell technology for understanding and
curing human diseases can be realized.
hESC research opens up new vistas in the fields of medicine. For instance,
treating hESCs with the right combination of growth factors may induce the formation of
dopaminergic neurons or perhaps insulin secreting beta cells of the pancreas. The
specialized neurons and beta cells derived from precursor hESCs can then be returned to
patients suffering from Parkinson’s disease or diabetes to correct the defects of the
malfunctioning organ or tissue. However, cellular therapy using hESC derived
6


specialized cell types will very likely be useful for treating only certain human diseases.
Diseases that involve and affect multiple cell types and organs may not be treatable using
a cell based therapeutic approach. Thus, Parkinson’s disease and Type I diabetes are
often singled-out as the two most promising targets for a cell based therapeutic approach.
Such treatments from stem cell research will be

“cell-based therapies”.

Presently, doctors administer fluids (injections), solids (pills) or surgical intervention.
For the first time, treatment may be by the administration of cells directly into the body.
Thus, several added precautions have to be taken before cell-based therapeutic products
can be released into the market. Firstly, stringent tests have to be conducted to ensure
that the specialized cell types, which are returned to the patient, are totally pure. No
contaminating undifferentiated hESCs should be present because they have the potential
to divide and replicate and produce a tumor if an undetected renegade hESC is
accidentally injected into the patient. Secondly, specialized cell types derived from

hESCs must be rigorously tested in vitro and in animal models in vivo to show that they
can restore normal physiological function in disease models. Thirdly, several hurdles in
the manipulation and differentiation of hESCs must be overcome before the technology
can be successfully transferred to the bedside. Cell replacement therapies require the
growth of large numbers of hESCs. Thus, large-scale hESC culture strategies using
bioreactors need to be developed to generate sufficient numbers of cells. High efficiency
directed differentiation strategies via spontaneous, co-culture or genomics approaches,
safer and purer populations of hESCs and their differentiated progeny, clinically
compliant xeno-free hESC lines and xeno-free storage systems are very urgent areas that
need investigation.
The studies in this thesis address some of these urgent issues, more specifically
the derivation and propagation of xeno-free hESC lines, the xeno-free storage of hESCs
7


and the understanding of the molecular genetics of hESCs that can help identify genes
involved in the maintenance of pluripotency and commitment to differentiation events.

8


LITERATURE REVIEW
Regeneration in invertebrates and vertebrates
Man has long been fascinated by the regenerative abilities of certain animals.
Regeneration is a remarkable physiological process in which remaining tissues organize
to reform a missing body part. All species possess the ability to regenerate damaged
tissues, the degree of regeneration, however, varies considerably among species. Such
differences in regenerative capacity are perhaps indicative of specific mechanisms that
control the different types of regeneration.
Several invertebrates like the Planarian flatworm and the Hydra regenerate

tissues with speed and precision. Planarians are spectacular examples of whole body
regeneration by an invertebrate; a planarian sliced into 50 pieces will regenerate 50 new
planarians from each piece.
The majority of higher vertebrates are incapable of any form of whole organ
regeneration, even though they had all the necessary instructions and machinery to
generate the tissue during embryonic development (Wolpert et al 1971, Brockes 1997).
Of the higher vertebrates, mammals appear to have limited regenerative ability, a tradeoff perhaps for more proficient wound healing ability. The most striking examples of
whole organ regeneration in mammals are that of antler regeneration in Elks, and in
humans, liver regeneration after partial hepatectomy (Kiessling & Anderson 2003).
Most tissue repair events in mammals are dedifferentiation independent events
resulting from the activation of pre-existing stem cells or progenitor cells. In contrast,
some vertebrates like the salamanders regenerate lost body parts through the
dedifferentiation of specialized cells into new precursor cells. These dedifferentiated
cells then proliferate and later form new specialized cells of the regenerated organ. Stem
cells or progenitor cells are the common denominator for nearly all types of
9


regeneration. They are either already pre-existing, as in the case for mammals or created
by the process of dedifferentiation.
The process of retina and limb regeneration in urodele amphibians involves
complex dedifferentiation and redifferentiation events. Following limb amputation, the
wound is quickly covered by an epithelium that provides the necessary signals for the
underlying tissues to dedifferentiate, proliferate, and form the blastema also known as
the amphibian regeneration bud. Blastema tissues then undergo redifferentiation to form
muscle, bone and other mesodermal tissues to enable the reconstruction of the amputated
limb. Major cell signaling pathways activated in the blastema during this process are the
fibroblast growth factor (FGF) and transforming growth factor (TGF) pathways (Tsonis
1996). Additionally, the blastema appears to express the phosphorylated version of the
tumor suppressor gene, retinoblastoma (Rb) that is found highly expressed in many

diverse tumors. Cancer cells share similarities with blastema cells in that they are both
dedifferentiated and pluripotent. An animal with powerful regenerative capabilities is
often refractory to spontaneous or experimentally induced cancer; this is true for the
amphibia. Spontaneous tumors are difficult to find in this class of vertebrates (Tsonis and
Del Rio-Tsonis, 1988). Studies in the Hydra have identified a family of Wnt proteins,
produced during Hydra budding and at the tip of a decapitated Hydra when its head starts
to regrow (Hobmayer et al 2000). Thus, the FGF, TGF and Wingless-Type Mmtv
Integration Site Family (Wnt) signaling pathways appear to play important and
overlapping roles in developmental, cancer, regeneration and stem cell biology.

Plant meristems
Plants but not most animals have the remarkable capacity to regenerate from vegetative
parts. Many terminally differentiated plant organs, tissues and cells retain their capacity
10


to regenerate. For example, a stem segment broken off from Opuntia will regenerate a
new plant. Stem cells derive their name from their similarity to the stem of a plant.
Indeed, the etymological origins of the term “stem cell” can be traced back to early
botanical monographs documenting the regenerative competence of plant meristems
(Kiessling & Anderson 2003). Stem cells that are totipotent are also found in plants in
the shoot apical meristems (SAM).
Whole plants can be regenerated by either organogenesis or somatic
embryogenesis by tissue culture. In somatic embryogenesis, regenerating tissue
recapitulates embryonic development while in organogenesis, organs form directly
without embryogenesis. Studies have shown that whole carrot plants can be regenerated
from single vegetative cells (Stewart 1958). This means that differentiated plant
vegetative cells retain the ability to revert into a totipotent state.

Definition of a stem cell

Three basic categories of cells make up the human body viz., germ cells, somatic cells
and stem cells. Somatic cells include the bulk of the cells that make up the human adult
and each of these cells in their differentiated state has its own copy or copies of the
genome with the only exception being cells without nuclei viz., red blood cells. Germ
cells are cells that give rise to gametes viz., eggs and sperm. The canonical definition of
a stem cell is a cell with the ability to divide indefinitely in culture and in the living
organism whilst retaining the potential to give rise to mature specialized cell types
(Alison et al 2002). When a stem cell divides, the daughter cells can either enter a path
leading to the formation of a differentiated specialized cell or self-renew to remain a
stem cell, thereby ensuring that a pool of stem cells is constantly replenished in the adult
organ. This mode of cell division characteristic of stem cells is asymmetric and is a
11


necessary physiological mechanism for the maintenance of the cellular composition of
tissues and organs in the body. Other attributes of stem cells include the ability to
differentiate into cell types beyond the tissues in which they normally reside. This is
often referred to as stem cell plasticity. Stem cells are also believed to be slow cycling
but highly clonogenic and generally represent a small percentage of the total cellular
make up of a particular organ (Gardner 2002).
While there is still much to discover about the molecular mechanisms that govern
stem cell fate decisions and self-renewal, transcriptome profiling studies have
highlighted several properties believed to be common to all stem cells at the molecular
level. These essential attributes of “stemness” are proposed to include (i) active Janus
kinase signal transducers and activators of transcription, TGFβ and Notch signaling, ii)
the capacity to sense growth hormones and interaction with the extracellular matrix via
integrins, iii) engagement in the cell cycle, either arrested in G1 or cycling, iv) a high
resistance to stress with upregulated DNA repair, protein folding, ubiquitination, and
detoxifier systems, v) a remodeled chromatin, acted upon by DNA helicases, DNA
methylases, and histone deacetylases and vi) translation regulated by RNA helicases of

the Vasa type (Ramalho-Santos et al 2002, Ivanova et al 2002).

Totipotency, pluripotency and multipotency
Stem cells can also be classified as totipotent, pluripotent and multipotent. Totipotency is
the ability to form all cell types of the conceptus, including the entire fetus and placenta.
Such cells have unlimited capability. They basically can form the whole organism. Early
mammalian embryos are clusters of totipotent cells.
In mammals, the fertilized egg, zygote and the first 2, 4, 8 and 16 blastomeres
resulting from cleavage of the early embryo are examples of totipotent cells. Proof that
12


these cells are indeed totipotent arises from the observation that identical twins are
produced from splitting of the early embryo. However, the expression “totipotent stem
cell” is perhaps a misnomer because the fertilized egg and the ensuing blastomeres from
early cleavage events cannot divide to make more of them. Although these cells have the
potential to give rise to the entire organism, they do not have the capability to self-renew
and by strict definition therefore, the totipotent cells of the early embryo should not be
called stem cells.
Multipotency is the ability of giving rise to a limited range of cells and tissues
appropriate to their location for example, blood stem cells give rise to red blood cells,
white blood cells and platelets, while skin stem cells give rise to the various types of skin
cells. Some recent reports suggest that adult stem cells such as haemopoietic stem cells,
neuronal stem cells and mesenchymal stem cells could cross boundaries and differentiate
into cells of a different tissue (Bjornson et al 1999, Jackson et al 1999, Clarke et al 2000,
Karuse et al 2001). This phenomenon of unprecedented adult stem cell plasticity has
been termed “transdifferentiation” and appears to defy canonical embryological rules of
strict lineage commitment during embryonic development.
Pluripotency is the ability to form several cell types of all three germ layers
(ectoderm, mesoderm and endoderm) but not the whole organism. Pluripotent stem cells

have in theory the ability to form all the 200 or so cell types in the body. There are four
classes of pluripotent stem cells in humans, other primates and mice. These are
embryonic stem cells, embryonic germ cells, embryonic carcinoma cells and recently the
discovery of a fourth class of pluripotent stem cell, the multipotent adult progenitor cell
from bone marrow (Smith 2001).
It is generally assumed that the range of potential fates for human embryonic
germ cells (hEGCs) will be limited compared to human embryonic stem cells (hESCs)
13


because hEGCs are much further along in the schema of embryonic development. The
number of groups working with hESCs continues to expand rapidly and this coupled
with the deluge of exciting experimental reports on hESCs appears to have
overshadowed much of the interest in hEGCs.
Human embryonal carcinoma (hEC) cell lines are derived from tumors of germ
cell origin and have long served as the human counterpart of murine EC cells for
studying human development and differentiation in vitro (Andrews 2002). hEC cell lines
are capable of multi-lineage differentiation in vitro but being of tumor origin are
unfortunately mostly aneuploid making them unsuitable for cell replacement therapies.
Both hESC and hEC cell lines express similar stage specific embryonic antigens and
tumor rejection antigens on the surfaces of their cells. hEC lines also express the
pluripotency controlling transcription factor Oct-4, grow in colonies and are
morphologically similar to hESC with individual cells displaying a high nucleus to
cytoplasmic ratio. Several hEC cell lines also require the support of a feeder layer to
retain pluripotent characteristics. Not all hEC cell lines are pluripotent and some feederindependent hEC lines have been reported to be nullipotent.
A new class of pluripotent adult stem cells from the bone marrow has been
recently discovered. In a series of experiments, Jiang et al (2002) isolated mouse
multipotent adult progenitor cells (MAPCs) from murine bone marrow and demonstrated
that these cells express telomerase and that a single MAPC could be expanded clonally
into a large number of daughter cells. Additionally, under appropriate conditions,

MAPCs differentiate into ectoderm, endoderm and mesoderm and are capable of
generating chimaeric mice when injected into mouse blastocysts. Also, reporter gene
marked MAPCs contribute to adult tissues when injected into the veins of adult mice
(Jiang et al 2002). Although extremely promising, MAPCs are rare cells in the bone
14


marrow and difficult to isolate. It is also still unclear if these cells are truly biologically
equivalent to hESCs and if they can be expanded indefinitely whilst retaining their longterm differentiation potential. More data needs to be collected from human MAPCs as
most of the current experimental data are derived from studies in the murine model.

Classification of stem cells
Germ cells of the gonads
Mammalian stem cells are usually classified according to their tissue of origin. The
ovary and testis contain oogonia and spermatogonia which have been referred to as the
stem cells of the gonads. In adult mammals, only the germ cells undergo meiosis to
produce male and female gametes which fuse to form the zygote that retains the ability
to make a new organism thereby ensuring the continuation of the germ line. In fact the
zygote is at the top of the hierarchical stem cell tree being the most primitive and
producing the first two cells by cleavage. This unique characteristic of germ cells is
termed as developmental totipotency. Intriguingly, Oct-4 an embryonic transcription
factor critical for the maintenance of pluripotency continues to be expressed in the germ
cells but is absent in other peripheral tissues (Yoshimizu et al 1999, Pesce et al 1998).

Adult stem cells
Adult stem cells also known as somatic stem cells can be found in diverse tissues and
organs. The most common adult stem cells are the hematopoietic stem cells, mesechymal
stem cells and neuronal stem cells. Adult stem cells have also been isolated from several
other organs such as the brain (neuronal stem cells), skin (epidermal stem cells), eye
(retinal stem cells) and gut (intestinal crypt stem cells) (Spradling et al 2001). However,

15


not all organs and tissues may contain stem cells. The molecular marking and lineage
tracing of pancreatic cells have revealed that some organs like the islet component of the
pancreas do not contain any stem cells (Dor et al 2004).
Although many somatic stem cells have been very well characterized and isolated
in rodents, the human equivalents of these adult stem cells have been difficult to identify
and difficult to expand in vitro. This could reflect innate differences between human and
rodent cell physiology. Human somatic stem cells appear to display telomere dependent
replicative senescence while rodent stem cells do not (Wright & Shay 2000).

Hematopoietic stem cells
The best-studied adult stem cell is the hematopoietic stem cell (HSC). HSCs have been
used widely in clinical settings for over 40 years and form the basis of bone marrow
transplantation success. Unfortunately, HSCs like many other adult stem cells are rare
and difficult to isolate in large numbers from their in vivo niche. For example, only
approximately 1 out of 10,000 bone marrow cells is a HSC (Spradling et al 2001).

Mesenchymal stem cells
Mesenchymal stem cells (MSCs) are another well-characterized population of adult stem
cells. MSCs are prevalent in bone marrow at low quantities (1 out of 10,000 - 100,000
mononuclear cells). It is thought that they respond to local injury by dividing to produce
daughter cells that differentiate into multiple mesodermal tissue types, including bone,
cartilage, muscle, marrow stroma, tendon, ligament, fat and a variety of other connective
tissues (Short et al 2003). The ease of culture has greatly facilitated the characterization
of MSCs. In addition, recent studies have shown that the MSCs can also differentiate
into neuron-like cells expressing markers typical for mature neurons, suggesting that
16

adult MSCs may be capable of overcoming germ layer commitment. Several reports hint
that MSCs can form a variety of cell types and tissues including fat cells, cartilage, bone,
tendons and ligaments, muscle cells, skin cells and even nerve cells (Short et al 2003).

Neuronal stem cells
The discovery of neuronal stem cells has indicated that cell replenishment is possible
within the brain. Neuronal stem cells have been isolated from various regions of the
brain including the olfactory bulb (Pagano et al 2000) as well as the spinal cord
(Shihabuddin et al 2000), and can even be recovered from cadavers soon after death
(Palmer et al 2001). Several studies have shown that neuronal stem cells can produce not
only mature neurons but also other tissues, including blood and muscle (Bjornsen et al
1999, Galli et al 2003, Clarke et al 2000, Galli et al 2000, Rietze et al 2001, Englund et
al 2002) Some animal studies have shown that adult neural stem cells can participate in
repair of brain damage after stroke via endogenous neuronal precursor (Arvidsson et al
2002) as well as transplanted neural stem cells (Reiss et al 2002). Neural stem
cells/neural progenitor cells may also show low immunogenicity thereby raising the
possibility for use of donor neural stem cells to treat degenerative brain conditions.
Neural stem cells have also been used to investigate potential treatments for Parkinson’s
disease (Liker et al 2003, Kim et al 2003). Pluchino et al (2003) recently used adult
neural stem cells to test potential treatment of multiple sclerosis lesions in the brain.
Using a mouse model of chronic multiple sclerosis, they injected neural stem cells either
intravenously or intracerebrally into affected mice. Donor cells entered damaged,
demyelinated regions of the brain and differentiated into neuronal cells. Remyelination
of brain lesions and recovery from functional impairment were also seen in the mice.

17

Fetal stem cells
Fetal stem cells are primitive cell types in the fetus that eventually develop into the
various organs of the body. Research with fetal stem cells has thus far been limited to
only a few cell types because of the unavailability of abortuses. These include neural
crest stem cells, fetal hematopoietic stem cells and pancreatic islet progenitors (Beattie et
al 1997). Fetal neural stem cells are abundant in the fetal brain and have been shown to
differentiate into both neurons and glial cells (Brustle et al 1998, Villa et al 2000). Fetal
blood, placenta and umbilical cord are rich sources of fetal hematopoietic stem cells.
Several commercial enterprises trying to capitalize on the theoretical potential of fetal
hematopoietic stem cells as a source of stem cells for cell replacement therapy have
surfaced in the last few years.
Umbilical cord blood (UCB) contains circulating stem/progenitor cells, and the
cellular contents of UCB are known to be quite distinct from those of bone marrow and
adult peripheral blood. Over the past two decades, the presence and characteristics of
hematopoietic stem cells in UCB have been clarified (Nakahata & Ogawa 1982,
Broxmeyer et al 1989, Gluckman et al 1989). The frequency of UCB hematopoietic
stem/progenitor cells equals or exceeds that of bone marrow and surpasses that of adult
peripheral blood (Mayani & Lansdorp 1998). Compared with adult cells, UCB
hematopoietic stem cells produce larger hematopoietic colonies in vitro, have different
growth factor requirements, are able to expand in long-term culture in vitro, and have
longer telomeres (Smith & Broxmeyer 1986, Salahuddin et al 1981, Gluckman 2000).
UCB transplantation for various hematopoietic diseases has resulted in successful
hematopoietic reconstitution and a lower incidence of graft-versus-host disease than
expected with conventional therapies (Wagner et al 1995, Kurtzberg et al 1996).
Recently, it has been reported that UCB contains mesenchymal progenitor cells capable
18


of differentiating into marrow stroma, bone, cartilage, muscle, and connective tissues
(Erices et al 2000). Furthermore, UCB provides no ethical problems for basic studies and

clinical applications. UCB cells can be collected without any harm to the newborn infant,
and UCB hematopoietic stem cell grafts can be cryopreserved and transplanted to a host
after thawing without losing their repopulating ability (Rubinstein et al 1995). For these
reasons, UCB could be a prominent source of cells for transplantation in various
diseases. It remains obscure, however, whether UCB contains stem/progenitor cells
leading to endodermal cells, including hepatocytes. Although working with umbilical
cord blood appears to circumvent the majority of the ethical issues associated with
research on fetal material, fetal stem cell research is in many ways underdeveloped and
is still in its infancy.

Primordial germ cells
Primordial Germ Cells (PGCs) are diploid germ cell precursors that transiently exist in
the embryo before they enter into close association with the somatic cells of the gonad
and become irreversibly committed as germ cells. Human Embryonic Germ (hEG) cells,
also a form of stem cells are isolates of PGCs from the developing gonadal ridge of 5 to
9 week old fetuses of elective abortions. Shamblott et al (1998) reported the successful
isolation and characterization of hEG cell lines. hEG cells are pluripotent and are capable
of forming all 3 primordial germ layers (Shamblott et al 1998; 2001).

Embryonic stem cells
Embryonic stem cells on the other hand are derived from the isolated inner cell masses
(ICM) of mammalian blastocysts. The continuous in vitro sub-culture and expansion of
an isolated ICM on an embryonic fibroblast feeder layer (human or murine) leads to the
19


development of an embryonic stem cell line. In nature however, embryonic stem cells
are ephemeral and present only in the ICM of blastocysts. The cells of the ICM are
destined to differentiate into tissues of the three primordial germ layers (ectoderm,
mesoderm and endoderm) and finally form the complete soma of the adult organism.

Embryonic stem cells can be expanded in vitro very easily and under optimal
culture conditions divide symmetrically to give two daughter cells. Embryonic stem cell
lines express the telomerase gene, the protein product of which ensures that the telomere
ends of the chromosomes are retained at each cell division preventing the cells from
undergoing senescence. These cells also retain a normal karyotype after continuous
passage in vitro thus making them truly immortal. The earliest hESC lines derived in our
laboratory have been maintained continuously in culture for over 300 population
doublings, a figure which surpasses the theoretical Hayflick limit of 50 population
doublings for normal cells (Bongso et al 1993; 1994, Richards et al 2002).
To qualify as a bona fide embryonic stem cell line, the following criteria must be
satisfied, i) immortality and telomerase expression, ii) pluripotentiality and teratoma
formation, iii) maintenance of stable karyotype after extended in vitro passage, iv)
clonality, v) Oct-4 expression and vi) ability to contribute to chimera formation through
blastocyst injection. hESCs have fulfilled all criteria with the exception of chimera
contribution (Ramalho-Santos et al 2002). For obvious ethical reasons, experiments
involving blastocyst injections and ectopic grafting in adult hosts cannot be performed in
the human.

Animal embryonic stem cells
Embryonic stem cells were first derived from certain strains of mice (Evans & Kaufman
1981). Embryonic stem cell lines have been established from the mouse, chicken,
20


hamster, rabbit, pig, bovine, fish (medaka) (Hong et al 1999), primate and humans.
However, only mouse and chicken embryonic stem cells appear to have germ line
competence and have the ability to contribute to chimera formation thus making them
true embryonic stem cell lines. Strikingly no group has yet been able to derive bona fide
rat embryonic stem cell lines. In general, rat rather than mouse physiology is believed to
be a closer parallel to human physiology. Therefore, the rat model would in theory be a

closer representative for the study of human disease. For example, there is a rat model
for hypertension but no similar mouse model. The rat asthma model also mimics many
features of human asthma and given the same level of cholesterol and triglycerides, the
rat atherosclerosis model demonstrates coronary artery disease and decreased survival
comparable to that of humans (Bice et al 2000, Stoll & Jacob 2001). Therefore, if rat
embryonic stem cell lines can be established they will perhaps benefit medical science in
more ways than mESCs have, through the use of knock-out technology.
Several lines of evidence suggest that hESCs and mESCs do not represent
equivalent embryonic cell types. In vitro differentiation of hESCs leads to the expression
of alpha-feto protein (AFP) and human chorionic gonadotropin (HCG), which are
typically produced by trophoblast cells in the developing human embryo, while mESCs
are generally believed not to differentiate along this extra-embryonic lineage. hESCs
express the stage-specific embryonic antigens (SSEA)-3, SSEA-4, tumor rejection
antigen (TRA)-1-60, and TRA-1-81 surface antigens prior to differentiation but only
SSEA-1 upon differentiation, while mESCs only express SSEA-1 prior to differentiation.
More strikingly, hESCs do not appear to have a perceivable LIF response unlike mESCs,
which can be maintained, in the undifferentiated pluripotent state in vitro with
exogenous LIF supplementation (Smith et al 1988, Willliams et al 1988). Transcription

21


profiling studies have shown that LIF and its cognate receptor are expressed at extremely
low levels in hESCs (Richards et al 2004).
Human embryonic stem cells
Human embryonic stem cell lines were first established in 1998. To date there are 78
National Institute of Health (NIH) USA registered hESC lines, all of which have been
derived on MEFs. The establishment of hESC lines is a highly efficient procedure, with
up to a 60% success rate from spare IVF blastocysts (Richards et al 2002, Thomson et al
1998, Reubonoff et al 2000). The quality of the donated embryos appears to be an

important determinant of success in deriving hESC lines. Nevertheless, protocols for
hESC line derivation have been reproduced in many labs and are relatively easy to
follow (Reubinoff et al 2000, Richards et al 2002, Cowan et al 2004).

History of human embryonic stem cell research
Pluripotent embryonal carcinoma (EC) lines were the first kind of stem cells that were
recognized in terminally differentiated tissues of spontaneously occurring murine
tumours (teratocarcinomas) (Andrews 2002). They can be stimulated to differentiate in
vivo as well as in vitro. Their similar characteristics and behaviour to embryonic stem
cells served as a model to isolate comparable cells from mammalian embryos.
The first report on the growth of ICMs and the isolation of stem cells from
human blastocysts was by Bongso et al 1993, 1994. In their study, 9 patients enrolled in
an IVF program donated 21 embryos for hESC production. All 21 embryos at the
pronuclear stage were co-cultured on human oviductal epithelial feeders to generate
blastocysts. The zona pellucida was then removed with pronase and zona-free blastocysts
cultured on irradiated human oviductal feeders as a whole embryo culture in the presence

22


of Chang’s medium supplemented with 1000 units/ml of hLIF. Nineteen of the 21
embryos produced “healthy” ICM lumps which were mechanically separated, trypsinised
and passaged further on fresh irradiated human feeders. Nest-like embryonic stem cells
colonies were produced which were mechanically cut with hypodermic needles,
disaggregated into single cells with trypsin-EDTA and seeded onto fresh irradiated
human feeders. It was possible to retain the typical hESC morphology of high nuclearcytoplasmic ratios, alkaline phosphatase positivity and normal karyotype for two
passages in 17 of the embryos (Bongso et al 1993;1994).
Later, primate embryonic stem cell lines were successfully produced from the
rhesus monkey (Thomson et al 1995) and the human (Thomson et al 1998). Irradiated
murine embryonic fibroblast (MEF) feeders, immunosurgery to separate the ICM and

passaging of clumps of hESCs instead of disaggregation into single cells were used.
Immunosurgery, mitomycin C treated MEFs and a similar ‘cut and paste’ method was
later used to derive and propagate hESC lines that would spontaneously differentiated
into neuronal cells (Reubinoff et al 2000). Amit and Itskovitz (2002) confirmed that the
whole embryo culture worked as well as the immunosurgery protocol to produce hESC
lines. Given the social nature of hESCs as known today, the disaggregration of ICM and
hESC colonies with trypsin into single cells during early passage rather than a ‘cut and
paste approach’ may have been responsible for the hESCs differentiating after two
passages in the early reports of Bongso et al (1993, 1994).
Reliance on a xeno-support system such as MEF introduces considerable
disadvantages with respect to exploiting the therapeutic potential of hESCs. A major
drawback is the risk of transmitting pathogens from the animal feeder cells or
conditioned medium to hESCs. The derivation of hESC lines on xeno-free support
systems in the presence of xeno-free proteins thus needs to be urgently developed.
23


Differentiation, transdifferentiation and stem cell plasticity
Differentiation is the process whereby an unspecialized early embryonic cell acquires the
features of a specialized cell such as that of a heart, liver or muscle. Differentiation in
vitro can be spontaneous or controlled. From a teleological perspective there appears to
be no limit to the types of cells that can be formed from hESC differentiation. This is in
contrast to the practical and theoretical constraints levied on somatic stem cells by virtue
of their position in embryonic development.
In vitro, hESCs spontaneously differentiate in high-density cultures or when
culture conditions are sub-optimal to yield a mixed milieu of differentiated cell types.
Cells and tissues representative of all 3 germ layers including neurons, cardiomyocytes
and primitive endoderm have been identified in differentiating hESC cultures. However,
to fully appreciate the plasticity of hESCs one has to look at the teratomas formed in
immune compromised mice when undifferentiated hESCs are injected into these hosts to

allow spontaneous differentiation and tumor formation. Histological sections of nonmalignant teratomas reveal complex, well-organized organ-like structures representative
of tissues from the ectoderm, endoderm and mesoderm. Gut-like structures, bone and
cartilage, neural rosettes and glandular epithelium with secretions are commonly found
in hESC formed teratomas.
Several groups have reported controlled in vitro differentiation of hESCs.
Typically, hESCs are induced to form embryoid bodies (EBs) by removal of the feeder
layer and the disaggregation into single cells in suspension culture. Alternatively, the
hanging drop method is used to induce EB formation. EBs and hESCs have been found
to differentiate in response to treatment with an array of protein-based cytokines and
growth factors (Schuldiner et al 2000). However, in these studies homogenous

24


differentiation into specific cell types was not achievable, instead, the final population of
cells consisted of mixed cell types representative of two or three germ layers.
To date, several studies have been published on the targeted differentiation of
hESCs. Kehat et al (2001) described a reproducible method based on spontaneous
differentiation to derive cardiomyocytes while Mummery et al (2002) used co-culture
techniques with isolates of primitive visceral endoderm to induce cardiomyocyte
formation in hESCs. Reubinoff et al (2001) and Zhang et al (2001) described methods
for the isolation of neural precursors from differentiating hESC cultures and showed
incorporation of these precursor cells in animal hosts while Assady et al (2001) reported
the ability of hESCs to differentiate into insulin-secreting cells. More recently, hESCs
have also been shown to be capable of differentiating into germ cell-like derivatives
(Clark et al 2004).
Although these studies represent reproducible and convincing examples of
controlled in vitro hESC differentiation, in many respects much of this work will not be
applicable in a clinical setting due to the low efficiencies of the procedures and the
difficulties involved in isolating pure and specific precursor cell types. Furthermore,

none of these reports describe a truly efficient directed differentiation strategy.
Currently, the best example of a directed differentiation strategy is probably that of bone
morphogenetic protein 4 (BMP4) induction of hESCs into the formation of trophoblast
cells where up to 40% conversion of hESCs into trophoblasts was reported (Xu et al
2002).
Nevertheless, these early reports of hESC differentiation lay an excellent
framework for the establishment of true efficient directed differentiation strategies for
the large-scale derivation of differentiated specialized cell types from hESCs and the
subsequent functional testing of these cells in primate models.
25