Progress and Challenges of Cell Replacement Therapy for Neurodegenerative Diseases Based on Direct Neural Reprogramming

Abstract

Neurodegenerative diseases are characterized by protein aggregation and progressive degeneration of neurons, causing severe functional deficiency in cognition, behavior, and movement. Until now, there has been no effective treatment available in the clinic. Considering the selective loss of specific neurons in the human brain in the pathogenesis of these diseases, generating functional neurons in vitro or in vivo to replace the lost neurons represents a novel strategy to treat neurodegenerative diseases. Human embryonic stem cells and induced pluripotent stem cells have good potential for cell replacement therapy. However, limitations, such as the possibility of tumor formation, have hindered its applications. Recently, a novel approach, direct neural reprogramming, in which somatic cells are reprogrammed to functional neurons without a stem-cell state, has emerged an alternative for cell replacement. Specific human somatic cells can be reprogrammed to functional subtype neurons via the introduction of transcription factors, microRNAs, or small molecules in vitro and in vivo, thereby reducing the risk of carcinogenesis. Studies demonstrated symptomatic relief when induced neurons were transplanted into animal models. Although the direct neural reprogramming holds great promise for cell replacement therapy, there remain a number of challenges for its clinical application, including low efficiency, unclear mechanisms, and safety concerns. This review highlights the progress and challenges of this technique, and discusses perspectives for its applications in cell replacement.

Introduction

Neurodegenerative diseases are a type of central nervous system disorder characterized by protein aggregation and progressive degeneration of neurons, finally resulting in functional deficiency in cognition, behavior, or movement. Few therapies are currently available in clinical practice, and most of them aim to relieve symptoms but fail to stop or slow the progression of the disease.¹ Given the age-related property of neurodegenerative diseases and the increasingly aging population, effective and safe treatment strategies are urgently needed. The progressive degeneration of specific neurons plays a significant role in the pathogenesis of neurodegenerative diseases. For example, the selective loss of dopaminergic (DA) neurons in the substantia nigra region of the midbrain is involved in the pathogenesis of Parkinson's disease (PD), a chronic and progressive movement disorder.² Given that neurons have a limited capacity for self-regeneration, cell replacement therapy has become a promising approach to treat neurodegenerative diseases.

The first attempt at cell replacement therapy was the isolation and in vivo transplantation of mouse teratocarcinoma cells. However, they exhibited limited differentiation potential. The isolation and differentiation of human embryonic stem cells (ESCs)³ led to another useful method to generate human neurons. Mouse ESCs transplanted into the striatum of rat models of PD fully differentiated into DA neurons and restored cerebral function and behavior in the animal PD model.⁴ However, the difficulty in establishing human ESC lines has limited the application of this approach. Recently, the discovery of human induced pluripotent stem cells (iPSCs) by introducing four transcription factors—Oct3/4, Sox2, c-Myc, and Klf4—into mouse embryonic or adult fibroblasts⁵ has overcome this limitation, and iPSCs have replaced ESCs in the generation of neurons.⁶ Before iPSC-derived neurons can be transplanted into humans to treat neurodegenerative diseases, certain hurdles, such as the possibility of tumor formation, must be overcome.⁷ In contrast, direct reprogramming of human somatic cells to generate specific cell types by introducing transcription factors or microRNAs reduces the risk of carcinogenesis without the intermediate stem-cell state.

This review highlights the progress and development of direct reprogramming in cell replacement therapy for neurodegenerative diseases, especially PD. The clinical challenges related to neuron conversion are discussed, and further perspectives are provided.

Direct Reprogramming Into iPSCs, Cardiomyocytes, Pancreatic B-Cells, and Hepatocytes

The overexpression of four specific transcription factors—Oct3/4, Sox2, c-Myc, and Klf4—reprogrammed human fibroblasts into iPSCs. Further studies demonstrated that in addition to fibroblasts, direct reprogramming toward iPSCs could be achieved for other specific human somatic cell types, such as liver and stomach cells, hematopoietic cells, and terminally differentiated mature lymphocytes.^8
–10 Meanwhile, further interventions were introduced in the direct reprogramming toward iPSCs. In an attempt to reprogram mouse B lymphocytes, in addition to introducing the four classical transcription factors, the transcriptional state maintaining B-cell identity was also disrupted by either ectopic expression of the myeloid transcription factor CCAAT/enhancer-binding-protein-alpha (C/EBPalpha) or specific knockdown of the B-cell transcription factor Pax5.¹⁰ Together, these studies raised the possibility of reprogramming specific somatic cell types into other lineages, without a stem-cell state, by introducing defined combinations of factors.

In an attempt to achieve direct reprogramming toward cardiomyocytes, Leda et al.¹¹ introduced 14 candidate transcription factors into cardiac fibroblasts via retroviruses, and detected the successful activation of cardiac markers in some cells after 2 weeks. Further experiments revealed that three cardiac transcription factors—Cata4/Mef2c/Tbx5 (GMT)—were sufficient for the direct reprogramming of fibroblasts toward cardiomyocytes. The induced cardiomyocytes were similar to neonatal cardiomyocytes in terms of their global gene expression, and exhibited spontaneous contraction, indicating that they were functional. Furthermore, additional combinations of defined factors were found to be sufficient for cardiac reprogramming and could promote the process, such as transcription factors Hand2,¹² microRNAs,¹³ and other small molecules.¹⁴ More recently, in vivo direct cardiac reprogramming was achieved successfully, which revealed that the in vivo environment could promote reprogramming.^15,16 Therefore, the generation of cardiomyocytes by the direct reprogramming of somatic cells has been achieved and represents a promising approach for cardiac disease treatment.

Direct reprogramming of a specific somatic cells t has also been studied to generate other cell types as a treatment strategy for various diseases. Huang et al.¹⁷ generated human induced hepatocytes from fibroblasts using lentiviral expression of FOXA3, HNF1A, and HNF4A. Human induced hepatocytes, expressing hepatic gene repertoires and displaying characteristics of mature hepatocytes, including cytochrome 450 enzyme activities and biliary drug clearance, could restore liver function and prolong survival when transplanted into acute liver failure mouse models. Recently, the direct in vivo reprogramming approach conducted by Sharma's group¹⁸ demonstrated that the in vivo expression of the transcription factors FOXA3, GATA4, HNF1A, and HNF4A reprogrammed myofibroblasts in fibrotic mouse livers into hepatocyte-like cells and reduced liver fibrosis. Thus, the direct reprogramming toward specific mature and functional cell types has been achieved in vitro and in vivo, greatly promoting the development of cell replacement therapy.

Direct Reprogramming into Functional Neurons using Transcription Factors

The successful direct reprogramming of mouse and human fibroblasts into iPSCs raised the question of whether other specific cell types could be generated using this approach. Considering the pathogenesis of neurodegenerative diseases, direct neural reprogramming has become a hot topic in research. In 2010, Vierbuchen et al.¹⁹ hypothesized that a combination of neural lineage-specific transcription factors could directly reprogram fibroblasts into neurons. They identified a combination of three factors—Ascl1, Brn2, and Myt1l (ABM)—which, when overexpressed using lentiviruses, could convert mouse embryonic and postnatal fibroblasts into functional neurons rapidly and efficiently in vitro. Further studies demonstrated that these induced neuron cells expressed multiple neuron-specific proteins, generated action potentials, and formed functional synapses. One year later, the same group demonstrated the direct neural reprogramming of human fibroblasts in vitro using the ABM factors plus the basic helix-loop-helix transcription factor NeuroD1.²⁰ The human fibroblast–derived cells by ABMN factors could form functional synapses and integrated into pre-existing neuronal networks, which indicated that they had potential to be biologically functional and become a cell resource for cell replacement therapy. More recently, they further identified that Ascl1 alone was sufficient to reprogram mouse and human fibroblasts into functional neurons, most of which were excitatory.²¹ Thus, Ascl1 was revealed as the key driver of the direct reprogramming of fibroblasts toward functional neurons. Compared with the induced neurons derived from ABM factors, Ascl1-induced neurons were predominantly excitatory, which hinted at the possibility of generating certain subtypes of neurons.

Considering that fibroblasts and neural crest lineages share the same germ layer, the possibility of direct neural reprogramming of cell types derived from different germ layers remained uncertain, and generated considerable research interest. Recent studies demonstrated that besides fibroblasts, many kinds of somatic cells are capable of being converted into functional neurons, including astrocytes, hepatocytes, adipocyte progenitors, and human urine cells (as shown in Table 1).^{22

–29} The direct conversion of mouse terminally differentiated hepatocytes²² and human hematopoietic cells²⁵ into neurons indicated that somatic cells derived from other germ layers, and even terminally differentiated cells, could act as source cells to generate neurons. In addition, adipocyte progenitor cells may also be used as an alternative resource for neuronal induction. Such cells are easily accessed and show steady proliferation in vitro compared with fibroblasts.²⁴ Zhang et al.²⁶ achieved neural conversion from human urine cells, which represented a noninvasive approach to obtaining source cells. In vivo direct reprogramming holds significant promise for cell replacement therapy of neurodegenerative diseases. Ultimately, whether the somatic cells in the human brain could be converted directly into functional neurons is a key question. One attempt at direct reprogramming of astrocytes revealed that they could be converted into neurons.²⁹ Moreover, Karow et al.²⁸ demonstrated that induced neurons could be generated from the cells from the adult human cerebral cortex, which expressed characteristic pericyte markers. These results indicated the possibility of direct reprogramming of endogenous cells in the human brain toward functional neurons and provided the basis for in vivo conversion. Surprisingly, Zhao et al.²³ demonstrated that human glioma cells could be converted into neurons, after which the proliferation of tumor cells was dramatically inhibited in vitro and in vivo, which might represent a new approach to treat human brain tumors.

Table 1.

The source of somatic cells to be reprogrammed into neurons

Reference	Source cells	Reprogramming factors	Induced neurons
Vierbuchen et al.¹⁹	Mouse embryonic and postnatal fibroblasts	Ascl1, Brn2, Myt1l	Excitatory for majority
Pang et al.²⁰	Human fetal and postnatal fibroblasts	Ascl1, Brn2, Myt1l, NeuroD1	Mixed (glutamatergic over half)
Marro et al.²²	Mouse terminally differentiated hepatocytes	Ascl1, Brn2, Myt1l	Not reported
Zhao et al.²³	Human glioma cells	Ascl1, Brn2, Ngn2	Not completely mature
Yang et al.²⁴	Mouse adipocyte progenitor cells	Ascl1, Brn2, Myt1l	Glutamatergic for 90% and GABAergic for 10%
Castano et al.²⁵	Human hematopoietic cells	SOX2 and c-MYC using Sendai virus	Mixed (glutamatergic and GABAergic)
Zhang et al.²⁶	Human urine cells	Ascl1, Brn2, NeuroD, c-MYC, Myt1l	Mixed (glutamatergic and GABAergic)
Hu et al.²⁷	Human retinal pigmented epithelial cells	SOX2 delivered by RPARPAR	Not reported
Karow et al.²⁸	Human cerebral pericyte-derived cells	SOX2 and Mash1	Glutamatergic
Addis et al.²⁹	Mouse astrocytes	Ascl1, Lmx1b, Nurr1	Dopaminergic

Direct Reprogramming into Functional Neurons using microRNAs and Small Molecules

The mechanism of the direct reprogramming of differentiated somatic cells toward neurons via overexpression of cell lineage–specific transcription factors might be the promotion of neurogenic signaling pathways. In addition to transcription factors, various signaling molecules play important roles in neurogenesis. However, it remains to be determined whether they could be used in the direct reprogramming process. In 2011, Yoo et al.³⁰ demonstrated microRNA (miRNA)-mediated conversion of human fibroblasts to neurons. MicroRNA-9* (miR-9*) and microRNA-124 (miR-124) play an important role in neuronal differentiation and function by inducing compositional changes of BAF complexes. The expressions of miR-9/9* and miR-124 in human fibroblasts were sufficient to reprogram them into functional neurons. This process was aided by neurogenic transcription factors, but was ineffective in the absence of miRNAs. Furthermore, the combination of miRNAs and transcription factors in human neural conversion has been reported. When combined with Myt1l and Brn2, miR-124 could accomplish the direct reprogramming of postnatal and adult human primary dermal fibroblasts into functional neurons. The introduction of miRNAs provides an alternative approach for direct neural conversion and enables a deeper understanding of the underlying mechanisms.

Besides miRNAs themselves, the processes regulated by microRNAs also play significant roles in neuronal differentiation. For instance, the polypyrimidine tract binding (PTB) protein was used in direct neural reprogramming. PTB induces a programmed switch in the process of neuronal differentiation and miR-124 reduces the expression of PTB to control the switch. The repression of PTB, which is similar to inducing the function of miR-124, proved to be sufficient to generate functional neurons from fibroblasts.³¹ This paved the way for the direct reprogramming of neurons and revealed that enthetic interventions to control gene expression in specific cell types might alter cell fate. Further studies are needed to identify further intervention strategies to achieve more efficient and safe direct neuronal reprogramming.

Although the introduction of transcription factors and microRNAs induced the direct reprogramming of somatic cells toward neurons, several disadvantages have limited their application, such as low efficiency, technical challenges, and the introduction of transgenes. Therefore, a safer and more feasible strategy for direct neural conversion is needed. The introduction of small molecules, as replacements for transcription factors, was reported to be sufficient to generate pluripotent stem cells from mouse somatic cells. Considering the convenience of using small molecules, they represent an alternative strategy for neural generation. Using six chemical compounds (inhibitors of TGF-β, BMP, GSK-3β, MEK-ERK, and P53, respectively, and Forskolin), postnatal human fibroblasts were reprogrammed into neuronal cells with >80% efficiency.³² Using a cocktail of seven small molecules, human fibroblasts, even those from Alzheimer's disease patients, were reprogrammed efficiently into functional neurons, which resembled the neurons induced from iPSCs and fibroblasts using transcription factors.³³ More recently, the combined application of transcription factors plus small molecules was assessed for direct neural conversion from fibroblasts. The combination of Forskolin and Ascl1 induced approximately 80% of the mouse fibroblasts into parvalbumin neurons, only a few of which were generated using the conventional strategy.³⁴

Although the successful neural conversion of fibroblasts by small molecules has been reported, the conversion of other cell types using small molecules remains uncertain. Cheng et al. demonstrated that a cocktail of small molecules (VPA, CHIR99021, and Repsox) could reprogram astrocytes directly into functional neurons in vitro, which might be achieved through the activation of NeuroG2 and NeuroD1 expression.³⁵ Although direct reprogramming to generate functional neurons through the overexpression of neurogenic transcription factors has been achieved for hepatocytes and hematopoietic cells, more studies are needed to assess the feasibility of using small molecules to convert these somatic cells into neurons.

The precise expression of genes is governed by transcription factors, microRNAs, and signaling pathways. These factors jointly and concordantly control the expression of genes and therefore determine the specific cell fate. Intervention to change the process might change cell fate, which is the main basis of direct neural conversion through transcription factors, microRNAs, and small molecules. Among these approaches, small molecules have several advantages. For example, the application of small molecules is convenient and reversible, which is beneficial for research and clinical practice. Moreover, small molecules are vector-free, which avoids the introduction of transgenes from viral vectors. Additionally, small molecules can be used at various concentrations to provide precise effects. Therefore, their application could be a convenient, safe, and precise approach for direct reprogramming.

As mentioned above, many protocols, including the introduction of transcription factors, microRNAs, and small molecules, have achieved direct neural reprogramming of somatic cells without a stem-cell state. Although these protocols contain different materials and techniques, they have some commonalities concerning their generation and mechanisms. For instance, the specific transcription factors, microRNAs, and small molecules have been identified from a pool of candidate factors that play significant roles in neural differentiation. From a total of 19 candidates that were selected because of their specific expression in neural cell types, Vierbuchen et al.¹⁹ identified the ABM factors. The generation of neurons in the human brain is a complex process, and the proneural genes, such as Ascl1, contribute to promoting cell cycle exit and neural differentiation in neural progenitor cells.³⁶ MicroRNAs miR-9/9* and miR-124 regulate the neurogenesis and neuronal functions by targeting the subunits of BAF complexes and other essential genes, including components of the REST complex.³⁰ Similarly, small molecules applied in direct neural reprogramming are involved in regulating neural differentiation. For example, Forskolin can induce functional neural differentiation of human adipose tissue–derived stem cells when combined with β-EGF.³⁷ Therefore, the genes or signaling pathways that are involved in neural differentiation have become key factors in direct neural reprogramming. In the past few years, researchers have developed several protocols for this technique. However, these protocols are usually generated from a couple of candidates after a wide screening program. Among recent studies, it is not clear why the reported factors could induce functional neural reprogramming and why other candidates could not. Moreover, limitations of screening make it unclear whether other more efficient factors exist for functional neural reprogramming. Therefore, this represents a bottleneck for the further development of direct neural reprogramming. To overcome this problem and to develop accurate and gold-standard protocols, further studies are needed to explore the underlying mechanisms and identify specific targets. Furthermore, whether the applications of these factors are transient treatments becomes another question for this technology. Fortunately, Pang et al.²⁰ demonstrated that induced neurons were stable without continued transgene expression as the mRNA expression levels of related endogenous genes were rapidly induced and increased over time, even after dox withdrawal. Importantly, neural identity of the induced neurons using microRNAs was assessed to be stable after the removal of exogenous expression of miR-9/9*-124.³⁰ Even though induced neurons could maintain neural identity for several weeks after the treatment of transcription factors and microRNAs, some questions remain. For example, in recent studies, it was proved that induced neurons could maintain the identity for several weeks after the withdrawal of treatment. However, several weeks still fails to meet the need for clinical practice. Which interventions could help to extend the time limitations becomes another question. Moreover, the application of small molecules is not a transient treatment: somatic cells were treated with these molecules for several weeks until their neural characteristics were detected.³³ It remains uncertain whether the induced neurons can act as normal and functional neurons after withdrawal of the molecule treatment. If not, long-term treatment could become a formidable challenge for the clinical application of this technology because of technological difficulties and safety concerns.

Direct Neural Reprogramming in vivo

The direct neural reprogramming of somatic cells in vitro provides a new strategy to treat neurodegenerative diseases by cell replacement therapy and helps disease modeling. However, the utility of this approach in vivo, which is the key step for its clinical application, remains unclear. Torper et al.³⁸ engineered human fibroblasts and human astrocytes to express neural transcription factors (ABM factors) in vitro and then transplanted them into the striatum and hippocampus of adult rats. When the transcription factors were activated after transplantation, these cells were converted directly into neurons. Meanwhile, they found that the endogenous astrocytes of a transgenic mouse model, which specifically expressed the ABM factors, were converted into neurons in situ. These results revealed that direct neural reprogramming could be achieved in the rodent brain, which is the theoretical basis of in vivo experiments.

Considering the reverse generation of reactive gliosis and scarring after brain injury and in neurodegenerative diseases, which occurs frequently following the loss of neurons, the in vivo direct reprogramming of glia to neurons has become a hot research topic and significant progress has been made in recent years. Guo et al.³⁹ demonstrated that using retroviral expression of a neural transcription factor, NeuroD1, glial cells in the cortex of stab-injured and Alzheimer's disease model mice could be converted into functional neurons. In addition to the brain, such an approach could also be used for spinal cord injury. Su et al.⁴⁰ demonstrated that in an adult mouse model of an injured spinal cord, the resident astrocytes were reprogrammed into neuroblasts after the ectopic expression of SOX2. Importantly, these neuroblasts matured into functional neurons after subsequent exposure to a histone deacetylase inhibitor, valproic acid (VPA), in vivo.

Although in vivo direct reprogramming has been achieved, there remain several shortcomings. For example, previous studies revealed that in addition to the expression of neural transcription factors, cocktails of small molecules might also induce the direct neural conversion of somatic cells in vitro, and the latter have more advantages compared with transcription factors. However, at present, the main approach to achieving in vivo reprogramming remains the introduction of transcription factors; the possibility of using small molecules to generate functional neurons remains uncertain. Additionally, in the above-mentioned studies, the induced neurons were observed without relief of symptoms. More recently, Chen et al. ⁴¹ demonstrated that the lentiviral expression of ABN factors (Ascl1, Brn2, Ngn2) in AD mouse models relieved Alzheimer's disease symptoms, with improved spatial learning and memory in the water maze test and altered expression levels of inflammatory antibodies. Unfortunately, they failed to assess the generation of neurons in the brain, and it was unclear whether the relief of Alzheimer's disease symptoms resulted from direct neural conversion. Therefore, further studies that link in vivo direct neural conversion with the observed changes in clinical symptoms are required.

Direct Reprogramming into Functional DA Neurons

The human nervous system is a complicated network comprising various specific subtypes of neurons. Each subtype has certain functions, and their deficiency might lead to different diseases. Pfisterer et al. ⁴² reported that distinct functional subtypes of neurons could be induced directly from fibroblasts when the appropriate transcription cues were provided via the ABM factors. Recently, several subtypes of functional neurons have been induced directly by reprogramming somatic cells, including DA neurons, spinal motor neurons, GABAergic neurons, peripheral sensory neurons, serotonergic neurons, and parvalbumin neurons.^{34,42

–46} Considering the important roles of DA neurons in PD, direct reprogramming to generate DA neurons selectively is mainly discussed.

PD is a common age-relative neurodegenerative disease with the pathological hallmarks of selective loss of DA neurons in the midbrain. At present, several clinical approaches are used to manage the symptoms of PD, including levodopa and deep brain stimulation (DBS). However, these methods fail to stop or slow the progression of diseases and are ineffective for advanced PD patients. Given the pathological hallmarks of PD, cell replacement and transplantation of DA neurons hold promise for disease treatment. In fact, it has been reported that the transplantation of ESC-derived DA neurons could restore the motor symptoms induced by dopamine deficiency.⁴⁷ Direct reprogramming to generate induced DA (iDA) neurons represents an alternative approach for cell replacement therapy. By combining the expression of ABM factors with two additional DA-related genes—LMX1A and FOXA2—Pfisterer et al.⁴² converted human embryonic and postnatal fibroblasts successfully into functional iDA neurons with the positive TH staining. Without the use of ABM factors, Caiazzo et al.⁴⁸ identified a minimal set of three transcription factors—Mash1, Nurr1, and Lmx1a—which were sufficient to generate functional iDA neurons from mouse and human fibroblasts. Importantly, it was also achieved in prenatal and adult fibroblasts from PD patients. More recently, it was reported that the iDA neurons induced by defined transcription factors in mouse tail tip fibroblasts alleviated symptoms when transplanted into a mouse model of PD.⁴⁹ In addition, Liu et al.⁵⁰ showed that a combination of five transcription factors—Mash1, Ngn2, SOX2, Nurr1, and Pitx3—could reprogram human fibroblasts directly into iDA neurons in vitro, and after transplantation they induced symptomatic relief in a rat PD model.

In the studies mentioned above, Kim and Liu et al. detected the relief of the PD symptoms in mouse and rat PD models, respectively. However, the degree of alleviation and the promise they provide in clinical application require discussion. In detail, they found that compared with controls, mouse and rat PD models injected with iDA neurons showed a significant stabilization in rotational behaviors. Further staining experiments indicated that animals transplanted with induced cells contained large number of DA neurons with positive TH staining, while the control models exhibited a significant reduction of DA neurons after PD modeling. This evidence proved the relief symptoms such as rotational behaviors, and indicated that induced DA neurons can survive in PD models for a long time. However, PD is a complex disease with various symptoms. The rotational test assay cannot represent a clinical phenotype of PD. Therefore, other behavioral assays to assess the ability of induced DA neurons in suppressing PD symptoms are needed. Meanwhile, standardization of neurons induced in vitro or in vivo that have the potential to function in the complex neural circuitry remains a significant question. Recent studies that created induced neurons detected several neural characteristics of the cells and demonstrated that they were functional in expressing multiple neuron-specific proteins, generating action potentials and forming functional synapses. However, neurons have complex functions in the human brain, and form circuits that control the activities of almost the whole body. Furthermore, there might be some neural functions that are currently unknown. Thus, the neural functions of the induced cells need to be further assessed, and it remains uncertain whether they can function in complex circuits.

Despite the generation of functional iDA neurons and the relief of symptoms after transplantation into an animal PD model, direct reprogramming still shows low efficiency (<10% of transplanted cells⁴⁹), which represents a barrier to its clinical practice. Therefore, it is necessary to increase the efficiency of direct reprogramming from human fibroblasts to generate functional iDA neurons. Considering the important role of some tumor-related genes, such as p53, in the generation of iPSCs, Liu et al.⁵¹ assessed the effect of p53 suppression on the direct reprogramming of iDA neurons, using a lentivirus-encoded dominant-negative p53 (p53-DN) gene. Compared with the direct reprogramming induced by transcription factors alone, their combination with p53-DN dramatically increased the conversion efficiency by 5–20-fold in IMR90 cells. The suppression of p53 enhanced the reprogramming by improving cellular survival rather than causing cell proliferation, which avoids the higher cancer risk caused by p53 inhibition. More surprisingly, it was reported that depletion of p53 alone could convert fibroblasts into three neural lineages.⁵² Recently, Jiang et al.⁵³ observed an increased conversion efficiency resulting from p53 inhibition in the direct reprogramming of human fibroblasts to iDA neurons by Ascl1, Nurr1, Lmx1a, and miR124. Furthermore, they demonstrated that another two factors could affect the direct reprogramming and increase the efficiency. First, they found that cell cycle exit was required for the reprogramming of fibroblasts toward neurons, as G1 arrest was induced by serum withdrawal, and the CDK2 inhibitor SU9516 and the mTOR inhibitor Torin1 remarkably enhanced the process. Moreover, they demonstrated that during reprogramming, cells required an appropriate extracellular environment. In summary, they proposed that Tet1 plays a critical role in the conversion progress because the conditions that enhanced the conversion efficiency induced Tet1 synergistically. Above all, certain conditions that alter cell signaling pathways are beneficial for the direct reprogramming of human fibroblasts toward iDA neurons. As well as biochemical stimulations, biophysical cues might also modulate cellular behavior. Yoo et al.⁵⁴ reported that nanoscale biophysical stimulation could promote the direct reprogramming of fibroblasts to generate iDA neurons. Specific histone modifications and transcriptional changes related to the mesenchymal-to-epithelial transition might contribute to this process.

Challenges and Prospects

This review highlights the recent discoveries and progress in direct reprogramming of somatic cells to neurons. A variety of studies have generated functional neurons via the introduction of transcription factors, microRNAs, or small molecules. In addition, the direct neural reprogramming can induce certain subtypes of neurons, which meet the need for specific diseases. Furthermore, recent studies demonstrated that functional neurons can be induced in vivo and can alleviate several symptoms caused by neurodegenerative diseases. Thus, direct neural reprogramming holds the promise for cell replacement therapy for these diseases.

Although direct reprogramming to generate certain subtypes of neurons is a novel strategy for cell replacement therapy for neurodegenerative diseases, there remain certain challenges for the further application and development of this approach. The underlying mechanisms of fate conversion at the gene or protein level remain unclear. The p53 signal pathway plays a significant role in direct neural conversion, and its suppression increases the conversion efficiency, even leading to the conversion of fibroblasts into three neural lineages.^51
–53 However, its specific function in cell fate conversion is unclear. The close relationship between p53 mutation and carcinogenesis⁵⁵ raises the question of whether carcinogenesis is involved in the direct reprogramming process. In addition to p53, several signaling pathways and proteins are involved in direct reprogramming, such as the Wnt/β-catenin,⁵⁶ TGF-β,⁵⁷ and polycomb repressive complex2 pathways.⁵⁸ Therefore, further research into the involvement of these pathways in the mechanisms of direct reprogramming is required. The induced neurons are similar to the endogenous neurons in some characteristics and functions. However, the cellular features examined in previous studies remain limited, and how much the induced neurons resemble the endogenous neurons is a key point in the further application and development of this approach. Additionally, there are technical and safety concerns for its application in humans. Genomic insertions can result in dysfunction and even tumorigenesis. Therefore, producing functional neurons without genomic insertion is important. Adler et al.⁵⁹ delivered plasmids encoding ABM factors into primary mouse embryonic fibroblasts using a bioreducible linear polymer (amido amine) and generated induced neurons that were observed to fire multiple and spontaneous action potentials. These results demonstrated the possibility of non-viral direct neural reprogramming, and hold the promise for a safer technique to generate functional neurons. More studies are needed to develop this non-viral reprogramming.

In summary, direct neural reprogramming is a promising technique for cell replacement therapy and has attracted considerable research attention. However, further studies are needed for the safety and efficiency of this approach, including in vivo reprogramming.

Footnotes

Acknowledgments

This work was supported by the National Nature Science Foundation of China (81400933 and 81520108010) and the Zhejiang Medical Science and Technology Plan project (2016KYB119).

Author Disclosure

The authors declare that there is no conflict of interest regarding the publication of this article and the funding that they have received.

References

Budd Haeberlein

, Harris

. Promising targets for the treatment of neurodegenerative diseases. Clin Pharmacol Ther, 2015; 98:492–501.

Koerts

, Konig

, Tucha

, et al. Working capacity of patients with Parkinson's disease—a systematic review. Parkinsonism Relat Disord, 2016; 27:9–24.

Thomson

, Itskovitz-Eldor

, Shapiro

, et al. Embryonic stem cell lines derived from human blastocysts. Science, 1998; 282:1145–1147.

Bjorklund

, Sanchez-Pernaute

, Chung

, et al. Embryonic stem cells develop into functional dopaminergic neurons after transplantation in a parkinson rat model. Proc Natl Acad Sci U S A, 2002; 99:2344–2349.

Takahashi

, Yamanaka

. Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell, 2006; 126663–676.

Sanchez-Danes

, Consiglio

, Richaud

, et al. Efficient generation of a9 midbrain dopaminergic neurons by lentiviral delivery of lmx1a in human embryonic stem cells and induced pluripotent stem cells. Hum Gene Ther, 2012; 23:56–69.

Miura

, Okada

, Aoi

, et al. Variation in the safety of induced pluripotent stem cell lines. Nat Biotechnol, 2009; 27:743–745.

Aoi

, Yae

, Nakagawa

, et al. Generation of pluripotent stem cells from adult mouse liver and stomach cells. Science, 2008; 321:699–702.

Okabe

, Otsu

, Ahn

, et al. Definitive proof for direct reprogramming of hematopoietic cells to pluripotency. Blood, 2009; 114:1764–1767.

10.

Hanna

, Markoulaki

, Schorderet

, et al. Direct reprogramming of terminally differentiated mature B lymphocytes to pluripotency. Cell, 2008; 133:250–264.

11.

Ieda

, Fu

, Delgado-Olguin

, et al. Direct reprogramming of fibroblasts into functional cardiomyocytes by defined factors. Cell, 2010; 142:375–386.

12.

Nam

, Lubczyk

, Bhakta

, et al. Induction of diverse cardiac cell types by reprogramming fibroblasts with cardiac transcription factors. Development, 2014; 141:4267–4278.

13.

Jayawardena

, Egemnazarov

, Finch

, et al. Microrna-mediated in vitro and in vivo direct reprogramming of cardiac fibroblasts to cardiomyocytes. Circ Res, 2012; 110:1465–1473.

14.

Ifkovits

, Addis

, Epstein

, et al. Inhibition of TGFβ signaling increases direct conversion of fibroblasts to induced cardiomyocytes. PLOS One, 2014; 9:e89678.

15.

Qian

, Huang

, Spencer

, et al. In vivo reprogramming of murine cardiac fibroblasts into induced cardiomyocytes. Nature, 2012; 485:593–598.

16.

Jayawardena

, Finch

, Zhang

, et al. Microrna induced cardiac reprogramming in vivo: evidence for mature cardiac myocytes and improved cardiac function. Circ Res, 2015; 116:418–424.

17.

Huang

, Zhang

, Gao

, et al. Direct reprogramming of human fibroblasts to functional and expandable hepatocytes. Cell Stem Cell, 2014; 14:370–384.

18.

Song

, Pacher

, Balakrishnan

, et al. Direct reprogramming of hepatic myofibroblasts into hepatocytes in vivo attenuates liver fibrosis. Cell Stem Cell, 2016; 18:797–808.

19.

Vierbuchen

, Ostermeier

, Pang

, et al. Direct conversion of fibroblasts to functional neurons by defined factors. Nature, 2010; 463:1035–1041.

20.

Pang

, Yang

, Vierbuchen

, et al. Induction of human neuronal cells by defined transcription factors. Nature, 2011; 476:220–223.

21.

Chanda

, Ang

, Davila

, et al. Generation of induced neuronal cells by the single reprogramming factor ASCL1. Stem Cell Rep, 2014; 3:282–296.

22.

Marro

, Pang

, Yang

, et al. Direct lineage conversion of terminally differentiated hepatocytes to functional neurons. Cell Stem Cell, 2011; 9:374–382.

23.

Zhao

, He

, Zhou

, et al. Neuronal transcription factors induce conversion of human glioma cells to neurons and inhibit tumorigenesis. PLoS One, 2012; 7:e41506.

24.

Yang

, Jiao

, Gao

, et al. Direct conversion of adipocyte progenitors into functional neurons. Cell Reprogram, 2013; 15:484–489.

25.

Castano

, Menendez

, Bruzos-Cidon

, et al. Fast and efficient neural conversion of human hematopoietic cells. Stem Cell Rep, 2014; 3:1118–1131.

26.

Zhang

, Ma

, Qian

, et al. Modeling neurological disease by rapid conversion of human urine cells into functional neurons. Stem Cells Int, 2016; 2016:2452985.

27.

, Chen

, Teesalu

, et al. Reprogramming human retinal pigmented epithelial cells to neurons using recombinant proteins. Stem Cells Transl Med, 2014; 3:1526–1534.

28.

Karow

, Sanchez

, Schichor

, et al. Reprogramming of pericyte-derived cells of the adult human brain into induced neuronal cells. Cell Stem Cell, 2012; 11:471–476.

29.

Addis

, Hsu

, Wright

, et al. Efficient conversion of astrocytes to functional midbrain dopaminergic neurons using a single polycistronic vector. PLoS One, 2011; 6:e28719.

30.

Yoo

, Sun

, Li

, et al. Microrna-mediated conversion of human fibroblasts to neurons. Nature, 2011; 476:228–231.

31.

Xue

, Ouyang

, Huang

, et al. Direct conversion of fibroblasts to neurons by reprogramming PTB-regulated microRNA circuits. Cell, 2013; 152:82–96.

32.

Dai

, Harada

, Takamatsu

. Highly efficient direct conversion of human fibroblasts to neuronal cells by chemical compounds. J Clin Biochem Nutr, 2015; 56:166–170.

33.

, Qiu

, Guan

, et al. Direct conversion of normal and Alzheimer's disease human fibroblasts into neuronal cells by small molecules. Cell Stem Cell, 2015; 17:204–212.

34.

Shi

, Zhang

, Chen

, et al. Conversion of fibroblasts to parvalbumin neurons by one transcription factor, Ascl1, and the chemical compound forskolin. J Biol Chem, 2016; 291:13560–13570.

35.

Cheng

, Gao

, Guan

, et al. Direct conversion of astrocytes into neuronal cells by drug cocktail. Cell Res, 2015; 25:1269–1272.

36.

Castro

, Martynoga

, Parras

, et al. A novel function of the proneural factor ascl1 in progenitor proliferation identified by genome-wide characterization of its targets. Genes Devel, 2011; 25930–945.

37.

Jang

, Cho

, et al. Functional neural differentiation of human adipose tissue-derived stem cells using bFGF and forskolin. BMC Cell Biol, 2010; 11:25.

38.

Torper

, Pfisterer

, Wolf

, et al. Generation of induced neurons via direct conversion in vivo . Proc Natl Acad Sci U S A, 2013; 110:7038–7043.

39.

Guo

, Zhang

, Wu

, et al. In vivo direct reprogramming of reactive glial cells into functional neurons after brain injury and in an Alzheimer's disease model. Cell Stem Cell, 2014; 14:188–202.

40.

, Niu

, Liu

, et al. In vivo conversion of astrocytes to neurons in the injured adult spinal cord. Nat Commun, 2014; 5:3338.

41.

Chen

, Yan

, Wang

, et al. Transfer of three transcription factors via a lentiviral vector ameliorates spatial learning and memory impairment in a mouse model of Alzheimer's disease. Gene, 2016; 587:59–63.

42.

Pfisterer

, Kirkeby

, Torper

, et al. Direct conversion of human fibroblasts to dopaminergic neurons. Proc Natl Acad Sci U S A, 2011; 108:10343–10348.

43.

Son

, Ichida

, Wainger

, et al. Conversion of mouse and human fibroblasts into functional spinal motor neurons. Cell Stem Cell, 2011; 9:205–218.

44.

Colasante

, Lignani

, Rubio

, et al. Rapid conversion of fibroblasts into functional forebrain gabaergic interneurons by direct genetic reprogramming. Cell Stem Cell, 2015; 17:719–734.

45.

Blanchard

, Eade

, Szucs

, et al. Selective conversion of fibroblasts into peripheral sensory neurons. Nat Neurosci, 2015; 18:25–35.

46.

Vadodaria

, Mertens

, Paquola

, et al. Generation of functional human serotonergic neurons from fibroblasts. Mol Psychiatry, 2016; 21:49–61.

47.

Kim

, Auerbach

, Rodriguez-Gomez

, et al. Dopamine neurons derived from embryonic stem cells function in an animal model of Parkinson's disease. Nature, 2002; 418:50–56.

48.

Caiazzo

, Dell'Anno

, Dvoretskova

, et al. Direct generation of functional dopaminergic neurons from mouse and human fibroblasts. Nature, 2011; 476:224–227.

49.

Kim

, Su

, Wang

, et al. Functional integration of dopaminergic neurons directly converted from mouse fibroblasts. Cell Stem Cell, 2011; 9:413–419.

50.

Liu

, Li

, Stubblefield

, et al. Direct reprogramming of human fibroblasts into dopaminergic neuron-like cells. Cell Research, 2012; 22:321–332.

51.

Liu

, Huang

, Li

, et al. Enhancing the efficiency of direct reprogramming of human primary fibroblasts into dopaminergic neuron-like cells through p53 suppression. Sci China Life Sci, 2014; 57:867–875.

52.

Zhou

, Zhang

, He

, et al. Conversion of fibroblasts to neural cells by p53 depletion. Cell Rep, 2014; 9:2034–2042.

53.

Jiang

, Xu

, Zhong

, et al. Cell cycle and p53 gate the direct conversion of human fibroblasts to dopaminergic neurons. Nat Commun, 2015; 6:10100.

54.

Yoo

, Noh

, Kim

, et al. Nanogrooved substrate promotes direct lineage reprogramming of fibroblasts to functional induced dopaminergic neurons. Biomaterials, 2015; 45:36–45.

55.

Levine

, Finlay

, Hinds

. P53 is a tumor suppressor gene. Cell, 2004; 116:S67–69, 61 p following S69.

56.

Sanges

, Romo

, Simonte

, et al. Wnt/β-catenin signaling triggers neuron reprogramming and regeneration in the mouse retina. Cell Rep, 2013; 4:271–286.

57.

Wang

, Li

, Huang

, et al. TGFβ signaling regulates the choice between pluripotent and neural fates during reprogramming of human urine derived cells. Sci Rep, 2016; 6:22484.

58.

Yaqubi

, Mohammadnia

, Fallahi

. Predicting involvement of polycomb repressive complex 2 in direct conversion of mouse fibroblasts into induced neural stem cells. Stem Cell Res Ther, 2015; 6:42.

59.

Adler

, Grigsby

, Kulangara

, et al. Nonviral direct conversion of primary mouse embryonic fibroblasts to neuronal cells. Mol Ther Nucleic Acids, 2012; 1:e32.