九州大学学術情報リポジトリ
Kyushu University Institutional Repository
Development of genetically engineered PA6 feeder cells for neural differentiation of mouse and human iPS cells
パルハティ, パルヴェン
http://hdl.handle.net/2324/2236048
出版情報：九州大学, 2018, 博士（システム生命科学）, 課程博士 バージョン：
権利関係：
Development of genetically engineered PA6 feeder cells for neural differentiation of mouse and human iPS cells
Paerwen Paerhati
March 2019
TABLE OF CONTENTS
Chapter 1...1
Introduction...1
1.1 Neurodegenerative diseases...1
1.2 Regenerative medicine...2
1.3 Stem cells...4
1.4 Research purpose...7
1.5 Thesis components...8
Chapter 2...9
Background...9
2.1 Neural differentiation of iPS cells...10
2.1.1 Neurons...10
2.1.2 General methods for pluripotent stem cells neural differentiation...13
2.1.3 SDIA...15
2.1.4 Feeder cells...16
2.2 Cadherins...18
2.2.1 Cadherin subtypes...18
2.2.2 Cadherin expressing feeder cells...20
2.3 Neuromuscular junction...20
2.4 Conclusion...22
Chapter 3...23
Neural differentiation of mouse induced pluripotent stem cells using cadherin gene-engineered PA6 feeder cells...23
3.1 Introduction...23
3.2 Material and Methods...25
3.2.1 Cell culture...25
3.2.2 Construction of E-cadherin or N-cadherin gene-engineered PA6 cells...25
3.2.3 Immunocytochemical analysis...27
3.2.4 Neural differentiation...28
3.2.5 Co-culture assay with myotubes...29
3.2.6 Statistical analysis...30
3.3 Results and Discussion...30
3.3.1 Cadherin expression in genetically engineered PA6 feeder cells...30
3.3.2 Neural differentiation efficiency of miPS cells on cadherin gene-engineered PA6 feeder cells...32 3.3.3 Effect of cadherin gene-engineered PA6 feeder cells on miPS cell differentiation into motor
neurons...36
3.4 Conclusion...40
Chapter 4...41
Neural differentiation of mouse and human iPS cells using constitutive E-cadherin expressing mouse PA6 feeder cells...41
4.1 Introduction...41
4.2 Materials and Methods...43
4.2.1 Cell culture...43
4.2.2 Plasmid construction...44
4.2.3 Transfection of PA6 cells...47
4.2.4 Puromycin resistant drug screening...47
4.2.5 Reverse transcription PCR analysis...48
4.2.6 Measurement of cell growth...48
4.2.7 Immunocytochemical analysis...48
4.2.8 Neural differentiation of mouse iPS cells...49
4.2.9 Neural differentiation of human iPS cells...49
4.2.10 Statistical analysis...50
4.3 Results and Discussion...50
4.3.1 Constitutive E-cadherin expressing genetically engineered PA6 feeder cells...50
4.3.2 Neuron and motor neuron differentiation of miPS cells on constitutive E-cadherin expressing genetically engineered PA6 cells...59
4.3.3 Neural differentiation of hiPS cells on constitutive E-cadherin expressing genetically engineered PA6 feeder cells...63
4.4 Conclusion...65
Chapter 5...66
Conclusion...66
References...69
Acknowledgement...76
LIST OF FIGURES
1. Figure 1-1 Process of regenerative medicine……….3
2. Figure 1-2 Types of stem cells………...4
3. Figure 1-3 Establishment of ES cells………....5
4. Figure 1-4 Establishment of iPS cells………..….6
5. Figure 2-1 Neurons structure………...10
6. Figure 2-2 Classification of neurons based on their role………..11
7. Figure 2-3 Classification of neurons based on their structure……….12
8. Figure 2-4 SDIA method……….15
9. Figure 2-5 Cadherin………19
10. Figure 2-6 Nueromuscular junction………21
11. Figure 3-1 Dox inducible gene expression system………..24
12. Figure 3-2 Retroviral vector constructs encoding for the rtTA, E-cad, and N-cad gene….26 13. Figure 3-3 Generation of cadherin gene-engineered PA6 cells………31
14. Figure 3-4 Neural differentiation of iPS cells on PA6 feeder cells for 6 days……….34
15. Figure 3-5 Neural differentiation of iPS cells on PA6 feeder cells for 10 days………35
16. Figure 3-6 Motor neuron differentiation of iPS cells on PA6 feeder cells………..37
17. Figure 3-7 Co-culture assay with myotubes………38
18. Figure 4-1 DNA transposon………42
19. Figure 4-2 Flowchart for the construction of PB/Ecadherin vector……….46
20. Figure 4-3 PB/Ecadherin vector construction……….51
21. Figure 4-4 Sequence alignment………...53
22. Figure 4-5 Analysis of new generated constitutive PA6/E cells………..58
23. Figure 4-6 Neural differentiation of miPS cells on PA6 feeder cells………60
24. Figure 4-7 Neurite number evaluation ………62
25. Figure 4-8 Neural differentiation of hiPS cells on PA6 feeder cells……….64
LIST OF TABLES 1. Table 4-1 Primers used for sequencing………46
2. Table 4-2 Different electroporation conditions………47
Abstract
The neurodegeneration of the CNS is caused by chronic, progressive disorders caused by the gradually loss of neurons in the central nerve system (CNS). However, the CNS, which controls the activities of the body, in vertebrates lack of ability to regenerate. As the elderly population has increased in recent years, the age-related neurodegenerative diseases are becoming increasingly prevalent. Neurodegenerative diseases like Alzheimer’s disease and Parkinson’s disease not only affect patience’s behavior, memory, language, but also every organ in the body. Since, CNS of vertebrate lack of regenerative ability, rescue the remaining neurons, increase the number of neurons or replace the lost neurons became the potential therapeutic strategies for treating neurodegenerative diseases.
Regenerative medicine, especially stem cell technology, take advantages of the body’s natural ability to heal itself, offers realistic way to restore diseased, injured tissues and whole organs. By this, regenerative medicine can not only relieving patience’s symptoms, but addressing the root cause of the pain. Pluripotent stem cells including embryonic stem (ES), induced pluripotent stem (iPS) cells are characterized by self-renewal and differentiation potential for all cell types, which lead pluripotent stem cells become a frontline source of regenerative medicine. Therefore, to study early neural development and also offer a potential source of cells for nerve regeneration, investigating neural differentiation of ES/iPS cells is of important significance. Stromal cell-derived inducing activity (SDIA) using mouse stromal PA6 cells promotes neural differentiation of iPS cells. Thus, we hypothesized that cadherin gene-engineered PA6 feeder cells will enhance the performance of SDIA by facilitating cell- cell interactions. Consequently, we created two types, Dox inducible and constitutive, cadherin expressing PA6 feeder cells. Neural differentiation efficiency of mouse and human iPS cells on genetically engineered feeder cells was significantly higher compared with parental PA6 feeder cells. In addition, motor neuron differentiation efficiency of mouse iPS
cells on Dox inducible cadherin expressing feeder cells was significantly higher compared with parental PA6 feeder cells. Altogether, these results indicate that cadherin gene-
engineered feeder cells are a potent tool for promoting neural differentiation of pluripotent stem cells.
Chapter 1
Introduction
By the increasing of elder people populations, the age-related neurodegenerative diseases are becoming increasingly prevalent. Different neurodegenerative disease involved the lost or degeneration of different types of neurons and proteins in the nervous system. Although there is great progress has been made in traditional medicine, few effective methods to treat
neurodegenerative diseases. Regenerative medicine, which is a branch of translational research, opened a new avenue by taking advantages of the body’s natural ability to heal itself and offers realistic way of restore diseased, injured tissues and organs. ES/iPS cells with their ability to self-renewal and differentiate into various types of cells becoming an important tool for studying early neural development, providing a potential cell source for regenerative medicine. In order to study and treat different types of neurodegenerative disease it is important to investigate efficient method for specific types of neuronal
differentiation. In this chapter, general knowledge about neurodegenerative disease, field of regenerative medicine, especially stem cells are presented.
1.1 Neurodegenerative diseases
In vertebrates, the nervous system consists of two main parts, the central nervous system (CNS) and peripheral nervous system (PNS). Among them, the CNS consists of the brain and spinal cord, the PNS including any part of the nervous system that lies outside of CNS.
Although CNS and PNS are interconnected, they are separated from each other. Main differences between CNS and PNS is their regeneration ability. Majority of the PNS has the ability to regenerate, however, the CNSs does not have this ability. Both the CNS and PNS
contains two basic types of cells: neurons and glial cells. The neuron is an electrically
excitable cell which in terms of the communicative function of the nervous system. Glial cell is able to provide a framework of tissue to supports the neurons and their activity. Neurons would not be able to function without the important roles fulfilled by glial cells.
Neurodegenerative diseases refer to a group of chronic, progressive disorders caused by the gradual loss of neurons in discrete areas of the CNS. It has become a major threat to human health. These neurodegenerative diseases including Parkinson’s disease (PD), Alzheimer’s disease (AD), amyotrophic lateral sclerosis (ALS), Huntington’s disease and multiple system atrophy. These age-dependent disorders are diverse in their pathophysiology, which not only affect patience’s behavior, memory, language, but also every organ in the body. Since, CNS of vertebrates has little or lack of regenerative ability, the generation of neurons in the mammalian brain had been restricted to early embryonic development stage [1]. Therefore, rescue the remaining neurons, increase the number of neurons or replace lost neurons are potential therapeutic strategies for treating neurodegenerative diseases.
1.2 Regenerative medicine
Physiological function of human tissues and organs damage by age, disease or trauma. The demand for organ transplantation and artificial organ rapidly increased all over the world during the past decade due to accident and congenital diseases. However, the unavailability of adequate organs for transplantation to meet the existing demand has resulted in major organ shortages crises. As a result, the number of patients who want organ transplantation are increasing. Worst of all patients dying while on the waiting list. Even patients find matching donors, there are still high possibility that transplanted organ rejected by own body.
Regarding artificial organ, as their function’s are still incomplete, periodically maintenance is needed. There are also many physical complications that goes with artificial organ; the patient is tends to get infections and device malfunction. Therefore, we are forced to look elsewhere in an effort to meet the need.
Even though great progress has been made in traditional medicine, there is few effective methods to treat diseases and condition mentioned above. In many cases, traditional treatments can only provide medications or devices, which only can temporary relief
patience’s symptom. Regenerative medicine is a branch of translational research, which has the great potential to deals with engineering or regenerating human cells, tissues or organs to recover or form normal function (Figure 1-1) [2]. It opens a new research avenue by taking advantages of the body’s natural ability to heal itself, offers realistic way to restore diseased, injured tissues and whole organs. Regenerative medicine could regenerate native-like cells, tissues and organs [3], which could effectively solve the raising problem like organ shortage, organ transplant rejection, etc [4]. By this, it can not only relieving patience’s symptoms, but addressing the root cause of pain as well.
Figure 1-1 Process of regenerative medicine
1.3 Stem cells
Stem cells defined as undifferentiated cells that be able to self-renewal through replication, as well as differentiate into other cell lineages. These great characteristics make stem cells become a frontline source of regenerative medicine, and repair tissues and organs due to congenital defects, disease and age-related effects [2].
Figure 1-2 Types of stem cells
According to the basis of differentiation potential, stem cells could divide into four types:
unipotent, multipotent, totipotent and pluripotent stem cells (PSCs) (Figure 1-2) [5].
Unipotent stem cells can differentiate along only one lineage. Compare with other types of stem cells, unipotent stem cells has the lowest differentiation potential; Totipotent stem cells can develop into any of cell types found in an embryo as well as extra-embryonic cells.
Zygote is the only totipotent stem cell in human body; Multipotent stem cells can
differentiate into a limited number of cell types in a particular lineage; Pluripotent stem cells can form tissues from all three embryonic germ layers (endoderm, mesoderm, and ectoderm) [5].
Embryonic stem (ES) cells are pluripotent stem cells, which derived from the inner cell mass of a blastocyst (Figure 1-3). ES cells are distinguished by their ability of differentiate into any cell types and by their ability to propagate. Mouse ES (mES) cells were first isolated in 1981 from mouse blastocyst and had been utilized as cell source for studying early
development of cell differentiation [6].
Figure 1-3 Establishment of ES cells
mES cells were originally maintained it’s undifferentiation state by culture on mouse embryonic fibroblasts (MEFs) as feeder cells and cytokine leukaemia inhibitory factor (LIF) in culture medium [7]. Human ES (hES) cells were first isolated in 1988 from blastocysts produced by in vitro fertilization (IVF) [8]. Although, mES cells and hES cells showed similar development potential, however, there is significant differences exist between the two cell lines [8]. hES cells are more valuable for study early human embryology and developing cell replacement strategies for the treatment of human diseases. However, because
establishment of hES cells involves the destruction of human embryos, hES cell research is ethically controversial. Furthermore, ES cells are incompatible with the immune system of patients [9].
In order to overcome these deficiencies, scientists tried to develop a variety of techniques to reverse somatic cells into stem cell-like state. In 2006, Takahashi and Yamanaka
discovered that reprogramming of mouse fibroblasts back to iPS cells could be obtained by introducing four pluripotency-associated transcription factors – Oct3/4, Sox2, c-Myc and Klf4 (Figure 1-4) [10]. Subsequent reports demonstrating iPS cells were highly similar to ES cells [11]. In 2007, direct reprogramming was achieved in human cells, which provides great contribution to the field of regenerative medicine [12].
Figure 1-4 Establishment of iPS cells
The development of ES and iPS cell research makes cell transplantation a promising therapy for the diseases of central nervous system, including traumatic brain injury and neurodegenerative disorders like AD, PD [13]. PD is considered as the best candidate for the replacement therapy because only dopaminergic neurons are affected. The first attempt of cell replacement for PD was to use embryonic mesencephalic tissue, and the results were
successful [14]. Recent research reviled dopaminergic neurons derived from ES cells are functional after grafted into parkinsonian rats [15]. Regarding other degenerative diseases, such as AD, various cell types are involved, including neurons, astrocytes, oligodendrocytes etc. Which makes cell replacement therapy is more complicated for AD compare with PD.
Recent study reviled neurons and glial cells could be differentiated from iPS cells in vitro, and transplantation of iPS cell derived neurons into brain improved the behavior in PD rat model [16]. The distinct advantages of hiPS cells, such as they can be patient specific, makes hiPS cells had been widely use in regenerative medicine, disease modeling and drug
screening.
1.4 Research purpose
Investigating neural differentiation of pluripotent stem cells, including iPS, is of important for studying early neural development and providing a potential cell source for nerve
regeneration. Stromal cell-derived inducing activity (SDIA) using mouse stromal PA6 cells promotes neural differentiation of iPS cells. Thus, we hypothesized that cadherin gene- engineered PA6 feeder cells will enhance the performance of SDIA by facilitating cell-cell interactions. First, we created Dox inducible cadherin gene-engineered PA6 cells. Efficiency of neural and motor neuron differentiation of mouse iPS cells on genetically modified feeder cells was significantly higher compared with parental PA6 cells. In this experiment, Dox, which derived from tetracycline, need to add to the medium to keep cadherin over-expressing in feeder cells. Besides, Dox is an antibiotic and it may cause toxicity to mammalian cells when used at high concentration. Therefore, we constructed constituently cadherin expressing feeder cells to improve neural differentiation of both mouse and human iPS cells by SDIA method. The purpose of this study is, whether cadherin gene transfer to PA6 cells can improve SDIA and induce accelerated neural differentiation of both mouse and human iPS cells.
1.5 Thesis components
In Chapter 1, general knowledge about neurodegenerative diseases, and the most promising way to cure these diseases- regenerative medicine, particularly stem cells, are introduced. In addition, the research purpose and strategy of the study in this thesis are briefly introduced.
In Chapter 2, techniques and mechanisms related to the study in this thesis are reviewed.
In Chapter 3, a study on neural differentiation of miPS cells by co-culturing with genetically modified cadherin over-expressing cells is demonstrated, which feeder
significantly improved neural differentiation efficiency of miPS cells compared with parental PA6 cells.
In Chapter 4, mouse and human iPS cells were co-cultured with an improved feeder cells, constitutively E-cadherin expressing cells. New feeder showed advantages over previous feeder on enhancement of neural differentiation efficiency of both mouse and human iPS cells.
In Chapter5, the contents of this and future perspective are summarized.
Chapter 2
Background
With the discovery of ES/iPS cells, it is possible to induce differentiation of ES/iPS cells into different types of adult tissues. Especially, the neural differentiation of ES/iPS cells hold great promise in regenerative medicine for the treatment of neurodegenerative diseases.
However, the neural differentiation of ES/iPS cells lack of optimal protocol. Therefore, investigation of a novel technique for the efficient neural differentiation of ES/iPS cells is of important significance. In this chapter, ES/iPS cells neural differentiation and general
methods which were used for neural differentiation, transmembrane protein-cadherin that was critical to improve neural differentiation in the study were reviewed.
2.1 Neural differentiation of iPS cells
In order to understand early neural development and also offer a potential source of cells for nerve regeneration, investigating neural differentiation of ES/iPS cells is significantly important [17]. Animal ES/iPS cells help us to establish cellular models of neurodegenerative conditions, particularly those lack appropriate animal models. Furthermore, at the
undifferentiated stage, human ES/iPS cells cannot be deployed into patients because of their tumorigenic potential [18]. Therefore, developing an efficient stepwise differentiation protocols for directing ES/iPS cells into neural lineages is an essential prerequisite for both therapeutic and basic research application. Here we describe general methods for ES/iPS cells neural differentiation.
2.1.1 Neurons
Neuron, which is the basic functional unit of the brain and nervous system, responsible for receiving signals or information from the external world, for integrate incoming signals and communicate signals to target cells. The generation of new neurons is termed as
neurogenesis. Usually a neuron has three main parts including a cell body (or soma), dendrites and axon (Figure 2-1). The nucleus of the neuron is located in the cell body. Most of the neuronal proteins are synthesized in the cell body as well. There are various processes extend from the cell body, including dendrite and axon. Dendrite, which refers to the shorter branching processes, receive and process incoming signals; axon, which refers to a separate process typically longer than the dendrites, transmit signals to different neurons, muscles and glands. Synapse refers to the contact region where a neuron transferring information to another cell. The term neurite, which is any projection from the cell body of a neuron, usually used when it refers to immature or developing neurons, especially the cells in culture.
Figure 2-1 Neuron structure
According to the role of neurons, they can be divided into three types: sensory neurons, motor neurons, and interneurons (Figure 2-2). Sensory neurons responsible for converting external or internal information and convert it into internal electrical impulses. Motor neurons, which are located in the spinal cord and the brain, are responsible for transmitting information to muscle and enabling muscle contraction. Interneuron refers to a broad class of neurons found integrative areas of the CNS, they usually create neural circuits to enabling the communication between sensory neurons or motor neurons, and the CNS.
Figure 2-2 Classification of neurons based on their role
Neurons can also be anatomically characterized based on the number of processes extend out from the cell body, they are: multipolar neurons (have three or more processes), bipolar neurons (have only two processes), unipolar neurons (have a single, short process) (Figure 2- 3).
Figure 2-3 Classification of neurons based on their structure
Base on the neurotransmitter production, neurons can also be divided into cholinergic neurons (release acetylcholine), GABAergic neurons (release gamma aminobutyric acid), glutamatergic neurons (release glutamate), dopaminergic neurons (release dopamine) and serotonergic neurons (release serotonin).
As neurons are post-mitotic and lake of regeneration ability might expose them to
accumulation of toxic substances and damaged organelles inside of the neurons which could remaining through future cell division in replicating cells [19]. Neurodegenerative diseases, including AD, PD, Huntington’s disease (H), amyotrophic lateral sclerosis (ALS), are caused by the loss or dysfunction of specific types if neurons. For example, AD caused by loss of hippocam-pal regions pyramidal neurons (type of multipolar neuron) loss [20,21]; one of the cause of PD is degeneration of dopaminergic neurons [22]; HD caused by loss of the
neostriatum medium spiny neurons (a special type of inhibitory GABAergic neurons)
[23,24]. Therefore, it is essential to investigate effective neural differentiation of ES/iPS cells towards specific neuron types.
2.1.2 General methods for pluripotent cells neural differentiation
The capacity of ES/iPS cells to keep self-renewal and differentiate to any of the three germ layers make ES/iPS cells a promising cell sources for studying disease mechanisms and for drug screening in the file of neurodegenerative diseases. Until now researchers have been demonstrated various techniques to in vitro differentiate animal and human ES/iPS cells to different types of neuronal cells. Here the most common methods which are using for neural differentiation of ES/iPS cells were reviewed.
One of the popular approach to initiate neural differentiation of ES/iPS into specific cell lineage is the use of embryoid bodies (EBs) formation. EB is a three-dimensional aggregate formed in suspension of ES/iPS cells. With longer culture duration, EB is capable of turning into a multilayered structure contains a mixed type of cells including neural cells. EBs can not only differentiate into other cell types, but also can provide an in vitro model system to understand the early lineage fate determination and organogenesis in mammalian [25]. Both glial cells and neurons could be generated by high concentrations of Retinoic acid (RA) induced neurosphere production from murine and human EBs [26-28]. RA is a morphogen, which derived from retinol (vitamin A), responsible for patterning of the neural plate and neural tube during embryogenesis [29,30]. Furthermore, RA plays role in mammalian nerve regeneration [31] and is the most utilized morphogen to generate neurons from stem cells [32]. However, one of the drawback of EB formation is presence of numerous cell types in the final product, including non-neuronal subtypes and undifferentiated cells [33].
Another most used approach to induce neural differentiation of ES/iPS cells is the addition of small molecules (SMs). SMs can maintain the self-renewal state of ES/iPS cells and also induce lineage-specific differentiation of ES/iPS cells. Neural differentiation of ES/iPS cells rely on the interaction of activation and inhibition of multiple signaling pathways which controlled by growth factors and cytokines. By selectively activate or inhibit the specific signaling pathways SMs could induce differentiation of ES/iPS cells towards neuronal
lineage cells. Further understanding of signaling pathways guided researchers designed neural differentiation protocols. Some researchers showed crucial role for BMP/SMAD signaling during neural induction [34,35]. Bone morphogenetic proteins (BMPs) are a group of growth factors, they have an important role during embryonic development on the
embryonic patterning and early skeletal formation. Pera M F et al., found by blocking BMP pathway using SMs BMP-2 and its antagonist Noggin, neural precursor induction of hES cells was increased [34]. SMADs is a family of structurally similar proteins which are the main signal transducers for receptors of the transforming growth factor beta (TGF-beta) superfamily. They are critically important for regulating cell growth and development.
Chambers SM et al., showed by the inhibition of SMAD signaling pathway using small molecules Noggin and SB431542, hES cells neural induction were dramatically increased [35]. Researchers found Wnt signaling pathway also plays important role in neurogenesis, some Wnt family members expressing during the CNS development. Yuko M et al., showed Wnt-3 promoted neural stem cells into neuronal and glial cells [36]; Besides, Chambers S M., found Wnt signaling activation under dual SMAD inhibition generated more neurons [35].
Furthermore, researchers hypothesis that exogeneous expression of a critical transcription factor (TF), for example Nurr1, could accelerate the differentiation of ES cells toward neural cell lineages [38]. By constitutively express Nurr1 in mES cells, proportion of dopaminergic (DA) neurons was increased 4- to 5-fold [37].
Regulation of ES/iPS cell development relies on cell-cell and cell-matrix interactions that involved in guiding the extracellular matrix (ECM). It has been proved that both natural or synthetic ECM is critical in regulating the proliferation, survival and differentiation of neural cells [38]. Jaroslaw Czyz et al., found ECM signaling could influences the fate determination of ES/iPS cells [39]. Aniruddh Solanki et al., showed various patterns of ECM would also modulate the neuronal and glial differentiation of NSCs even without SMs and exogenous transcription factors [40].
Kawasaki et al., demonstrated mouse stromal PA6 cells could remarkably induce neural differentiation of mES cells, and this method has been termed as stromal cell derived inducing activity (SDIA) (Figure 2-4) [41]. Stromal cells are loose connective tissue cells existed in many organs, such as bone marrow. They provide matrix-support to other cells in the organ. The SDIA method shows their advantages over other methods, especially in terms of its technical simplicity and high efficiency. Furthermore, Mizuseki et al. demonstrated SDIA-based methods for systematic induction of the neural crest, and dorsal-most CNS and ventral-most CNS differentiation [42]. The sonic hedgehog (Shh) protein addition to the SDIA method promotes ventral CNS tissue differentiation, whereas RA addition induces motor neurons differentiation [42].
Figure 2-4 SDIA method
2.1.3 Stromal cell derived inducing activity (SDIA)
Kawasaki et al., screened various feeder cells (such as MDCK, MEF, COS, OP9, and NIH3T3 cells) for mES cells neural induction. As the result, mouse stromal PA6 cells remarkably induced neural differentiation of ES cells [41]. Even after fixation of PA6 cells with paraformaldehyde (PFA), it still remained SDIA [41], which suggesting that SDIA might mediated through the ECM of the cell surface and /or membrane-related soluble
growth factors. Furthermore, while PA6 cells are physically separated from co-culture mES cells using the filter membrane of cell culture insert, PA6 cells are still capable of induce neural differentiation of mES cells, which suggesting that PA6 cells produce soluble growth factors for neural differentiation [41]. However, PA6-conditioned medium was failed to induce neural differentiation of mES cells [41] and its mechanism still unknown.
Mizuseki et al. demonstrated SDIA-based methods for systematic induction of the neural crest, and dorsal-most CNS and ventral-most CNS differentiation [42]. The sonic hedgehog (Shh) protein addition to the SDIA method promotes ventral CNS tissue differentiation, whereas RA addition induces motor neurons differentiation [42]. Shh was originally
determined as a morphogen which has critical role in neural development in embryogenesis.
Later it was found that Shh signaling plays important role in regulating developmental process in the nervous system [43,44]. Furthermore, in order to investigate the potential application of SDIA method in medical research, same team demonstrated the neural induction of ES cells, which derived from primate blastocysts, using SDIA method [45].
2.1.4 Feeder cells
Puck and Marcus demonstrated the first use of feeder cells to support large-scale colony production in 1955 [46] and it is different from traditional co-culture system because only one cell type is allowed to proliferate. Feeder cells not only support target cells by releasing growth factors, but also support by culture medium detoxification or provide extracellular matrix protein for cell growth [47]. Besides, feeder cells support target cells via paracrine interaction (e.g., cytokine release) and/or juxtacrine interactions (e.g., integrin/cadherin binding) [48]. Even though, paracrine interactions could be mediate by using conditioned medium instead of feeder cells, but juxtacrine interactions can only be achieve by direct contact of target cells and feeder cells.
ES/iPS cell technology offers great promise to regenerative medicine and disease
modeling, however, they are not able to maintain original characteristics in monoculture on
standard tissue culture dishes without any supporting factors. In vitro, ES/iPS cells typically need feeder cells, which provide adhesion molecules and extracellular matrix (ECM), to maintain undifferentiated state or induce differentiation of ES/iPS cells. Since the
establishment of first mouse ES cells, mitotically inactivated fetal mouse fibroblasts have been used as maintaining undifferentiated state of mouse ES cell culture [6]. It has been proved including Leukemia Inhibitory Factor (LIF) and other feeder cells could maintain the undifferentiated state od ES/iPS cells.
The most commonly used feeder cells to support undifferentiated status of ES/iPS cultures is mitotically inactivated mouse embryonic fibroblast (MEF) cells. However, one drawback of using primary MEF cells is their limited proliferation ability, which needs repeated isolation from pregnant mice embryos to support feeder cells [49]. STO cells, which isolated from Sandoz inbred mouse (SIM)-derived fibroblasts [50], used as an alternative cell source of MEF cells. STO cells usually used as alternative cell source of MEF cells. However, the iPS cells undifferentiation state maintaining ability of STO cells in subordinate to MEF cells [51]. The detailed mechanism underlying maintenance the undifferentiated state of ES/iPS cells on a MEF feeder layer remains unclear.
Regarding neural differentiation of stem cells, mouse stromal cell line PA6 has been proved to induce dopaminergic differentiation of both mouse ES [41] and human ES cells [52]. This ability of PA6 cells has been termed as stromal cell-derived inducing activity (SDIA) [41].
PA6 cell is a stromal cell line derived from newborn mouse calvarium. Even though PA6 cell culture contains mainly preadipocytes, they also contain a heterogeneous population of cells, including preadipocytes, spontaneously differentiated adipocytes, fibroblasts, muscle cells, osteoblasts and endothelial cells [53]. It is also found that the level of SDIA varies depending on batches of PA6 [54]. Therefore, it has been hypothesized that only some types of cells in the PA6 cell cultures trigger differentiation of ES cells into DA neurons [55].
2.2 Cadherins
2.2.1 Cadherin subtypes
Cell-cell adhesion is an important factor involved in cell survival, proliferation and differentiation. In multicellular organisms, cell-cell adhesion occurs both in homotypic and heterotypic way. For example, homotypic cell-cell adhesion could be found between two neighboring epithelial cells and they have almost no intercellular spaces. In epithelial cells, cell-cell adhesion is controlled by abundant specialized junctions, including tight junctions, adherens junctions and gap junctions. Among all these bonds, adherens junctions are prior to other junction. It is known that when adherencs junction were collapsed other junction also collapse. Here, cadherin is known as adhesion protein has central role in adherens junction [56,57].
Furthermore, regulation of stem cell development depends on cell-cell and cell-matrix interactions that are involved in guiding the adhesion molecule signaling [58]. Among cell adhesion molecules (CAM), cadherins are a class of type-1 transmembrane homophilic adhesion proteins, which play important roles in cell adhesion, colony formation and differentiation of iPS cells (Figure 2-5). Extracellular cadherin domains mediate cell-cell adhesion, whereas the intracellular cytoplasmic tail associates with a large number of
signaling proteins and adaptors. It has been observed that cells containing a specific cadherin subtype tend to cluster together [59]. For example, cells expressing E-cadherin tend to cluster with other E-cadherin-expressing cells. Later several studies observed different types of cadherin also binding together [60,61].
Figure 2-5 Cadherin
The cadherin molecules at adherent junctions have multiple isoforms, which can be divided into four groups according to their sequence similarity: classical, demosomal, protocadherins, and unconventional. The first identified cadherin subtype is classical cadherins, their structure/function studies provided detailed understanding into molecular regulation of cadherins. Classical cadherins subdivided into type and type . They were originally named on the basis of the tissues within where they were first identified. Type classical cadherins include E (epithelial)-cadherin and N (neuronal)-cadherin; Type classical cadherin include VE (vascular endothelial)-cadherin and K (kidney)-cadherin.
Among them, E-cadherin and N-cadherin are involved in ES/iPS cells pluripotency and neural cell development, respectively [62].
Karpowicz et al., demonstrated the importance of E-cadherin adhesion for neural stem cell differentiation, in which neuronal differentiation was reduced by blocking cell-cell
interactions through E-cadherin [63]. In contrast, Amrabul et al., reported that N-cadherin
reduced Rho/ROCK activation and β-catenin expression, which led to stimulation of neurite outgrowth [64]. Furthermore, neural differentiation of miPS cells is reportedly associated with an E- to N-cadherin switch [65,66].
2.2.2 Cadhrein expressing feeder cells
Haque et al., established a cadherin-based artificial ECM substrate by immobilizing both E-cadherin and N-cadherin. As the result, they demonstrated specific differentiation of mouse ES/iPS cells into neural cells [66]. Alternatively, cell-cell interactions between ES/iPS cells and co-cultured cells are thought to have major implications for tissue differentiation [67,68].
We previously reported that magnetic accumulation of magnetically labeled STO feeder cells onto mouse ES/iPS cells improves undifferentiated growth of pluripotent stem cells [69,70].
Accordingly, we demonstrated improved performance that was comparable to a conventional mouse embryonic fibroblast (MEF) feeder layer [69,70]. These observations suggest that physical contact (specifically, tight and close) between target cells and feeder cells is
essential for efficient provision of growth factor signals. Therefore, it has been hypothesized that cadherin gene-engineered feeder cells will enhance the performance of co-culture by facilitating cell–cell interactions.
2.3 Neuromuscular junction (NMJ)
Synapses are fundamental units in nervous system in which presynaptic membrane release neurotransmitter molecules and activate postsynaptic membrane receptors, by this
establishing neuronal communication. Synaptic transmission plays important role in brain function and motor control. The neuromuscular junction (NMJ) is a chemical synapse between motor neurons and skeletal muscle fibers, and is covered by Schawnn cells (SCs) (Figure 2-6). While action potentials arriving, acetylcholine (ACh), which activates muscle fibers’ ACh receptors (AChRs) and induce calcium release to initiate muscle contraction, release from motor neuron terminals. Synaptic dysfunction, including NMJs dysfunction possibly initiates MN disease progression [71,72].
Figure 2-6 Neuromuscular junction
Development of NMJ requires the close communication among presynaptic motor neurons, postsynaptic muscle fibers and SCs. In seeking to generate in vitro effective culture models, which could correctly demonstrate the physiology and function of in vivo tissues, there is an urgent need to develop platforms where the main cell types could interact accurately.
Therefore, in order to effectively model skeletal muscle, generated tissue should enable to demonstrate biologically correct representations of the NMJs. The successfully developed skeletal muscle models could be used as tool for disease modelling, drug screening and mechanistic studies [73].
It has been shown ES/iPS cell-derived MNs exhibited molecular and physiological characteristics associated with mature MNs [74,75]. Furthermore, while transplanted these MNs into peripheral nerve of mice, they seem to enable of extending axons towards
peripheral muscles [76]. Harrison R G et al., first showed the co-culture of nerve and muscle tissue [77]. Umbach J A et al., developed a low-density co-culture system which induce the active NMJ formation between mES cells derived MNs and C2C12 myotubes in vitro [78].
2.4 Conclusion
As the elderly population has increased, the age-related neurodegenerative diseases are becoming increasingly prevalent. However, neurons in the CNS lack of regenerative ability.
The capacity of ES/iPS cells to keep self-renewal and differentiate to any of the three germ layers make ES/iPS cells a promising cell sources for studying disease mechanisms and for drug screening in the file of neurodegenerative diseases.
Therefore, the investigation of efficient neural differentiation method is of important significance. Among the general methods for ES/iPS cells neural differentiation, SDIA method shows advantages like technical simplicity and high efficiency. Hence, SDIA method could be expected to provides source for the drug screening and studying early
neurodegeneration.
Chapter 3
Neural differentiation of mouse induced pluripotent stem cells using cadherin gene-engineered PA6 feeder cells
3.1 Introduction
Investigating neural differentiation of pluripotent stem cells, including induced pluripotent stem iPS cells, is of important for studying early neural development and providing a
potential source of cells for nerve regeneration. Kawasaki et al., screened various cells (such as MDCK, MEF, COS, OP9, and NIH3T3), and demonstrated that mouse stromal PA6 cells induced remarkably efficient neural differentiation of mouse ES cells, which termed as stromal cell derived inducing activity (SDIA) [41]. SDIA method offers advantages over other methods, especially in terms of its technical simplicity and high efficiency. We hypothesized that cadherin gene-engineered PA6 feeder cells will enhance the performance of SDIA by facilitating cell-cell interactions.
In our previous study, the idea of genetically modified feeder cells was applied in
maintaining of undifferentiation state of mouse ES/iPS cells. The STO cells was genetically modified to express E-cadherin and LIF (anti-differentiation factor) [69,70]. STO/EL cells were established using plasmid vector with an expression cassette for E-cadherin and LIF [70]. In the present study, inducible E- or N-cadherin expressing PA6 feeder cells (designed as PA6/E or PA6/N) were generated by incorporation of the Tet-On system into the retroviral vectors. Then whether genetically modified feeder cells can improve SDIA and induce accelerated neural differentiation of mouse iPS (miPS) cells was investigated.
Tet-On system is a regulatory system, which allows activation of gene expression by the addition of doxycycline (Dox) [79]. As illustrated (Figure 3-1), transcription factor rtTA is capable of binding the operator only if it is bound by tetracycline. Therefore, the introduction of Dox to the system initiates the transcription of the genetic product. Compare with other inducible mammalian expression systems, gene regulation in the Tet-On Systems is highly specific and fast in response times.
Figure 3-1 Dox inducible gene expression system
In present study we both early neural differentiation and motor neuron differentiation of mouse iPS cell by SDIA method were detected. Motor neurons are differentiated cells which maintain control over voluntary actions of vertebrates’ body. Pluripotent stem cell-derived motor neurons provide a promising tool in disease modeling, drug screening and
development of therapeutic approaches towards motor neuron diseases (MNDs) and spinal cord injuries [80,81]. In this study, motor neurons were successfully differentiated from miPS by the combination of SDIA method and addition of purmorphamine, RA
(SDIA/purmorphamine/RA treatment). After motor neuron formation, it’s functionality was
determined by neuromuscular junction (NMJ) formation while co-culture with C2C12- derived myotubes.
3.2 Material and Methods 3.2.1 Cell culture
Mouse embryonic fibroblast (MEF) cells were isolated from fetuses of 14-d pregnant BALB/c mice and cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and antibiotics (100 U/mL penicillin G potassium and 50 μg/mL streptomycin sulfate). Animal experiments were approved by the Ethics Committee
for Animal Experiments of Faculty of Engineering, Kyushu University (A28-173-1).
Undifferentiated miPS cells (Ips-MEF-Ng-20S-17 iPS cell line; Riken BioResource Center, Ibaraki, Japan)[74] were maintained on mitotically inactivated MEF feeder cells, which were treated with 1 μg/mL mitomycin C for 2 h. Undifferentiated iPS cells were maintained in iPS medium composed of Knockout-DMEM™ (Invitrogen, Carlsbad, CA, USA) supplemented with 4 mM L-glutamine, non-essential amino acids (Invitrogen), 100 μM 2-mercaptoethanol, antibiotics, 15% Knockout Serum Replacement (Invitrogen), and 1,000 U/mL leukemia inhibitory factor (LIF) (ESGRO™; Millipore, Billerica, MA, USA). Mouse stromal PA6 cells (Riken BioResource Center) were grown in alpha-MEM (Invitrogen) supplemented with 10%
FBS and antibiotics. Mouse myoblast C2C12 cells (American Type Culture Collection, Manassas, VA, USA) were grown in DMEM supplemented with 10% FBS and antibiotics (i.e., myoblast growth medium). To induce myogenic differentiation, the medium was changed to DMEM supplemented with 2% calf serum and antibiotics (i.e., myogenic differentiation medium). Cells were cultured at 37°C in a 5%(v/v) CO2 incubator.
3.2.2 Construction of E-cadherin or N-cadherin gene-engineered PA6 cells
The Tet-On system (Clontech, Mountain View, CA, USA) was incorporated into retroviral vectors for inducible expression of the E- or N-cadherin gene. Fragments of full-length murine E-cadherin [83]or N-cadherin [84] cDNA were obtained from the RIKEN
Bioresource Center, and ligated into pQMSCV/EGFP-TREtight-LETF-WPRE vector [85] to generate cDNApQMSCV/EGFP-TRE-Ecadherin-WPRE or pQMSCV/EGFP-TRE-
Ncadherin-WPRE, respectively. For retroviral vector production, two retroviral vector plasmids were used: pQMSCV/EGFP-CMV-rtTA-WPRE [85] and either pQMSCV/EGFP- TRE-Ecadherin-WPRE or pQMSCV/EGFP-TRE-Ncadherin-WPRE. pQMSCV/EGFP-CMV- rtTA-WPRE encodes a constitutive expression cassette for a transactivator (rtTA) that is activated by addition of doxycycline (Dox), while pQMSCV/EGFP-TRE-Ecadherin-WPRE and pQMSCV/EGFP-TRE-Ncadherin-WPRE include a tet-responsive element and encode expression cassettes for the E- and N-cadherin gene, respectively (Figure 3-2). Consequently, cadherin gene expression is induced via rtTA activation by Dox addition. Furthermore, these plasmids include an enhanced green fluorescent protein (EGFP) gene under control of the viral long terminal repeat (LTR) promoter. Retroviral vectors pseudo-typed using vesicular stomatitis G protein (VSV-G) were produced by transient transfection of 293FT cells with Lipofectamine 2000 reagent (Life Technologies, Carlsbad, CA, USA) and the retroviral vector plasmids, pcDNA4-gag/pol and pLP/VSV-G.
Figure 3-2 Retroviral vector constructs encoding for the rtTA, E-cad, and N-cad genes
For retroviral infection, PA6 cells were cultured for 24 h, and then the medium was replaced with retroviral solution containing retroviral vectors encoding the rtTA gene and inducible cadherin gene expression cassettes. Cells were then cultured in the presence of polybrene (Sigma–Aldrich, St. Louis, MO, USA) for 7 h, resulting in generation of PA6 cells capable of Dox-inducible E- or N-cadherin gene expression: PA6/E or PA6/N cells,
respectively. Viral titers against PA6 cells were determined by flow cytometry using a cell sorter (SH800; Sony, Tokyo, Japan). Viral titers were approximately 1.2 × 106 IU/mL (multiplicity of infection, 40).
3.2.3 Immunocytochemical analysis
Cells were fixed with 4% paraformaldehyde (PFA) for 15 min at room temperature, permeabilized with 0.2% Triton-X-100, washed three-times with phosphate buffered saline (PBS), and blocked in 1% bovine serum albumin solution in PBS for 30 min. Primary antibodies were added and samples were incubated for 3 h. After washing three-times with PBS, samples were incubated with Alexa Fluor 488- and/or 546-conjugated secondary antibody (Life Technologies) for 45 min followed by washing three-times with PBS. The following primary antibodies were used: anti-E-cadherin monoclonal antibody (36/E- Cadherin; BD Biosciences, San Jose, CA, USA), anti-N-cadherin monoclonal antibody (H- 63; Santa Cruz Biotechnology, Dallas, TX, USA), anti-neuron specific β-III tubulin (TuJ) monoclonal antibody (TuJ-1; R&D Systems, Minneapolis, MN, USA), anti-HB9 polyclonal antibody (bs-11320R; Bioss, Boston, MA, USA), and anti-sarcomeric α-actinin antibody (EA-53; Sigma–Aldrich). For visualization of acetylcholine receptors (AChRs), cells were incubated with Alexa Fluor 643-conjugated α-bungarotoxin (Life Technologies) for 1 h. In some experiments, samples were also stained with 4',6-diamidino-2-phenylindole (DAPI).
Cells were observed using a BZ-9000 fluorescence microscope (Keyence, Osaka, Japan) or Fluoview FV10i confocal laser-scanning microscope (Olympus, Tokyo, Japan).
3.2.4 Neural differentiation
Neural differentiation of miPS cells based on the SDIA method was performed according to a published method [41]. PA6/E(+) and PA6/N(+) cells were prepared by adding Dox (1 μg/mL) at two days before the beginning of neural differentiation culture (Day −2). On Day
−1, parental PA6 cells, PA6/E cells without Dox addition (PA6/E[−]), PA6/N cells without
Dox addition (PA6/N[−]), PA6/E(+) cells, or PA6/N(+) cells were seeded (2.0 × 105 cells/well) as feeder cells onto wells of collagen-coated 6 well-plates (AGC Techno Glass, Shizuoka, Japan). On Day 0, iPS cells were seeded (1.1 × 103 cells/well) onto confluent feeder cells and cultured in iPS medium without LIF addition (i.e., neural differentiation medium). The medium was replaced every other day until Day 6. Cell clusters with diameters
> 100 μm were defined as a colony. On day 6, TuJ-positive (TuJ+) colonies were counted in five fields of view from three individual wells per sample. Single colony areas were
measured using BZ-II measurement module software (Keyence). Neural differentiation efficiency was determined by the following equation: Differentiation efficiency [%] = (number of TuJ+ colonies) / (number of initially seeded cells) × 100.
For long-term culture, a papain dissociation system (Worthington Biochemical, Lakewood, NJ, USA) was used for detachment of iPS cell-derived colonies. On Day 6 of neural
differentiation culture, cells were detached from the culture surface and pre-plated twice to remove feeder cells. Next, iPS-derived cells were passaged to freshly prepared feeders (parental PA6 cells, PA6/E[+] cells, or PA6/N[+] cells) and co-cultured in neural
differentiation medium with N2 supplement (Thermo Fisher Scientific, Waltham, MA, USA) until Day 10. TuJ+ area in each well was measured using BZ-II measurement module
software (Keyence).
For differentiation of miPS cells to motor neurons, 0.2 μM RA (Wako Pure Chemical Industries, Osaka, Japan) and 1 μM purmorphamine (Sigma–Aldrich) (Shh signaling pathway agonist [86]) were added to neural differentiation medium on Day 4 [55], and cells cultured
until Day 9. The number of both TuJ- and HB9-double positive (TuJ+HB9+) colonies was counted in five fields of view from three individual wells per sample. Single colony areas were measured using BZ-II measurement module software (Keyence). Differentiation efficiency of iPS cells to motor neurons was determined by the following equation:
Differentiation efficiency [%] = (number of TuJ+HB9+ colonies) / (number of initially seeded cells) × 100.
3.2.5 Co-culture assay with myotubes
On Day 8 of motor neuron differentiation, iPS cell colonies were detached from feeder cells by papain treatment and pre-plated onto 100-mm collagen coated dishes (AGC Techno Glass) for 1 h. Next, 3.2 × 105 of these cells along with 3.2 × 105cells C2C12 cells were plated into 35-mm tissue culture dishes and cultured in myogenic differentiation medium. In some experiments, agrin (100 μg/mL; R&D System), a proteoglycan involved in
neuromuscular junction (NMJ) development during embryogenesis [79], was added to myogenic differentiation medium. After incubating for 4 d in myogenic differentiation medium, immunocytochemical staining of cells and measurement of contractile activity of myotubes were performed. Width of α-actinin-positive myotubes was measured from images using BZ-II measurement module software (Keyence). To estimate mean myotube width, the five largest myotubes from five fields in each of three separate wells per sample were chosen and measured. To assay contractile activity of myotubes, electrical pulse stimulation was applied to cells. Carbon electrodes were placed 18 mm apart at opposite sides of a tissue culture dish. Cells were stimulated with electric pulses for 30 min with the following properties: voltage, 0.3 V/mm; width, 4 ms; and frequency, 1 Hz [88]. After a 30-min electrical pulse stimulation period, electric pulses were applied again, and myotube movement recorded at a speed of 15 frames/s for 25 s using a BZ-9000 fluorescence
microscope (Keyence). For synapse blocking experiments, a nicotinic cholinergic antagonist, (+)-tubocurarine chloride pentahydrate (also known as curare) (50 μM; Sigma–Aldrich) [89],
was added to the medium to block AChRs present in NMJs. A single myotube was set in the captured image and tracked during electrical pulse stimulation. Displacement was analyzed using motion analyzer software (Keyence). For estimating displacement range, the three myotubes displaying highest contractile activity in each of three fields in three separate dishes were chosen and measured using motion analyzer software (Keyence).
3.2.6 Statistical analysis
Statistical comparisons were performed using the Mann–Whitney U rank sum test. Values of P < 0.05 were considered to indicate significant differences.
3.3 Results and Discussion
3.3.1 Cadherin expression in genetically engineered PA6 feeder cells
The E-cadherin or N-cadherin gene with a Dox-inducible expression cassette was introduced into stromal PA6 cells (designated PA6/E or PA6/N cells). Consequently, transgene expression was induced by Dox addition to the medium (PA6/E[+] or PA6/N[+]).
Genetically engineered PA6 cells were evaluated for E-cadherin and N-cadherin expression (Figure 3-3). Immunocytochemical staining revealed that parental PA6 cells exhibited
extremely low levels of E-cadherin expression with slight N-cadherin expression. In contrast, PA6/E(+) cells and PA6/N(+) cells exhibited greater expression levels of E-cadherin and N- cadherin, respectively.
Figure 3-3 Generation of cadherin gene-engineered PA6 cells. Immunocytochemical staining of E-cadherin and N-cadherin expression in PA6 cells with anti-cadherin antibodies. PA6, parental PA6 cells; PA6/E(−), E- cadherin gene-transduced PA6 cells without Dox addition; PA6/N(−), N-cadherin gene-transduced PA6 cells without Dox addition; PA6/E(+), E-cadherin gene-transduced PA6 cells with Dox addition; and PA6/N(+), N- cadherin gene-transduced PA6 cells with Dox addition.
3.3.2 Neural differentiation efficiency of miPS cells on cadherin gene-engineered PA6 feeder cells
miPS cells were cultured on a PA6 feeder monolayer and allowed to differentiate for 6 days (Figure 3-4). Control miPS cells in feeder-free cultures were negative for the neuron specific marker, TuJ. Meanwhile, TuJ+ colonies formed on PA6/E(+) feeders, with PA6/N(+) feeder cells being larger than parental PA6 cells and exhibiting distinctive neurite outgrowth on Day 6 (Figure 3-4A). Quantitative analysis revealed that differentiation efficiency (Figure 3-4B), the number of TuJ-positive cells (Figure 3-4C) and TuJ+ colony area (Figure 3-4D) were significantly higher on PA6/E(+) and PA6/N(+) feeders compared with parental PA6 feeder, suggesting that cadherin gene-engineered feeder cells enhance SDIA.
Further, PA6/E(+) feeder cells showed the greatest effects on both differentiation efficiency and TuJ+ colony growth. The effect of cadherin expression on neural
differentiation was further evaluated by quantifying the number of neurites formed by iPS- derived cells on PA6 feeders (Figure 3-4E). miPS cells cultured on the parental PA6 cell layer formed an extremely low number of TuJ+ colonies with neurite outgrowth, whereas both PA6/E(+) and PA6/N(+) feeders increased the number of TuJ+ colonies with more than five neurites. Specifically, the percentage of TuJ+ colonies with more than 5 neurites was 35.5 ± 4.4% and 49.9 ± 3.3% for PA6/E(+) and PA6/N(+) feeders, respectively, suggesting that PA6/N(+) feeder cells are effective for inducing neurite outgrowth. For long-term neural differentiation culture, the co-culture period was extended to Day 10. iPS-derived cells cultured on a PA6 feeder layer were reseeded onto a freshly prepared PA6 feeder layer on Day 6, and co-cultured until Day 10 (Figure 3-5). Elongated neurites were observed on Day 10 (Figure 3-5A). As shown (Figure 3-5B), TuJ+ area was significantly larger in co-cultures with cadherin gene-engineered PA6 feeders compared with the parental PA6 feeder. The most effective protocols involved iPS-derived cells cultured on PA6/E(+) feeder and reseeded
onto either PA6/E(+) feeder (Protocol E→E) or PA6/N(+) feeder (Protocol E→N) on Day 6 and co-cultured until Day 10.
Figure 3-4 Neural differentiation of induced pluripotent stem (iPS) cells on PA6 feeder cells for 6 days. (A) Anti-neuron specific β-III tubulin (TuJ) staining of iPS cells cultured on PA6 feeders. (B) Quantitative analysis of differentiation efficiency, (C) number of TuJ-positive colony, (D) single colony area, and (E) number of colonies with neurite outgrowth. Inset graph of D shows the percentage of colonies with outgrowth of more than 5 neurites. Data are expressed as mean ± SD of triplicate experiments. *P < 0.05 versus parental PA6 feeder.
Figure 3-5 Neural differentiation of induced pluripotent stem (iPS) cells on PA6 feeder cells for 10 days. (A) Protocols and anti-neuron specific β-III tubulin (TuJ) staining of iPS cells cultured on PA6 feeders. (B) Quantitative analysis for TuJ-positive area. Data are expressed as mean ± SD of triplicate experiments. *P <
0.05 versus Protocol P→P.
3.3.3 Effect of cadherin gene-engineered PA6 feeder cells on miPS cell differentiation into motor neurons
To investigate the effect of PA6/E(+) and PA6/N(+) feeder cells on differentiation into motor neurons, SDIA treatment of miPS cells was combined with addition of purmorphamine and RA (SDIA/purmorphamine/RA treatment). Cells were co-cultured until Day 9 (Figure 3- 6). A larger number of TuJ+HB9+ colonies with neurite outgrowth formed on PA6/E(+) and PA6/N(+) feeders compared with parental PA6 feeder (Figure 3-6A). Quantitative analysis revealed that PA6/N(+) feeder had the greatest effect on differentiation of miPS cells into motor neurons (Figure 3-6B) and TuJ+HB9+ colony growth (Figure 3-6C). For functional analysis, motor neurons generated with SDIA/purmorphamine/RA treatment on PA6/N(+) feeder cells were co-cultured with C2C12-derived myotubes (Figure 3-7). As shown in Myotube width was increased by co-culture with SDIA/purmorphamine/RA-induced cells (Figure 3-7A, B). In co-cultures, AChR clustering alongside TuJ+ neurites were observed (Figure 3-7C). For assay of myotube contractile activity, electrical pulse stimulation was applied, and myotube movement measured (Figure 3-7D). In co-cultures,
SDIA/purmorphamine/RA-induced cells enhanced contractile activity of myotubes. Further, agrin, an AChR clustering agent, slightly enhanced displacement of myotubes under
electrical pulse stimulation. To assess the role of the neurotransmitter acetylcholine in these observed contractions, an AChR antagonist, curare, was used. Displacement of myotubes co- cultured with SDIA/purmorphamine/RA-induced motor neurons under electrical pulse stimulation was significantly decreased by curare treatment. These results suggest that motor neurons generated with SDIA/purmorphamine/RA treatment are functional, innervating C2C12-derived myotubes and causing AChR-dependent muscular contraction.
Figure 3-6 Motor neuron differentiation of induced pluripotent stem (iPS) cells on PA6 feeder cells. (A) Immunocytochemical staining of iPS cells cultured on PA6 feeders. (B) Quantitative analysis of motor neuron differentiation efficiency and (C) single colony area. Data are expressed as mean ± SD of triplicate experiments.
*P < 0.05 versus parental PA6 feeder.
Figure 3-7 Co-culture assay with myotubes. (A) Immunocytochemical staining of C2C12 myotubes. (B) Quantitative analysis of myotube width. Data are expressed as mean ± SD of triplicate experiments. *P < 0.05 versus monoculture. (C) Co-immunocytochemical staining of α-bungarotoxin that is specific for acetylcholine receptors (AChRs), and anti-neuron specific β-III tubulin (TuJ) for neurite outgrowth. (D) Quantitative analysis
of displacement range for myotube contractile activity with (+) or without (−) agrin/curare treatment. Data are expressed as mean ± SD of triplicate experiments. *P < 0.05.
3.4 Conclusion
In conclusion, we have shown that cadherin gene-engineered PA6 feeder cells can enhance SDIA for neural differentiation of miPS cells. Further, motor neurons generated with
SDIA/purmorphamine/RA treatment are also functional. Recently, in vitro construction of NMJs has attracted attention for drug screening of neuromuscular diseases [90]. miPS- derived motor neurons induced by SDIA/purmorphamine/RA treatment using cadherin gene- engineered PA6 feeders may be useful for studying NMJ formation. Investigating neural differentiation of miPS cells is of important significance to offer a potential source of cells for nerve regeneration. Cadherin gene-engineered PA6 feeders may be a powerful tool for increasing the yield of neural cells from miPS cells for preclinical investigations.
Chapter 4
Neural differentiation of mouse and human iPS cells using constitutive E-cadherin expressing mouse PA6 feeder cells
4.1 Introduction
Investigating neural differentiation of pluripotent stem cells, including induced pluripotent stem iPS cells, is of important for studying early neural development and providing a
potential source of cells for nerve regeneration. Kawasaki et al., screened various cells (such as MDCK, MEF, COS, OP9, and NIH3T3), and demonstrated that mouse stromal PA6 cells induced remarkably efficient neural differentiation of mouse ES cells, which termed as stromal cell derived inducing activity (SDIA) [41]. SDIA method offers advantages over other methods, especially in terms of its technical simplicity and high efficiency. We hypothesized that cadherin gene-engineered PA6 feeder cells will enhance the performance of SDIA by facilitating cell-cell interactions.
In previous study, doxycycline (Dox) inducible cadherin expressing mouse stromal PA6 cells (E-cadherin expressing PA6/E cells and N-cadherin over-expressing PA6/N cells) were generated and significantly improved the performance of SDIA for neural differentiation of mouse iPS cells. Tet-On System was incorporated into retroviral vectors for inducible
expression of the E- or N-cadherin gene. Tet-On System is a regulatory system, which allows activation of gene expression by the addition of Dox [79]. Therefore, Dox need to be
consistently added to the culture medium to keep cadherin expression. However, Dox is an antibiotic and it may cause toxicity to mammalian cells when used at high concentration. In order to simplify the experimental procedure and avoid from toxicity, constitutive cadherin
expressing feeder cells was generated to induce neural differentiation of both mouse and human iPS cells.
Constitutive gene expression refers to a gene which is transcribed continually. In present study, constitutive E-cadherin expressing PA6 feeder (PA6/E) cells was generated by
construction of PiggyBac (PB) transposon vector. Transposon is a DNA fragment, which can change it’s position in the entire genome of a cell. The PiggyBac (PB) transposon, which has been widely used as a functional tool in various human and mouse cell lines [91,92], has the capacity to efficiently transposes between vectors and chromosomes by a “cut and paste”
mechanism (Figure 4-1). PB transposon provides significant advantages over other traditional methods, including simplicity to generate stable cell line, virus-free integration.
Figure 4-1 DNA transposon
Although animal models have been widely used in the investigation of disease mechanism, fundamental developmental, physiological and biochemical differences exist between mouse and human. Compare with other animal cells, hiPS cells can offer physiologically related cells to provide more efficient medicines and accurate model for drug screening. In order to investigate the potential use of SDIA method in medical research, Kawasaki et al.,
demonstrated the neural differentiation of ES cells, which derived from primate blastocysts, using SDIA method [45].
After successfully improved neural differentiation efficiency of mouse iPS cells, next hiPS cells were co-cultured with genetically modified feeder cells to induce neural differentiation of hiPS cells.
4.2 Materials and Methods 4.2.1 Cell culture
Mouse embryonic fibroblast (MEF) cells were isolated from fetuses of 14-d pregnant BALB/c mice and cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and antibiotics (100 U/mL penicillin G potassium and 50 μg/mL streptomycin sulfate). Animal experiments were approved by the Ethics Committee
for Animal Experiments of Faculty of Engineering, Kyushu University (A28-173-1).
Undifferentiated miPS cells (Ips-MEF-Ng-20S-17 iPS cell line; Riken BioResource Center, Ibaraki, Japan)[82] were maintained on mitotically inactivated MEF feeder cells, which were treated with 1 μg/mL mitomycin C for 2 h. Undifferentiated iPS cells were maintained in iPS medium composed of Knockout-DMEM™ (Invitrogen, Carlsbad, CA, USA) supplemented with 4 mM L-glutamine, non-essential amino acids (Invitrogen), 100 μM 2-mercaptoethanol, antibiotics, 15% Knockout Serum Replacement (Invitrogen), and 1,000 U/mL leukemia inhibitory factor (LIF) (ESGRO™; Millipore, Billerica, MA, USA). Mouse stromal PA6 cells (Riken BioResource Center) were grown in alpha-MEM (Invitrogen) supplemented with 10%
FBS and antibiotics.
Undifferentiated hiPS cells (201B7 iPS cell line, Kyoto University) were maintained on Laminin coated dish (Thermo). Cells were cultured in mixed medium which components by adding the full volume of “Liquid A”, “Liquid B” and “Liquid C” (StemFit AK02N,
Ajinomoto).
4.2.2 Plasmid construction
E-cadherin insert fragment was extracted from pQMSCV/EGFP-TRE-Ecadherin-WPRE (previously constructed in our lab) vector by digestion of restriction enzymes EcoR and Not . PB513-B (System Biosciences) transposon vector was linearized by digestion of restriction enzymes Swa and Not . Then E-cadherin fragment and linearized PB513 vector fragment were blunt ligated to generate PB513/Ecadherin plasmid (Figure 4-2).
PB/Ecad plasmid was transformed into competent cell (DH-5α. Cat. No. DNA-913, Toyobo) and positive colonies were selected on LB plates containing 100 μg/mL ampicillin (Wako) after incubation overnight at 37°C.
After PB/Ecad plasmid transformation, plasmid was first small-scale isolated by
Minipreparation. Briefly, 12 colonies were picked up and a starter culture of 2ml LB medium containing 100 μg/mL ampicillin was inoculated. After shaking bacteria was harvested by spinning at 12000 rpm. Then the plasmid was extracted with base on the alkaline lysis method. After the minipreparation, colonies were digested with specific restriction enzyme and run electrophoresis to confirmation. Colonies which showed expected band size were storage as glycerol stocks.
Colonies, which showed expected band after restriction enzyme digestion, were large-scale isolated. Briefly, glycerol stocks of colonies were spread on the LB plate containing 100 μg/mL ampicillin and allow to grow 14-16 hours. Single colony was picked up and a starter
culture of 2ml LB medium containing 100 μg/mL ampicillin was inoculated. Then 2ml LB medium (containing 2ul ampicillin) with bacterial clone cultured with vigorous shaking at 37 degree for 5-6 hours. The starter culture was diluted and grow at 37°C for 12-16 h with
