When the International Society for Stem Cell Research
(ISSCR) first met in Washington, D.C. in 2003, a few hundred attendees
participated in the discussions. In June 2013, just 10 years later, a record
4,000 researchers from all over the world attended the society's meeting in
Boston. The ISSCR now has more than 3,000 members and three affiliated indexed
journals, one of which has one of the highest impact factors in the field. In
addition, the number of abstracts that utilized reprogramming technology
increased exponentially from basically none to more than 220 in just 5 years
(see graph). The ISSCR's rapid growth has run parallel with an unprecedented
display of general interest on the part of researchers and clinicians from
different backgrounds and levels of expertise. Both trends speak directly to the
potential impact of stem cell research.
We all begin our lives with one
major stem cell: a fertilized egg. That one stem cell then divides and forms new
cells that, in turn, also divide. Even though these cells are identical in the
beginning, they become increasingly varied over time. As a result of this
process, which we call cell differentiation, our cells become specialized for
their locations in the body. As we develop in the womb, our cells differentiate
into nerves, muscles, and so on, and the organs begin to organize and function
together.
Scientists long believed that a mature or specialized cell
could not "reprogram," or return to an immature state. A few researchers
challenged this view, however. In 1966, John Gurdon (Wellcome Trust/CRUK Gurdon
Institute, Cambridge, UK) was the first to show that if you removed the nucleus
containing the genetic material of a fertilized frog egg (stem cell) and
replaced it with the nucleus of a fully differentiated intestine cell from a
tadpole, the modified egg would grow into a normal frog with the same genetic
material as the original egg.[1]
Gurdon's findings were confirmed by others, including Robert Briggs and
Thomas King Jr., whose earlier works showed that normal hatched tadpoles could
be obtained by transplanting the nucleus of a blastula cell to the enucleated
eggs of a leopard frog (Rana pipiens).[2] In 1997, Ian Wilmut electrofused
(a technique used to fuse cells using electrical impulse) nuclei of cultured
sheep adult mammary gland cells into enucleated sheep eggs and produced a single
cloned sheep named Dolly.[3] These researchers sent the
scientific community this message: it was now possible to reprogram adult cells
to an immature state by exposing them to a yet-unknown combination of factors
that were present inside enucleated eggs. These reprogrammed cells became
pluripotent again, meaning they were capable of going through a new process of
maturing and specializing.
Even though the pioneering researchers
provided the proof of principle that reprogramming was possible, the cloning
experiments they performed were very time-consuming, difficult to reproduce,
extremely inefficient for mammalian cells, and ethically controversial when
envisioned for human cells. In addition, an important piece of the puzzle was
still missing: What made the reprogramming of adult cells possible? It was not
until 2006 that Japanese researcher Shinya Yamanaka and his postdoc Kazutoshi
Takahashi were able to answer this question.
The Reprogramming
Pioneers
When Yamanaka presented his first reprogramming results
at the 2006 ISSCR meeting, many scientists were skeptical. Yamanaka claimed that
with the addition of only four factors that are master regulators of cell
pluripotency, his team could induce an adult skin cell (fibroblast) to become a
pluripotent stem cell (then called an induced pluripotent stem cell, or iPS
cell) within only a month. Many thought his results were too good to be true,
but later that year, when his procedure was published with a description of the
four factors he used for reprogramming experiments, dozens of labs around the
world (including ours) tried his protocol.[4,5] To our complete astonishment, it
worked in our lab the very first time-and it worked in many other labs as
well.[6,7] Yamanaka
and Takahashi's research results played a major role in popularizing and
disseminating stem cell research because by uncovering the basic factors and
principles of the reprogramming process, they made it possible for researchers
from other fields to work with pluripotent stem cells. The impact and potential
of their stem cell research earned Yamanaka and Gurdon the Nobel Prize in
Physiology or Medicine in 2012.
Using iPS Cells to Study Neurological
Diseases
Human iPS cells, which can, in principle, form any cell
in the body, could provide an attractive alternative when the traditional models
for neurological diseases are inadequate.
Nearly all of our current
knowledge about human neurodevelopmental and neurodegenerative diseases at the
cellular level is derived from studies in postmortem brain tissues. These
samples often represent the end stage of a disease and therefore are not always
informative representations of a disease's developmental path. Furthermore, the
pathology observed in these tissues is potentially not the authentic disease
cellular phenotype. Genetically modified ("transgenic") mice provide an
alternative way to reproduce human genetic forms of neurodegenerative diseases
to serve as models for observation as to their developmental course in a neurotrophic phrase. However, use of these models is
limited to monogenetic (the origin of diverse individuals or kinds by descent
from a single ancestral individual or kind) disorders in which the specific gene
mutations are known-disorders that represent the minority of neurological
diseases. And in some cases, mouse transgenic technology cannot adequately model
neurological disorders with defined genes because of intrinsic differences
between species.
For example, mice have much less of a complex brain
architecture than humans; there are a number of brain structures present in
humans that are not present in rodents. This suggests a need for advancement
toward human models of disease. Currently, many subtypes of disease-relevant
neurons can be developed from iPS cells using a combination of manual selection
and the addition of mixtures of different neurotropic factors to the culture.
Differentiation protocols can provide enriched populations of particular
subtypes of neurons that are relevant to specific diseases. These subtypes
include dopaminergic neurons for Parkinson's disease, hippocampal and
cholinergic neurons for Alzheimer's disease, motor neurons for amyotrophic
lateral sclerosis (ALS, or Lou Gehrig's disease), and inhibitory interneurons
for schizophrenia. [7-10]
To date, most
experiments involving disease modeling for neurological diseases utilize iPS
cells-derived neurons from patients with monogenetic disorders for which the
gene mutation is defined and well characterized. The modeling of monogenic brain
disorders has promoted rapid advancements in the field by helping to establish
the basic tools for culturing functional human neurons. In addition, initial
modeling research revealed meaningful neuronal phenotypes, such as differences
in synaptogenesis, neuronal size and arborization complexity, and connectivity
properties.11, 12 Importantly, monogenic disorder modeling presents
an opportunity to perform gain-of-function and loss-of-function studies and to
confirm the specificity of the neuronal phenotypes observed. In addition,
studying the in vitro phenotypic consequences of the mutation in specific genes
can highlight molecular mechanisms responsible for subtle alterations in the
nervous system, perhaps pointing to common mechanisms for more complex,
multi-gene diseases.
Nonetheless, the vast majority of neurological
disorders (for example, autism spectrum disorders, schizophrenia, Parkinson's
disease, Alzheimer's disease, and Lou Gehrig's disease) are complex in
nature and likely multifactorial: a combination of mutations in several
genes and extrinsic factors (such as influence of neighboring cells in the
neuronal niche and environment) is likely involved in the disease pathology
course. Recently, scientists have made successful attempts to detect a specific
neuronal phenotype using sporadic neurological disease models. Hopefully there
will be more advances in the near future as the technology becomes sensitive
enough to detect more subtle phenotypes.[11-13]
Finding
Clinically Relevant Drugs
Candidate compounds for treating
central nervous system (CNS) deficiencies fail in clinical trials in more than
90 percent of cases because of poor targeting (the drug does not target the
affected area of the brain efficiently), lack of efficacy, and unacceptable side
effects.[14] Pluripotent stem cells derived
from patients with CNS diseases offer a significant advantage, as researchers
can take into consideration the patient's genetic background and the
developmental course of the disease. Importantly, these stem cells allow for the
generation of both genetic and sporadic forms of the disease.
Before
developing a screening platform with the aim of discovering new drugs for a
treatment, a consistent abnormal phenotype needs to be identified and reproduced
on a large scale. Researchers are making progress in this process, and as large
pharmaceutical companies move into stem cell-related drug research, more
systematic progress is expected.[15-17] The
best examples so far are coming from partnering between research organizations
(universities and institutes) with industry and start-up companies that have
scientists as advisors. A few months ago, a group from iPierian Inc. configured
a high-content chemical screen using an indicator of ALS pathology in human
motor neurons derived from iPS cells from patients with ALS. The group
identified small molecule compounds (i.e. digoxin) that alleviated the
disease-related phenotype in iPSC-derived patient neurons, thus demonstrating
the feasibility of iPS cell-based disease modeling for drug screening. The
general strategy for drug screening is to identify a reliable disease-related
phenotype and to develop high-throughput screening platforms to test bioactive
compounds (such as proteins and small molecules) that protect the patient
neurons from either developing or progressing through the disease course. After
rigorous testing, these screenings will likely unearth therapeutic compounds
that could benefit a group of patients.
Finally, iPSCs may also be used
to assess developmental as well as cell-type-specific drug toxicities. Indeed,
existing commercially available human iPS-derived hepatocytes, cardiomyocytes,
and neural cells may provide the basis for humanized assays to detect off-target
activity and side effects of drugs in a tissue-specific manner.[18] We
firmly believe that reprogramming technology can be a valuable, additional tool
for screening and validating CNS compounds for pharmaceutical companies in the
near future, ultimately culminating in the discovery of new therapies.
Cautionary Notes
iPS cell lines and their derived
progeny bear a significant intrinsic variability, as revealed by abnormal
expression of imprinted genes, differential expression profiles, and
inconsistent neuronal differentiation competence.[19-21] For that reason, researchers
still need to conduct comparative experiments with well-established human
embryonic stem (HES) cell lines as a benchmark for complete reprogramming and
ideal differentiation protocols. It is our expectation that the use of HES cell
lines may decrease over time, but studying reprogramming without them would be
unconceivable at this point.
This variability can become a real hurdle
for disease modeling, especially when comparing cells from patients with
sporadic forms of diseases that have multifactorial etiologies. The differences
observed have been generally attributed to random integration of viral vectors
causing potential insertional mutagenesis, reactivation of reprogramming
transgenes, and persistency of donor cell gene expression.[22] New technology that promotes the
delivery of reprogramming factors in a non-integrative way is available and
becoming more popular among disease modeling groups.[23,24] Reprogramming can also be
achieved by using synthetic genes and small molecules, and further improvement
of these methodologies will promote widespread use by the scientific
community.[25] As more research groups use
nonintegrative approaches, we anticipate that the iPS cell lines generated will
have decreased intrinsic variability.
Identifying disease-relevant
phenotypes requires researchers to compare experimental cells with "healthy"
control cells. New gene-targeting technologies in iPS cells can enable more
efficient and less variable rescue from monogenetic alterations. In addition,
the generation of isogenic cell lines allows for more relevant controls that
take into account the individual's genetic background. Examples of methods
currently using iPS gene editing are zinc-finger nucleases (ZFNs), transcription
activator-like effector nucleases (TALENs), and clustered regularly interspaced
short palindromic repeats (CRISPR). [26-29]
For sporadic cases, alternative ways to decrease variability will include
using neurotypical family members as controls or including groups of patients
who present common clinical histories and/or respond to drugs in a similar
manner. New high-throughput genomic tools, such as genomic deep sequencing, are
beginning to reveal naturally occurring genetic variation that can help us to
understand the differences between cell lines. When possible, reprogramming
cells from genetically identical individuals, such as monogenetic twins who are
concordant or discordant for a specific neurological condition, will also help
us to understand variability and to generate relevant disease hypotheses.
The Road Ahead
Reprogramming technology has opened
the door for many new insights into the brain and brain-related conditions. The
recapitulation of early stages of human neural development made possible by
using iPS cells is an invaluable tool that can reveal the exact moment of the
disease onset, thus fostering the generation of new diagnostic tools and
potentially optimizing novel therapeutic interventions.
Although it has
been only seven years since the introduction of somatic reprogramming technology
to generate iPS cells, clinical studies that bring iPS cell-based therapy to
patients are already underway. In August 2013, the Japanese Ministry of Health,
Labour, and Welfare approved the first pilot clinical study using isogenic iPS
cells for age-related macular degeneration (AMD). The study will be conducted
mainly by the Takahashi group at the RIKEN Center for Developmental Biology in
Kobe, Japan. They plan to transplant sheets of iPS cell-derived retinal cells
into the subretinal space of AMD patients to rescue and restore the pigmented
epithelium responsible for absorbing visual stimuli.[30] If Takahashi's study is
conducted safely, it will be the first clinical demonstration of iPS cells for
medical use and will undoubtedly impact the outlook regarding the safety and
efficacy of iPS cell-based therapy. Advanced Cell Technology, an American
biotechnology company, is applying for Federal Drug Administration approval for
a less ambitious clinical trial of injecting human iPS cell-derived platelets as
a potential treatment of coagulopathies. Because platelet cells lack a nucleus,
scientists expect that the risks of tumors and tumor-associated immune responses
will decrease. Nonetheless, the main challenge in the field remains: Much more
groundwork is needed to improve understanding of the biology of reprogrammed
cells and their progenies. In addition, we need to be vigilant about avoiding
the dissemination of unproven applications.
Incorporation of bioengineering techniques making the use of bio scaffolds to
allow for cells to grow in three-dimensions will raise our level of
understanding of the different brain structures and eventually begin to dissect
out the birth of more complex neuronal networks. Earlier this year, an Austrian
group led by Jürgen
Knoblich assembled in vitro the first iPS cell-derived rudimentary
brain.[31] The cerebral organoids produced
by the researchers recapitulated early stages of human development (up to
approximately nine weeks of pregnancy) and modeled for microcephaly, a
neurological condition that is not efficiently modeled in rodents. More
refinement of the technique will be required in order to maintain the cells as
organoids or tissue in a viable and stable state for longer periods;
nevertheless, the tissue-engineering approach is a very promising and powerful
tool for understanding various aspects of human brain development.
Neuroscientists in the past could not have predicted a scenario in which
patient-derived, live functional neurons would be readily available for
research, and researchers in the future will not be able to imagine a scenario
without it.
