1
miRNAs;
exciting
new
prospects for glial cell
replacement therapies
Michael Whitehead
Supervisor: Dr Vivi Heine
2
Abbreviations
IPSC – induced pluripotent stem cells TF – transcription factors
miRNA – micro ribonucleic acid NT – nucleotides
VWM – vanishing white matter OPC – oligodendrocyte precursor cells MS – Multiple Sclerosis
UTR – untranslated region EIF – eukaryotic initiation factors GRP – glial resticted proginator cells GFP – green fluorescent protein NSC – neural stem cell
NPC – neural progenitor cell MRI – magnetic resonance imaging SPIO – superparamagnetic ion oxides MBP- myelin binding protein
HD – Huntington’s Disease ESC – embryonic stem cell
BMP- bone morphogenetic protein TGF-β – transforming growth factor beta
3
Abstract
Vanishing white matter disease (VWM) is a severe predominantly childhood disease which is characterised by dysfunctional glial cells. This immediately suggests that replacement of these glial cells could be an effective therapeutic. However this field has been hindered by the lack of an accurate disease model. The Van der Knaap and Heine labs have recently produced two mice with mutations associated with VWM. They show many of the characteristics of VWM and therefore represent an excellent model to test cell replacement therapeutics. This is supported by a number of successes in glial cell replacement for treating mouse models of multiple sclerosis (MS) and spinal cord injury. The field of cell replacement therapy is changing rapidly and it is now possible to produce pluripotent stem cells from somatic cells or even go straight from one somatic cell to a stem cell precursor or another somatic cell. However there are two significant problems in the way to translating this into the clinic. The first is that viral transfection of TF used to reprogram cells can cause tumourigenesis. The second is the relative inefficiency of producing the cells for replacement; especially glial cells. Recent studies have shown that miRNA’s have the potential to help negate both of these problems which is extremely exciting. This literature review will focus on our current knowledge of miRNAs in reprogramming somatic cells to pluripotent stem cells and trans-differentiation with a focus on producing glial cells for treating VWM.
4
Introduction
Ever since the discovery by Yamanaka (2006) that only four TFs can reprogram fibroblasts into pluripotent stem cells, there has been pronounced interest in personal cell replacement therapies. The notion that pluripotent stem cells could be made from any somatic cells is very exciting; especially blood as this represents a convenient non-invasive method. This is already starting to revolutionise disease models which has great potential for improving our understanding of many different diseases as well as in testing potential therapeutics. Already this potential has been recognised as the noble prize in medicine was awarded to Yamanaka for this discovery. Trans-differentiation offers an alternative approach to iPSC which also has great potenital, but research in this area is still in its infancy. Recent advancements have shown that these cells have the potential to treat a number of brain diseases, but there are still many issues stopping translation into the clinic. Here we will discuss the potential treatment options for VWM, a fatal disease which predominantly affects glial cells. We will then discuss the potential for miRNAs in reprogramming cells for therapies, to try and negate a number of current issues in the field including inefficiency and tumourigenesis.
Vanishing white matter
Vanishing white matter (VWM)(OMIM #603896) is a predominantly childhood disease (Hanefeld et
al 1993) which is characterised by white matter loss and diagnosed using magnetic resonance
imaging (MRI). The disease is ultimately fatal in all cases with episodes of rapid progression in response to head trauma and infections (Van der Knaap 1997). In some cases this can lead to coma where the patient may not awaken at all or will only partially recover. Affected individuals can also develop seizures and the main neurological signs are progressive cerebella ataxia and spasticity (Pronk et al 2006). VWM disease causes severe motor problems and leads to an inability to communicate (Van der Knaap et al 1997 and 98). Later onset is characterised by psychiatric disorders and cognitive decline. The affected white matter literally disappears, only to be filled with fluid, leaving behind the grey matter still relatively preserved. In cavitated areas there is severe myelin loss, axon loss and astrogliosis (Van der Knaap et al 2006). Abnormal oligodendrocytes called “foamy oligodendrocytes” develop, which are a diagnostic feature of the disease (Wong et al 2000). In some areas of the brain an increase in oligodendrocytes is observed with a conflicting increase in apoptotic oligodendrocytes, while in other areas they decrease (Van Haren et al 2004).
5
VWM is caused by mutations in the gene for eIF2B, a guanine exchange factor (GEF) which mediates the activity of eIF2. eIF2B has 5 non-identical subunits termed α-ε. Mutations have been found in all of these subunits (120 to date) which all cause VWM (reviewed in Pravitt and Proud 2009). eLF2 is very important for the initiation of translation. During cellular stress mRNA translation is inhibited through a number of mechanisms including the unfolded protein response (UPR) (Hetz 2012). Activation of the UPR has been identified in the white matter of VWM post-mortems and may contribute to the cell death seen in the disease, as chronic activation of the UPR activates cell death pathways (Hetz 2012 and Van Kollenburg et al 2006). This inhibition takes place by phosphorylating eIF2 which in turn should decrease the activity of eIF2B. However the mutations associated with VWM means this does not happen, explaining why episodes of deterioration are correlated with stress. Why this specifically affects only glial cells is still unknown.
Geva et al (2010) produced the first mouse model for VWM. This was associated with a less severe adult onset form of the disease, making the testing of potential therapeutics difficult. The Van der knaap and Heine lab subsequently produced two mice with mutations that correspond to severe phenotypes in humans, causing death two months and two years after birth (unpublished data). These mice appear normal in the first five months of life. After which they develop sporadic epileptic seizures and ataxia of the lower limbs. This ultimately leads to death after seven months. Histopathological analysis has identified a decrease in myelin in a number of brain regions including the corpus callosum. Astrocytes are increased but immature. This phenotype is extremely similar to that seen in VWM in humans which mean these models represent an excellent method for testing cell replacement therapies. It is unknown whether replacement of functional astrocyte and/or oligodendrocyte cells will be an effective treatment and whether the environment will allow for cell replacement to be an effective therapy
Stem cells of the brain
Stem cells can be found in the dendate gyrus and the subventricular zone of the brain where they can divide into neurons, astrocytes and oligodendrocytes. This division is asymmetric which allows for the production of one dividing cell and one stem cell, ensuring tissue homeostasis (Inaba and Yamashita 2012). This ability is maintained throughout adulthood but is nowhere near effective enough to stop neurodegenerative diseases on its own. Injection of neural stem cells (NSC) has been shown to treat stroke in mice, although through producing growth factors rather than cell replacement (Oki et al 2012). Human induced pluripotent stem cells (iPSC) have also been
6
differentiated into NSCs with a varying efficiency of 17-75% depending on the cell line (Hu et al 2010). This was in contrast to human ESCs which had a more consistent differentiation of 90-97%, suggesting differences between the cells. More emphasis has been placed on using neural progenitor cells (NPC). These cells are more mature forms of NSC and one of the main differences is that they are not immortalised. Immortalised cells are one of the characteristic hallmarks of cancer cells, which mean using NPCs is likely to reduce the chance of cancer formation. NPCs are still tripotent and have been especially affective for the treatment of traumatic brain injury (Bakshi et al 2006, Shear et al 2004, and Wallenguist et al 2009). In 2004 Shear et al completed one of the longest transplantation experiments using NPCs from E14.5 mice. These cells were injected into mice after traumatic brain injury which leads to functional recovery even after one year. The cells predominantly differentiated into oligodendrocyte precursor cells (OPCs), but not into oligodendrocytes. More recently NPCs have been produced from iPSC (Zhao et al 2012) although they are yet to be tested in vivo for their use in cell replacement therapy. For the replacement of oligodendrocytes more lineage restricted OPCs provide another option.
Oligodendrocyte replacement
On first thought one might presume that finding a way to produce oligodendrocytes in vitro and then injecting them into the brain would be an effective therapeutic. However mature oligodendrocytes do not have the same ability to migrate to different regions of the brain in response to chemical and physical cues as their precursors. Numerous studies have now shown that these precursors known as oligodendrocyte progenitors (OPC) have the ability to migrate and differentiate into fully functioning oligodendrocytes within the brain (reviewed in Franklin and constant 2008). These OPC come from neural progenitor cells (NPC) which can also be used for cell replacement therapy. However because they can differentiate into a number of different lineages OPCs provide a more efficient option for oligodendrocyte replacement. In 2004 Windrem et al proved that oligodendrocyte precursor cells (OPC) could be used to myelinate the shivering mouse and it is noteworthy that this was possible with OPC derived from the white matter of adult humans. Shivering mice lack the myelin basic protein (MBP) which leads to improper axonal myelination, shivering and premature death. Unfortunately although widespread myelination was observed this did not have a significant effect on the life expectancy of the mouse. It was not until 2008 that Windrem et al proved that injection of these cells rescued the phenotype of shivering mice and significantly extended their lifespan. This study showed for the first time that cell replacement of
7
oligodendrocytes is possible, but a large number of OPCs are needed. A number of studies have also attempted astrocyte cell replacement.
Astrocyte replacement
Much less research has been done into astrocyte replacement and as such we are not able to produce, in vitro, glial restricted progenitors (GRP) which can form both astrocytes and oligodendrocytes (reviewed in Noble et al 2011). The injection of GRPs into mice with spinal cord injury (SCI) leads to their differentiation into astrocytes and aids neural regeneration (Has et al 2012). OPCs can differentiate into type 2 astrocytes in vitro but astrocytes with a similar phenotype
in vivo are yet to be identified, which means NPC/NSC are currently needed for astrocyte
replacement (Baumann and Pham-Dinh 2001). It is currently not known whether injection of in vitro differentiated astrocytes would be an effective cell replacement therapy. Replacing astrocytes along side oligodendrocytes is very important for “maintaining white matter homeostasis”. In vitro astrocytes release factors into the culture medium which protect oligodendrocytes through the ERK and AKt pathways (Arai and low 2010 and Kato et al 2011). For treating VWM both oligodendrocytes and astrocytes are affected which means injection of both cell types maybe needed for functional recovery. The environment that the cells are injected into has a major affect on how successful the cell replacement will be (Tirrota et al 2012). Whether the environment would promote glial differentiation in VWM is still to be identified. ESC can be used to produce OPCs relatively efficiently (Selvaraj et al 2010 and Fadi et al 2010) but the ethical problems associated with obtaining these cells makes it unviable for the clinic. iPSCs may provide the answer.
Induced pluripotent stem cells (IPSCs)
Over the last six years there has been a revolution in our understanding of stem cells and their reprogramming for cell replacement therapies. In 2006 Yamanaka and colleagues made the startling discovery that fibroblasts can be reprogrammed to pluripotent stem cells (PSC) which they dubbed induced PSC. In 2007 along with Yu et al they showed this could be achieved in human fibroblasts. Pluripotent means the cells have the ability to change into any cell type in the body. This was a significant finding for regenerative medicine, drug research and disease research. This is because there are no longer any ethical problems with obtaining stem cells and the number of cells produced for replacement should not be as limited. In the case of genetic diseases homologous recombination can be used to fix the cells. Unfortunately this means sporadic diseases may not be treatable with
8
this technique as the replaced cells would have the same vulnerabilities, but functional recovery may still be possible. Finally because the fibroblasts can be obtained from the patient to be treated there are no issues with immune rejection (Figure 1 and reviewed in Yamanaka 2012). The significance of this discovery is highlighted by the fact that Yamanaka was awarded the 2012 noble prize in medicine. There is no reason why we should just be limited to reprogramming fibroblasts though. One of the most exciting alternative cell sources is blood, as this is a non-invasive procedure (Giorgetti et al 2010). To date iPSCs have been used to treat a number of mouse models for diseases such as β-thalassemia and Huntingtons’s disease (HD).
iPSC for cell replacement therapy
β-thalassemia (OMIM #613985) is associated with mutations which cause a reduction in haemoglobin A, causing abnormal globin production which ultimately affects erythropoiesis. The disorder is clinically heterogeneous with minor, intermediate and major disorders. In β-thalassemia major the individuals are severely anaemic, transfusion dependant and normally die by the age of 35, in the UK (Modell et al 2000). Other complications include diarrhoea, fever and spenomegaly,
Figure 1 Induced pluripotent stem cells for cell replacement therapies
A number of different somatic cells can be taken from a patient and reprogrammed into pluripotent stem cells for their induction into any cell type in the body. In the case of genetic diseases homologous recombination can be used to correct known mutations. The differentiated cells can then be used to replace lost or dysfunction cells in the patient. The cells can also be used as a disease model to understand pathogenic mechanisms and screen for drugs. Figure adapted from Chun et al 2010.
9
which is by no means an exhaustive list (Cao et al 2000). In 2008 Miccio et al were able to use stem cells to treat a lethal mouse model of the disease (th/3+). They used a lentiviral vector with an erythrocyte promoter to change the mutation back to normal. When they injected these hemopoietic stem cells they rescued the lethal phenotype, mice survived for over 500 days as opposed to 80 days in the untreated mice. Upon analysis the haemoglobin was increased which means the mice were able to produce functional erythrocytes. Wang et al (2012) later used this method to correct iPSC from a two year old with β-thalassemia. This was achieved successful and the resultant iPSC were differentiated into hematopoietic stem cells, which were injected into a non-lethal mouse model of β-thalassemia. This lead to an increase in haemoglobin which means these injected stem cells were able to undergo erythropoiesis. However this strategy needs to be used with a more severe lethal phenotype, to truly identify how effective this would be as a therapeutic. This technique shows a proof of principle which has the potential to effectively treat a wide ranging number of diseases. More specifically this could be very effective for treating neurodegenerative diseases, which are characterised by predominantly a loss in neurons.
Recently significant advancements have been made in using iPSC for the cell replacement of neurons for Huntington’s Disease (HD). HD is a neurodegenerative disease which predominantly affects the striatum and is caused by an abnormal number of CAG repeats. The disease is characterised by cognitive decline which leads to dementia and an increasing inability to control movement. Ultimately this disease is always fatal and although there are treatments available they are not very effective. Generation of striatal neurons and astrocytes from iPSCs derived from HD patients has been achieved recently (Juopperi et al 2012 and The HD iPSC consortium 2012). In itself this provides an essential cell model to further understand the disease as well as to test potential therapeutic drugs, as the phenotype of these cells is very similar to that of neurons with an induced mutation for HD. This was taken a step further when the iPSC derived from Huntington’s patients were corrected by homologous recombination (An et al 2012). Subsequently the neurons produced were saved from the HD phenotype. Interestingly they also injected NPC produced from these iPSC into a mouse model for HD. This indicated functional integration of the differentiated cells although analysis over a longer period is needed to assess how this helps to alleviate the phenotype of the mouse model. Overall this is a very exciting study which shows the potential of iPSC in alleviating neurodegenerative diseases. iPSC also have potential for the replacement of glial cells.
10
OPCs from iPSC
Only a very limited number of studies have produced OPCs from iPSC. One study compared the efficiency of producing OPC from ESCs compared to iPSCs which indicated similar efficiencies (14.4% and 12.6%). When they tried to differentiate these precursors into oligodendrocytes they noted that the efficiency was very different with 24.0% from the ESC and 2.3% from iPSC, suggesting subtle differences between the cells (Tokumoto et al 2010). The same group later improved the differentiation of iPSC into oligodendrocytes, using immunopanning, to 43.5% (Ogawa et al 2011). After 21 days 62.3% of these cells expressed the myelin binding protein (MBP). Czepiel et al (2011) are the only groups to inject OPC derived from iPSC into mice. They administered these mice with cuprizone which leads to demyelination of the corpus callosum. Although 80% of the cells injected into this area were lost, those that remained were able to remyelinate. Currently no studies have used iPSC to produce astrocytes, which should be focused on in the future. However Nori et al (2011) used NSCs derived from human iPSC and injected them into mice with SCI. This significantly improved recovery through differentiation into oligodendrocyes and astrocytes which indicates a proof of principle that could also be used for cell replacement in the central nervous system. iPSC have great potential for treating VWM disorder although there are still a number of problems with iPSC which need to be overcome before translation into the clinic is possible.
The differences and similarities between iPSCs and embryonic stem cells (ESC) are still hotly debated. In 2011 a number of epigenetic differences were identified (lister et al 2011), which is likely due to culturing processes. However more recently it has been shown that a lot of these changes pre-existed in the somatic cells used for reprogramming (Cheng et al 2012 and Young et al 2012). Current evidence shows the efficiency of iPSCs being induced into other cell types is lower than with ESC (Tokumoto et al 2010); suggesting subtle differences. Although Yamanaka argues that the iPSC being produced now are as different from ESCs as one ESC is from another (Blanpain et al 2012). The efficiencies of producing iPSCs from fibroblasts are also low, averaging at around 1%, with the four Yamanake TF’s (Table 1). This low efficiency will hinder the translation of iPSC into the clinic and it is therefore important to identify methods which can improve this. There are still a number of inherent problems associated with using iPSC that need to be overcome, which has prompted researchers to look for alternative methods.
11
Trans-differentiation
There are many advantages to producing iPSCs and then differentiating them into the cell of choice for therapies but this is an arduous and time consuming task. Consequently there has recently been a push to identify more efficient techniques which would in turn be more applicable in the clinical setting. In 1987 Davies et al showed that a single c-DNA (with protein sequence from the c-myc family) was able to trans-differentiate fibroblasts into myoblasts. The Oxford dictionary defines trans-differentiation as “the rare natural transformation of cells other than stem cells into a different cell type”. Trans-differentiation has the unique ability to miss out the pluripotent stage and go straight to a different somatic cell. Most notably for this review Vierbuchen et al (2010) were able to trans-differentiate mouse fibroblasts into neurons using only three factors, which was later achieved in human fibroblasts (Ambasudhan et al 2011 and Pang et al 2011 ). In vitro these cells had the ability to form functional synapses and membrane action potentials similar to their endogenous counterparts. Very recently Karow et al (2012) showed that pericytic cells from the cerebral cortex of adult humans could be differentiated into neurons with only two factors. These neurons also showed normal membrane properties and were able to form synapses. This study suggests that in
situ trans-differentiation of somatic cells in the brain could be used to replace neurons and
potentially glial cells also, although this line of research is only in its infancy. It is still debated as to whether trans-differentiation will take over iPSCs because changing somatic mutations (if treating a genetic disorder) is more challenging. In a recent expert review a number of authors voiced their concerns that trans-differentiation will produce only partially differentiated cells (Blanpain et al 2012). Extensive analysis is needed before a definitive answer can be made. Recent advances in trans-differentiation have now enabled the production of tripotent neural progenitor/stem cells (NPC) from fibroblasts, which could be used for glial cell replacement.
Induced neural progenitor cells
In 2011 Kim et al showed for the first time that trans-differentiation could produce induced neural progenitor cells from mouse fibroblasts. The difference between stem cells and progenitor cells is controversial, but the main difference is that progenitor cells are not immortalised. These cells showed great potential and a refined technique used in a later paper produced tripotent neural progenitor cells which could produce neural, astrocyte and oligodendrocyte cells (Lujan et al 2012). Interestingly they injected these cells in neonate shivering mice. Injection of these induced neural precursor cells lead to their differentiation into oligodendrocytes which indicated some
12
remyelination, by these cells, 10 weeks after injection. The paper did not state any improvements in the mice but it showed a proof of principle which could be improved. One of their most important findings is that these cells survived for 10 weeks in the brain which is by the far the longest reported time, using this technique, to date.
One of the most pressing questions about induced neural progenitor cells is how closely they resemble their endogenous counterparts. To date results indicate cells with a similar expression profile, epigenetic status and differentiation ability (Han et al 2012 and Thier et al 2012). Earlier this year Ring et al (2012) used trans-differentiation from human fibroblasts to neural stem cells, for the first time. This gives a further glimpse into the future of this emerging technique and suggests it has great potential. Interestingly this study, as well as the other publications noted, showed that iNPC/SC do not form tumours. Ring et al (2012) showed that NPC from iPSC had a percentage tumour formation of 62.5%-71.4% when injected into mice. Some of the TF needed to produce iPSC are also oncogenes; these oncogenes are not needed for trans-differentiation. The lack of a pluripotent stage also means contaminating tumourigenic iPSC will not be injected with the differentiated cells (Germain et al 2012). The efficiency and time taken to produce NPCs via trans-differentiation is superior to using iPSC (Table 1). Extensive analysis of how differentiated glial cells function, compared to their normal counterparts has not yet been done, although the use of common markers shows their potential. A recent study suggested in vitro injection of iNPCs lead to the differentiation of neurons that may have produced synapses. Further in vitro analysis showed the ability to form synapses and membrane properties similar to normal neurons (Sheng et al 2012). This study suggests fully functioning glial cells could also be produced from this technique, although more research is needed. There are still a number of issues to overcome with trans-differentiation (and iPSC) before they can be used as an effective therapeutic.
Viral transfection
Lentiviral and retroviral transfection are most commonly used for transfecting somatic cells with TFs to reprogram them into iPSC or somatic cells in trans-differentiation. This is a significant problem because their random insertion causes unwanted affects on genes which could promote tumourigenesis, invalidating their use in the clinic (Ghosh et al 2011 and Moriguchi et al 2010). A number of TFs used for this reprogramming are also oncogenes. Viral transfection means the TFs will be expressed long after the iPSCs have been produced, which will increase the likelihood of developing tumours. One advantage of trans-differentiation is that oncogenes are not needed for
13
reprogramming. Many studies have now developed strategies to try and overcome these problems. Stadtfeld et al (2008) used adenovirus transfection which does not lead to genome integration, using murine cells, but the process is very inefficient (<0.001%). These cells were pluripotent and resembled iPSCs produced using other methods. Importantly this was one of the first studies to prove that insertional mutagenesis is not needed for cell reprogramming. Another study used lentiviral transfection which was doxycycline dependant and could be excised using cre-recombinase (Soldner et al 2009). Upon analysis the iPSC produced more closely matched ESC, showing that genomic integration affects the quality of the cells produced. Unfortunately this technique is not perfect as some of the exogenous DNA is left behind. A number of techniques for non-viral transfection have now been developed to try and solve this problem.
Recombinant proteins can be directly inserted into a cell by being fused to a membrane permeable peptide. This is a novel idea which was successful in both murine and human cells (Kim et al 2009 and Zhou et al 2009) but is a time consuming method and again the efficiency is relatively low (<0.01%). Small molecules which can enter cells easily and efficiently, can also be used to significantly improve reprogramming. Recently a screen of 244 protein kinase inhibitors identified four which could independently produce iPSC, although the efficiency compared to TF reprogramming was not improved (Li et al 2012). Ladewig et al (2012) showed that small molecules which inhibited SMAD and GSK-3β and the addition of noggin significantly increased the trans-differentiation efficiency of fibroblasts to neurons, when combined with two pro-neuronal TFs. TFs alone gave a neuronal purity of <5% while the addition of the small molecules increased this to 37.5% with adult fibroblasts, in 23 days (identified by βII tub+ cells). This shows the significance of small molecule inhibitors but a replacement for TF is still needed.
Even though these methods avoid genomic integration the resultant iPSC are still tumourigenic, as indicated by the teratoma assay which is used to determine pluripotency. ESCs also have the ability to form tumours which means this is an intrinsic characteristic of pluripotent cells. Cells differentiated from iPSC, such as NSC, show an increased propensity for forming tumours, in vivo (Ring et al 2012). It is unclear how much of this propensity is due to the injection of contaminating iPSC along with the transfected cells and how much is due to the random insertion of TFs (Germain
et al 2012). Alternative reprogramming methods may also produce more viable cells as functionally
important proteins could also be affected. Using miRNAs is a promising new area of research which has the potential to negate these issues.
14
MiRNAs
When the human genome was first sequenced in full scientists were surprised to find that the majority of the DNA did not code for proteins. We now know that many of these non-coding regions still produce RNAs, but not proteins, and exert another layer of control on translation. One of the most ubiquitous non-coding RNAs is the microRNA (miRNA) so called for its 20- 25 nucleotide (nt) sequence length. MiRNA’s are normally transcribed into primary miRNA’s which undergo alterations by drosha complexes and the resultant hairpin loop is transported into the cytoplasm. The Dicer complex degrades one strand and the other strand associated with argonaute (Figure 2 and reviewed in Pasquinelli 2012). Repressing mRNAs then occurs through many functions. If the sequence is a direct match the mRNA is cleaved but if not a translational inhibition complex can form or the mRNA’s polyA tail can be removed which leads to its degradation (reviewed in Meister 2007). One miRNA has the unique ability to repress the translation of many mRNA’s, forming a complex regulatory network. Computational biology has given us great insights into the function of these networks. But the algorithms used are not perfect so experiments to determine miRNA targets have to be done which are complex and time consuming, but give an unrivalled insight into the function of these complex networks (Gennarino et al 2012). There is increasing evidence to suggest that miRNAs play an important role in cell development, differentiation and diseases in the brain (Im and Kenny 2012). This raised the unique idea that stopping the translation of proteins rather than up regulating the transcription of certain genes may be sufficient for cell differentiation.
The role of miRNA’s in pluripotency
Wang et al (2008) first coined the term embryonic stem cells specific cell cycle regulating miRNAs (ESCC), for a group of miRNAs which regulate the unique stem cell cycle. These are the miRNA
Figure 2 MiRNA biogenesis
MiRNA’s are transcribed as primary miRNA’s which are altered by drosha complexes and then transported into the cytoplasm by exportin 5. The hairpin structure is cleaved by dicer which results in one strand forming a complex with argonaute. This complex is able to inhibit the translation of mRNA’s by a number of different mechanisms, depending on the seuquence match (Pasquinelli 2012).
15
families’ miR-302, 367 and 391-395. These miRNAs are highly expressed in ESC and then decline prior to cell differentiation, which supports their role in maintaining pluripotency (Wang et al 2008, Subramanyam et al 2011 and Ren et al 2009). Since then expression profiling has identified a complex network of targets which promote “stemness”. The combination of miR-302 and 367 was shown to regulate 27 mRNA targets (which is not an exhaustive list)(Subramanyan et al 2011). These targets included cell cycle regulators, epigenetic regulators, cell signalling and vesicular transport (Subramanyam et al 2011). The miR-302 family of miRNAs are one of the most highly expressed in both iPSC and ESC (Lin et al 2008 and Sub et al 2004). MiR-302 causes epigenetic changes by targeting CpG binding proteins 1 and 2 and lysine-specific histone demethylases 1 and 2. In turn this destabilises DNA methyl transferase 1 (DNMT1), which causes genome wide demethylation; a hallmark of embryonic stem cells (Lin et al 2011, a, Subramanyam et al 2011 and Jackson et al 2004). MiR-302 also increases the expression of Oct4, Nanog and Sox. Oct4 and Sox are Yamanaka TFs and Nanog is a commonly used marker for iPSCs (Lin et al 2011, b). The combination of 302 and miR-372 also alters a number of cell cycle processes which promotes pluripotency (Rosa et al 2011). MiR-290-295 represent 70% of all the miRNAs expressed in ESC (Marson et al 2008) and appear to have similar downstream targets to c-myc (Judson et al 2011). This is by no means an exhaustive list of miRNAs and their downstream targets which shows the fundamental role that these miRNAs have in controlling a complex and diverse network to promote “stemness”. A number of studies have now used these miRNA’s in different combinations both with and without the commonly used Yamanaka factors to produce iPSC.
miRNA’s: a new way of producing pluripotent stem cells?
The miRNA families MiR291-95 showed a very modest increase in iPSC production efficiency in combination with TFs (Judso et al 2009). This was further improved when using MiR302/372 in combination with TFs (Subramanyam et al 2011 and Table 1). A significant step forward in this area of research was made by Anokye-Danso et al (2011). They showed that miR302/367 is sufficient to produce human iPSC. The time and efficiency was significantly improved from methods with TFs or miRNA’s in combination with TF (Figure 3) but viral transfection was still used (Anokye – Danso et al 2011). The authors reported integration into the genome which means the risk of tumourigenesis was not avoided in this study. Miyoshi et al (2011) took this one step further and transfected mature miRNA’s in successive 48 hour rounds, with miR200/302/369. The efficiency is very low (0.001%) and the process labour intensive but this study is noteworthy as it produced human iPSC without
16
genome integration. In the future studies need to identify common methods and the most effective set of miRNAs for reprogramming.
In all of these studies the teratoma assay was used to show pluripotency which means the cells were still tumourigenic. Injection into mice also leads to tumour formation. Whether cells differentiated from them are less likely to form tumours and also whether these cells are functionally closer to their endogenous counterparts needs to be determined. MiRNAs are sufficient to produce iPSC which means many of the genes needed for pluripotency are being transcribed into mRNA and then repressed at the post-transcriptional level. MiRNAs can remove their repressors which allows for deprogramming to the pluripotent state. This theory is supported by a recent study from ENCODE which showed up to 75% of the genome is transcribed at some point, in a number of different cell types (Djebali et al 2012). MiR-34a has been shown to inhibit reprogramming to iPSC (Choi et al 2011). This suggests that using methods to knock out miRNAs, such as anti-sense oligonucleotides, may provide alternative novel methods for cell reprogramming in the future. The exact mechanism behind how miRNAs reprogram cells is still poorly understood. Advancements in identifying miRNA targets will give many further insights into this process in the future (Baigude et al 2012). MiRNAs also play a fundamental role in the differentiation of many cell types.
1 7 iP S C m iR N A O ri g in a l ce ll V ir a l v e ct o r T im e (d a y s) E ff ic ie n ci e s (% ) P lu ri p o te n t P a ss a g e s T u m o u ri g e n ic R e fe re n ce - H u m a n fi b ro b la st re tr o v ir u s 3 0 0 .0 2 Y - Y T a k a h a sh i e t a l 2 0 0 7 - M o u se fi b ro b la st a d e n o v ir u s 3 5 -4 0 < 0 .0 0 1 % Y - Y S tr a d tf e ld e t a l 2 0 0 8 - M o u se fi b ro b la st L e n ti v ir u s 4 6 0 .0 1 % Y - Y S o ld n e r e t a l 2 0 0 9 - H u m a n fi b ro b la st - 5 6 < 0 .0 1 % Y - Y K im e t a l 2 0 0 9 M iR 3 0 2 / 3 6 7 ( n o T F ) H u m a n fi b ro b la st L e n ti v ir u s 1 5 1 0 % Y - Y A n o k y e -D a n so e t a l 2 0 1 1 M iR 2 0 0 / 3 0 2 / 3 6 9 (n o T F ) H u m a n fi b ro b la st - 2 0 0 .0 0 1 % Y M iy o sh i e t a l 2 0 1 1 M iR 2 9 1 -2 9 5 ( w it h T F ) M o u se fi b ro b la st s re tr o v ir u s 1 0 -1 5 * Y - Y Ju d so n e t a l 2 0 0 9 M iR 3 0 2 / 3 7 2 H u m a n fi b ro b la st s 2 1 -3 1 ‡ 1 Y - Y S u b ra m a n y a m e t a l 2 0 1 1 (i )N P C T ra n s-d if fe re n ti a ti o n - M o u se fi b ro b la st R e tr o v ir u s 1 9 0 .0 0 5 -0 .0 0 8 T ri p o te n t > 5 0 N T h ie r e t a l 2 0 1 2 - M o u se fi b ro b la st L e n ti v ir u s 2 5 5 N e u ro n a l a st ro cy ti c > 1 2 N L u ja n e t a l 2 0 1 2 - M o u se fi b ro b la st L e n ti v ir u s 2 5 1 1 .5 T ri p o te n t > 1 2 N L u ja n e t a l 2 0 1 2 - M o u se fi b ro b la st re tr o v ir u s 3 0 .0 0 1 -0 .0 0 2 T ri p o te n t N S h e n g e t a l 2 0 1 2 T a b le 1
1 8 * e n h a n ce d T F e ff ic ie n cy o f 0 .0 1 -0 .0 5 % t o 0 .1 -0 3 % ‡ 2 1 d a ys ( 4 Y a m a n k a T F ), 3 1 d a y s (3 Y a m a n a k a T F) 1 . 1 2 -1 7 f o ld i n cr e a se i n c o lo n ie s co m p a re d t o m o ck c o n tr o l Y – y e s N - n o N /A – n o t a p p li ca b le ( e xp e ri m e n t n o t d o n e ) iP S C - h iP S C - 1 5 d a y s 1 5 -7 9 T ri p o te n t - N / A H u e t a l 2 0 1 0 - h E S C - 1 5 d a y s 9 0 -9 7 T ri p o te n t - N / A H u e t a l 2 0 1 0 - h iP S C (4 1 1 a n d W T .9 ) - 1 4 d a y s U n k n o w n T ri p o te n t - 6 2 -7 1 % R in g e t a l 2 0 1 2
19
miRNA’s: neural and oligodendrocye progenitor cells
Early work on the role of the miRNA’s in brain development used knockdown models of important components in miRNA biogenesis such as Dicer and Argonaut 2 (Figure 2), which showed their fundamental role in brain development (Giraldez et al 2005 and Lui et al 2004). Further in vivo analysis showed that dicer knockdown inhibits both neuronal and gliogenesis which leads to premature death in mice, but does not severely affect neural stem cell/progenitor cell populations (Tonelli et al 2008). This was further verified by neural stem cells isolation from these mutant mice. Again they were not able to form neurons or glia cells, but this was restored with the introduction of exogenous dicer (Andersson et al 2010). This led to the assumption that miRNAs are not important for the maintenance of neural stem/ progenitor cells but only in their differentiation into other cell types. However in 2010 two studies identified a miRNA biogenesis pathway devoid of dicer function (Cheloufi et al 2010 and Cifuentes et al 2010). Consequently not much is known about miRNAs which are specific to NSC. More data is known about inhibitors of this fate. For instance the miRNA let-7a negatively regulates a transcription regulator called hgma2. In age the ability of neural stem cells to self-renew and differentiate decreases. Hgma2 is increased in NSC early in development and then decreases with age. (Nishino et al 2008). Knock down of miR-9 also maintains neural progenitor cells and stops neuronal differentiation (Delaloy et al 2010). Due to the lack of available promoters it has been difficult to identify the role of miRNAs in vivo (Shi et al 2010) although one study showed that Dicer knock down in the mouse retina lacked the development of late progenitor cells (Georgi and Reh 2010). A recent study has also suggested miR-125 plays an important role in NPC/NSC through Nestin signalling. (Cui et al 2012).
There is a growing body of evidence which supports the role of miRNA in glial cell development. OPC will be focused on as these are the cells of choice for oligodendrocyte replacement therapies. We have taken a big leap forward in this area of research through expression profiling of oligodendrocytes derived from rats as well as oligodendrocytes differentiated from human ESC. The latter has proven extremely useful because it most accurately resembles the process in humans and allows for expression profiling at different stages of differentiation. In 2008 Lau et al showed for the first time the importance of miRNAs in regulating the differentiation of oligodendrocytes and identified a dynamic change in 46 miRNAs. miR-199a5p has been identified as one of the most important due to its strong bias against CIIorf9. Currently this is thought to be the human orthologue of the myelin gene regulating factor (MRF) found in rodents (Emery 2009). This protein is critical for oligodendrocyte development and myelination. More recently Zhao et al (2012) showed that
20
miRNA-7a is important for the differentiation of both NPC and ESC into OPC by repressing pro-neuronal genes such as NeuroD4 and Pax6. This study suggests miRNA’s can be used to produce OPC from either iPSC or NPC. This is already possible but using miRNAs could both speed up the process and improve the quality of the cells produced. Further research will be needed to identify whether this could produce functional oligodendrocytes which could be used in cell replacement therapies.
miRNA’s: trans-differentiation
In 2005 Gokhan et al identified a number of transcription factors expressed in oligodendrocyte cells which they then transfected into fibroblasts but this was not sufficient to induce endogenous myelin expression. Later a similar technique was used but in combination with epigenetic modulation, however this attempt at trans-differentiation was again unsuccessful (Liu et al 2010). This suggests that miRNAs may be the missing ingredient. Vierbuchen et al 2010 and later Ambasudhan et al 2011 both showed that miRNAs could be used in combination with TFs to produce neurons using trans-differentiation. Crabtree et al 2011 then showed this was possible with miRNAs alone, using miR09 and miR-124. The efficiency was very low and the resultant neurons were substandard, but when used in combination with the transcription factor NeuroD1, this issue was resolved. The neurons produced were able to form action potentials and had very similar membrane properties to their endogenous counterparts. As already discussed there is some scepticism about whether trans-differentiation can produce completely differentiated cells so more analysis is needed to determine whether this is the case. If fibroblasts can be converted directly to neurons, using miRNAs alone, there must be miRNAs expressed in fibroblasts which are needed for neuronal fate, but are repressed post-translationally. Baf53a is one such example. Baf53a is expressed in both cell types and represses neuronal differentiation. Both miR-9 and miR-124 target its mRNA which leads to the up regulation of Baf53b and promotion to the neural fate (Shenoy and Blelloch 2012).
Trans-differentiation, using miRNAs, could be possible for producing both NSC/NPC and OPC. Lujan
et al (2012) showed that the Yamanaka TFs are sufficient for producing NPCs from fibroblasts, when
cultured in the right conditions. Although the cells were not tripotent it suggests miRNAs which are used to produce iPSC could also be used in this process. MiR-302, as already discussed, is central to reprogramming cells to pluripotency. However this family of miRNAs also inhibits differentiation to the neural fate, as shown by inhibitors of MiR-302-367 which increases the efficiency of differentiation to neural stem cells (lipchina et al 2012). One of the mechanisms for this inhibition is the positive effect that these miRNAs have on bone morphogenetic protein (BMP) and transforming
21
growth factor beta (TGF-β) signalling (lipchina et al 2012). Interestingly a key TF which maintains the “stemness” of NSC is SOX2, which negatively regulates these pathways (Bylund et al 2003). SOX2 knockout promotes NSC differentiation its expression can produce iNSC without any other TFs. (Bylund et al 2003 and Ring et al 2012). This suggests that miRNAs, or other methods, which up regulate SOX2 could replace viral transfection of SOX2 in this reprogramming. MiR-302 up regulates SOX2, whether there are miRNAs that can do this and not inhibit neural differentiation is still to be determined. Ichida et al (2009) screened for small molecules that could replace SOX2. They identified a number but the most effective inhibits TGF-β, suggesting it could also be useful in producing iNSC. This small molecule also led to an increase in nanog, which helps to maintain NSC (Po et al 2010). None of these small molecules actually led to an increase in SOX2 which positively regulates a number of proteins important for the NSC/NPC fate (Graham et al 2003). Consequently further research is needed to identify whether these molecules would be sufficient to trans-differentiate somatic cells into NPC/NSC. A number of miRNAs have been identified which negatively regulate SOX2 such as; miR-200 and miR-145. Anti-sense oligonucleotides (or other methods for knocking down miRNAs) could also be effective for increasing SOX2 function. It may also be possible to reprogram somatic cells straight to OPCs using miRNAs, in specific culture conditions which promote the OPC lineage. MiR-7a and miR-199a5p are likely to be the most important.
A window into the brain: using in vivo analysis of stem cells to
improve cell replacement therapies
One of the main areas of study which is currently lacking in stem cell therapy for brain disorders is the analysis of the differentiated cells in vivo, to identify how well they are able to function in comparison to normal cells. When stem cells are injected into the brain for cell replacement the majority of studies will then sacrifice the animal and analyse its brain slices. The problem with this method is that we only get a small snapshot of what these stem cells are doing and animals have to be sacrificed at different time points, increasing the number of animals which have to be used. This technique is also not possible in human trials. In vivo imaging allows the visualisation of these stem cells, non-invasively; to gain a much better understanding of what is really happening in the brain. Magnetic resonance imaging (MRI) is a technique which can be used effectively in both clinical and pre-clinical trials. Multi-photon microscopy offers far better resolution and has great potential for future research, although it cannot be used in humans. These techniques have revolutionised many fields of neuroscience and would help us to gain an unprecedented analysis of stem cells in the brain.
22
MRI, which is also called nuclear MRI, uses a large magnet to promote a rotating magnetic field in the nuclei of atoms in the body. This field can then be detected to produce a 3 dimensional image for non-invasive analysis of the brain. Superparamagnetic iron oxides (SPIO) have now been used in a number of studies to label stem cells for in vivo visualisation of transplantation. Bulte et al (2001) used magnetodendrimer 100 (MD), which is a SPIO, to label OPC. MD-100 is specifically targeted to endosomes which limits its cellular toxicity. This study isolated rat NSC and differentiated them into OPCs, before labelling them with MD-100. They were then injected into the lateral ventricle of Long Evans shaker rats which show hypomyelination due to a mutation in the myelin binding protein (MBP). The injected cells could then be analysed by MRI at different time points and subsequent post-mortem analysis identified myelin production from these cells. The signal faded within 42 days, which is not ideal, however visualisation of the early events are likely to be the most important. The SPIO technique can also be used in humans as was clearly shown in a very small clinical trial for traumatic brain injury. Cells from the brain were isolated from a patient with traumatic brain injury and then cultured which a selective bias for NSC. The NSC were then injected back into the individuals brain and MRI was used at multiple time points which showed movement of the cells from the injection site into the affected area (Zhu et al 2006). One problem with this technique is that macrophages can engulf the injected cells – which mean the MRI would actually be visualising macrophages. To combat this problem cells are virally transfected to express GFP. GFP will only be expressed in living cells which have not been engulfed by macrophages. It is then possible to compare the location of GFP cells to the cells which are visualised using MRI, which shows macrophages are not a significant issue (Thu et al 2012 and Zhu et al 2006). Another major advantage of this technique is that the number of injected cells in different areas of the brain can be approximated. Recently the SPIOs which can be used in humans have been taken off the market which means clinical trials are currently unable to use this technique. Zhu et al (2012) have recently developed a nanocomplex comprising of heparin, Ferumoxytol and protamine which are able to effectively label stem cells and can be visualised after in vivo injection using MRI. These drugs are already approved by the food and drug administration (FDA) which means its use in clinical trials should be possible in the near future. Multi-photon microscopy can also be used to analyse stem cells in the brain, at a much higher spatial resolution.
Multi-photon microscopy is a type of fluorescent microscopy which is unique due to its ability to view optical sections, which is useful when analysing a very complex structure such as the brain. For mouse studies in vivo multiphoton microscopy is an exciting technique which has many applications, only some of which have been realised to date (Dunn et al 2008). A craniotomy is performed on the
23
mouse and is replaced by a cover slip, literally making a window into the brain. Fluorescent proteins can then be visualised longitudinally. For stem cells simple techniques such as tagging them with a green fluorescent protein (GFP) to monitor their activity in vivo could be utilised. The production of the brainbow mouse shows the number of different fluorescent proteins available (Livet et al 2007). The same group recently showed, for the first time, that you can view green, yellow-orange and blue wavelengths at the same time, using multiphoton microscopy. This produced a 3D image from a 450µm thick slice of the cortex which allows for amazing spatial resolution (Figure 3). The applications for this, whether using an organtypic brain slice or in vivo visualisation is a very exciting prospect for stem cell mediated therapies. For glial cells different fluorescent proteins could be tagged to astrocyte or oligodendrocyte specific promoter sites to monitor their differentiation. Proteins important in the differentiation and function of these cells could also be tagged to monitor the functionality of the cells. Loannidou et al (2012) injected GFP expressing neurospheres into the
ex vivo spinal cord of shiverer mice. Using multi-photon microscopy (as well as other techniques)
they were able to visualise the interaction between neurons and oligodendrocytes. In vivo visualisation of stem cells using multi-photon microscopy will provide many new insights into the functional viability of the differentiated cells that they produce, in the brain.
B A
Figure 3 3D multi-photon microscopy image of the brainbow mouse
A) Visualising one of optical planes of the brainbow mouse cortex using multi-photon microscopy. This comes from image B. B) this shows a 3D image of a 450µm thick brain slice from the brainbow mouse. It is possible to produce optical slices or planes for more precise analysis, as seen in image A. Images adapted from supplementary video 1, Mahou et al 2012. Green, yellow-orange and blue wavelengths analysed.
24
Conclusions
There has been a revolution in our understanding of how stem cells could be used for glial cell therapies in the CNS. iPSC can be obtained from the patient to be treated which negates previous ethical issues with ESC and prevents immune rejection. Viral transfection is used to produce these cells which can lead to genomic integration and tumourigenesis. A number of techniques have been used to try and solve this problem however they are predominantly inefficient and time consuming; small molecule inhibitors have, however, shown promise. MiRNA’s are now intrinsically associated with a complex network of proteins which both allows for and promotes the pluripotent state. Recent findings have shown that miRNAs can be used to produce iPSC which has the potential to be far more efficient than the Yamanaka TFs. Whether these cells are less likely to form tumours and also whether the quality of the cells is improved needs to be analysed. More research is also needed to understand the complex networks which are altered by these miRNAs during reprogramming as our current knowledge of their targets is lacking. A common protocol also needs to be developed as different methods and different miRNAs make all of the current studies difficult to compare. An alternative to iPSC is trans-differentiation.
Trans-differentiation offers an alternative approach whereby somatic cells can be differentiated into another cell type without the pluripotent step. This technique is faster and more efficient than the alternative iPSC technique. However there are concerns that the cells will not be completely differentiated and homologous recombination (for the cell replacement therapy of genetic diseases) is more challenging. Fibroblasts can be directly reprogrammed into iNPC which are tripotent and can be injected for the treatment of vanishing white matter disorder as this would allow for the differentiation of both oligodendrocytes and astrocytes. MiRNAs could be used to mediate this process and may stop any tumourigentic potential of the cells. Although none has been identified so far; this is likely due to the lack of undifferentiated contaminating cells being injected.
It is unknown whether astrocytes or oligodendrocytes are affected first in VWM disease and so whether both or just one cell type needs to be introduced for treatment. MiRNAs are important for OPC and NPC development and maintenance. Therefore they could be used to produce iNPC or NPC from iPSC and subsequently OPC (Figure 3). This has the potential to increase the speed, efficiency and quality of the cells produced, compared with existing techniques. This knowledge may also make direct trans-differentiation to OPCs possible. We know much less about astrocyte development and so NPC are currently the best option for their replacement in disease. These cell replacement studies
25
will gives further insights into vanishing white matter disease by understand which cells need to be replaced as well as whether the environment in the brain will allow for cell replacement therapies. Future research should also focus on identifying more miRNAs which are important in the differentiation of NPC, OPC and astrocytes as well as try to identify their targets, which are still poorly understood. Finally there is a need for
differentiated cells are able to function in the CNS, to truly indentify how closely they resemble their endogenous counterparts.
Figure 3 Stem cell replacement therapy for VWM Astrocytes and glial cells are both affected
both cell types will be an effective therapeutic or whether the environment will be sufficient for their survival. NPCs can be produced from iPSC or
replacement of both astrocytes and oligodendrocytes. They can also be differentiated into OPC, for oligodendrocyte replacement. MiRNAs are fundame
reduce the chance of tumourigenesis as well improve the speed, efficiency and quality of the cells which are produced. MiRNAs can already be used to produce iPSC and trans
have suggested some of the most important miRNAs that could be from “saving the brain’s white matter with mutated mice
indicate alternative reprogramming. Dotted arrow indicates potential reprogramming method. *small mol have been identified which can replace some functions of SOX2 (Ichida
will gives further insights into vanishing white matter disease by understand which cells need to be replaced as well as whether the environment in the brain will allow for cell replacement therapies. ture research should also focus on identifying more miRNAs which are important in the differentiation of NPC, OPC and astrocytes as well as try to identify their targets, which are still poorly understood. Finally there is a need for in vivo techniques to monitor how well these differentiated cells are able to function in the CNS, to truly indentify how closely they resemble their
Stem cell replacement therapy for VWM
rocytes and glial cells are both affected in VWM disease. We do not know whether replacement of one or both cell types will be an effective therapeutic or whether the environment will be sufficient for their survival. PCs can be produced from iPSC or directly by trans-differentiation. NPCs can then be injected directly for replacement of both astrocytes and oligodendrocytes. They can also be differentiated into OPC, for oligodendrocyte replacement. MiRNAs are fundamental to all of these reprogramming
tumourigenesis as well improve the speed, efficiency and quality of the cells which are MiRNAs can already be used to produce iPSC and trans-differentiate fibroblasts into neurons. We have suggested some of the most important miRNAs that could be used for reprogramming.
saving the brain’s white matter with mutated mice”. Red arrows indicate TF reprogramming. Thick arrows indicate alternative reprogramming. Dotted arrow indicates potential reprogramming method. *small mol have been identified which can replace some functions of SOX2 (Ichida et al 2009).
will gives further insights into vanishing white matter disease by understand which cells need to be replaced as well as whether the environment in the brain will allow for cell replacement therapies. ture research should also focus on identifying more miRNAs which are important in the differentiation of NPC, OPC and astrocytes as well as try to identify their targets, which are still monitor how well these differentiated cells are able to function in the CNS, to truly indentify how closely they resemble their
We do not know whether replacement of one or both cell types will be an effective therapeutic or whether the environment will be sufficient for their survival. hen be injected directly for replacement of both astrocytes and oligodendrocytes. They can also be differentiated into OPC, for steps and may be able to tumourigenesis as well improve the speed, efficiency and quality of the cells which are differentiate fibroblasts into neurons. We used for reprogramming. Picture of rat taken Red arrows indicate TF reprogramming. Thick arrows indicate alternative reprogramming. Dotted arrow indicates potential reprogramming method. *small molecules
26
References
Ambasudhan R, Talantova M, Coleman R, Yuan X, Zhu S, Lipton SA, Ding S (2011). Direct reprogramming of adult human fibroblasts to functional neurons under defined conditions. Cell
Stem Cell. 9:113-8
An encephalopathy with unique features on magnetic resonance imaging and proton magnetic An MC, Zhang N, Scott G, Montoro D, Wittkop T, Mooney S, Melov S, Ellerby LM (2012). Genetic
Correction of Huntington's Disease Phenotypes in Induced Pluripotent Stem Cells. Cell Stem Cell. 11: 253-63
Andersson T, Rahman S, Sansom SN, Alsiö JM, Kaneda M, Smith J, O'Carroll D, Tarakhovsky A, Livesey FJ (2010). Reversible block of mouse neural stem cell differentiation in the absence of dicer and microRNAs. PLoS One. 5:e13453.
Anokye-Danso F, Trivedi CM, Juhr D, Gupta M, Cui Z, Tian Y, Zhang Y, Yang W, Gruber PJ, Epstein JA, Morrisey EE (2011). Highly efficient miRNA-mediated reprogramming of mouse and human somatic cells to pluripotency. Cell Stem Cell. 8:376-88.
Arai K, Lo EH (2010). Astrocytes protect oligodendrocyte precursor cells via MEK/ERK and PI3K/Akt signaling. J Neurosci Res. 88:758-63.
Arai K, Lo EH (2010). Astrocytes protect oligodendrocyte precursor cells via MEK/ERK and PI3K/Akt signaling. J Neurosci Res. 88:758-63.
Baigude H, Ahsanullah, Li Z, Zhou Y, Rana TM (2012). miR-TRAP: a benchtop chemical biology strategy to identify microRNA targets. Angew Chem Int Ed Engl. 51:5880-3.
Bakshi A, Shimizu S, Keck CA, Cho S, LeBold DG, Morales D, Arenas E, Snyder EY, Watson DJ, McIntosh TK (2006). Neural progenitor cells engineered to secrete GDNF show enhanced survival, neuronal differentiation and improve cognitive function following traumatic brain injury.
Eur J Neurosci. 23:2119-34.
Baumann N, Pham-Dinh D (2001). Biology of Oligodendrocyte and Myelin in the Mammalian Central Nervous System. Physiol Rev. 81:871-927.
Bernadette Modell, Maren Khan, Matthew Darlison,Mark A Westwood, David Ingram, and Dudley J Pennell (2008). Improved survival of thalassaemia major in the UK and relation to T2* cardiovascular magnetic resonance. J Cardiovasc Magn Reson.10: 42.
Blanpain C, Daley GQ, Hochedlinger K, Passegué E, Rossant J, Yamanaka S (2012). Stem cells assessed. Nature Reviews Molecular Cell Biology 13: 471-476
Bylund M, Andersson E, Novitch BG, Muhr J (2003). Vertebrate neurogenesis is counteracted by Sox1-3 activity. Nat Neurosci. 6:1162-8
C.Sheng et al (2012). Direct reprogramming of sertoli cells in multipotent neural stem cells by defined factors. Cell Research 22:208-218
Cheloufi S, Dos Santos CO, Chong MM, Hannon GJ (2010). A dicer-independent miRNA biogenesis pathway that requires Ago catalysis. Nature. 465 :584-9.
Cheng, L et al (2012). Low incidence of DNA sequence variation in human induced pluripotent stem cells generated by nonintegrating plasmid expression. Cell Stem Cell 10: 337–344.
Chun YS, Chaudhari P, Jang YY (2010). Applications of Patient-Specific Induced Pluripotent Stem Cells; Focused on Disease Modeling, Drug Screening and Therapeutic Potentials for Liver Disease.
Int J Biol Sci, 6:796-805
Cifuentes D, Xue H, Taylor DW, Patnode H, Mishima Y, Cheloufi S, Ma E, Mane S, Hannon GJ, Lawson ND, Wolfe SA, Giraldez AJ (2010). A novel miRNA processing pathway independent of Dicer requires Argonaute2 catalytic activity. Science. 328:1694-8.
Cui Y, Xiao Z, Han J, Sun J, Ding W, Zhao Y, Chen B, Li X, Dai J (2012). MiR-125b orchestrates cell proliferation, differentiation and migration in neuralstem/progenitor cells by targeting Nestin.
27
Czepiel M, Balasubramaniyan V, Schaafsma W, Stancic M, Mikkers H, Huisman C, Boddeke E, Copray S (2011). Differentiation of induced pluripotent stem cells into functional oligodendrocytes. Glia. 59:882-92.
Davis, R. L.; Weintraub, H.; Lassar, A. B. (1987). "Expression of a single transfected cDNA converts fibroblasts to myoblasts". Cell 51:987–1000
De Pietri Tonelli D, Pulvers JN, Haffner C, Murchison EP, Hannon GJ, Huttner WB (2008). miRNAs are essential for survival and differentiation of newborn neurons but not for expansion of neural progenitors during early neurogenesis in the mouse embryonic neocortex.
Development. 135:3911-21.
Delaloy C, Liu L, Lee JA, Su H, Shen F, Yang GY, Young WL, Ivey KN, Gao FB (2010). MicroRNA-9 Coordinates Proliferation and Migration of Human Embryonic Stem Cell–Derived Neural Progenitors. 6: 323–335.
Djebali S, Davis CA, Merkel A, Dobin A, Lassmann T, Mortazavi A, Tanzer A, Lagarde J, Lin W, Schlesinger F, Xue C, Marinov GK, Khatun J, Williams BA, Zaleski C, Rozowsky J, Röder M, Kokocinski F, Abdelhamid RF, Alioto T, Antoshechkin I, Baer MT, Bar NS, Batut P, Bell K, Bell I, Chakrabortty S, Chen X, Chrast J, et al(2012). Landscape of transcription in human cells. Nature 489: 101–108
Esau CC (2008). Inhibition of microRNA with antisense oligonucleotides. Methods. 44:55-60.
Gennarino VA, D'Angelo G, Dharmalingam G, Fernandez S, Russolillo G, Sanges R, Mutarelli M, Belcastro V, Ballabio A, Verde P, Sardiello M, Banfi S (2012). Identification of microRNA-regulated gene networks by expression analysis of target genes. Genome Res. 22:1163-72
Georgi SA, Reh TA (2010). Dicer Is Required for the Transition from Early to Late Progenitor State in the Developing Mouse Retina. The Journal of Neuroscience 30: 4048-4061
Germain ND, Hartman NW, Cai C, Becker S, Naegele JR, Grabel LB (2012). Teratocarcinoma formation in embryonic stem cell-derived neural progenitorhippocampal transplants. Cell Transplant. . [Epub ahead of print]
Geva M, Cabilly Y, Assaf Y, Mindroul N, Marom L, Raini G, Pinchasi D, Elroy-Stein O (2010). A mouse model for eukaryotic translation initiation factor 2B-leucodystrophy reveals abnormal development of brain white matter. Brain 133:2448-61.
Ghosh Z, M Huang, S Hu, KD Wilson, D Dey and JC Wu. (2011). Dissecting the oncogenic and tumorigenic potential of differentiated human induced pluripotent stem cells and human embryonic stem cells. Cancer Res 71:5030–5039.
Giorgetti A, Montserrat N, Rodriguez-Piza I, Azqueta C, Veiga A, Izpisúa Belmonte JC (2010). Generation of induced pluripotent stem cells from human cord blood cells with only two factors: Oct4 and Sox2. Nat Protoc. 5:811-20
Giraldez AJ, Cinalli RM, Glasner ME, Enright AJ, Thomson JM, Baskerville S, Hammond SM, Bartel DP, Schier AF (2005). MicroRNAs regulate brain morphogenesis in zebrafish. Science 308: 833–838 Graham V, Khudyakov J, Ellis P (2003). SOX2 functions to maintain neural progenitor identity.
Neuron. 39:749-6
Haas C, Neuhuber B, Yamagami T, Rao M, Fischer I (2012) Phenotypic analysis of astrocytes derived from glial restricted precursors and their impact on axon regeneration. Exp Neurol. 233:717-32 Hagan JP, Piskounova E, Gregory RI (2009). Lin28 recruits the TUTase Zcchc11 to inhibit let-7
maturation in mouse embryonic stem cells. Nat Struct Mol Biol. 16:1021-5
Han DW, Tapia N, Hermann A, Hemmer K, Höing S, Araúzo-Bravo MJ, Zaehres H, Wu G, Frank S, Moritz S, Greber B, Yang JH, Lee HT, Schwamborn JC, Storch A, Schöler HR (2012). Direct Reprogramming of Fibroblasts into Neural Stem Cells by Defined Factors. Cell Stem Cell 10: 465– 472
