Abstract
With the discovery and development of somatic cell nuclear transfer, cell fusion, and induced pluripotent stem cells, cell transdifferentiation research has presented unique advantages and stimulated a heated discussion worldwide. Cell transdifferentiation is a phenomenon by which a cell changes its lineage and acquires the phenotype of other cell types when exposed to certain conditions. Indeed, many adult stem cells and differentiated cells were reported to change their phenotype and transform into other lineages. This article reviews the differentiation of stem cells and classification of transdifferentiation, as well as the advantages, challenges, and prospects of cell transdifferentiation. This review discusses new research directions and the main challenges in the use of transdifferentiation in human cells and molecular replacement therapy. Overall, such knowledge is expected to provide a deep understanding of cell fate and regulation, which can change through differentiation, dedifferentiation, and transdifferentiation, with multiple applications.
Introduction
In 1981

Schematic of mouse embryo development and ESC differentiation. Solid arrows represent embryo development or direct acquisition. Dotted arrows represent cell differentiation or experimental treatment. *Indicates an important constituent. D, differentiated cell; ESC, embryonic stem cell; TSC, totipotent stem cell; PC, progenitor cell or precursor cell; U, unipotent stem cell.
Therefore, this pivotal experiment demonstrated that, under suitable conditions, cell differentiation could be altered, thereby breaking the past view that cell differentiation is irreversible. In 2007, researchers successfully generated human iPSCs (Takahashi et al., 2007; Yu et al., 2007), showing that even the phenotypes of human cells can be altered. However, despite these advances, the application of iPSCs still faces many challenges, such as inefficient transformation, iPSC immunogenicity and tumorigenicity, and incomplete cell differentiation (Yamashita et al., 2011).
Cell transdifferentiation is a phenomenon wherein cells lose their unique characteristics and change into another cell type. ASCs can generate various differentiated cells, and these differentiated cells can, in some cases, change their phenotype. These unique features highlight the potential importance of cell transdifferentiation research in the development of various treatments. In this review, the definition of ASCs and the classification of cell transdifferentiation are discussed. Further, the advantages, challenges, and prospects of cell transdifferentiation are also summarized, aiming to provide new insights for future research.
Differentiation of Stem Cells
Stem cells are a class of self-renewing and highly proliferative pluripotent cell population. According to their developmental stage or the different sources, stem cells can be divided into ESCs and ASCs. ASCs are present in a very small number of tissues, and are usually in the process of differentiation into first intermediate cells, called precursor cells or progenitor cells (Laplane and Solary, 2019).
ASCs are currently considered to possess self-renewal, and have substantial potential for clonogenic cell division, as well as ultimately being responsible for cell replacement and tissue regeneration. Under certain conditions, ASCs differentiate into new functional cells, thereby aiding tissues and organs in maintaining their function (Busch et al., 2023). It is well known that stem cells can self-renew, and through the process of cell proliferation, daughter cells can maintain proliferation ability. The self-renewal ability of stem cells occurs through asymmetric division. Asymmetric division of a stem cell creates a stem cell and a transient proliferation cell or directional progenitor cell. Each cell has different developmental potential as some proteins are not equally divided between the two daughter cells (Venkei and Yamashita, 2018).
Compared with stem cells, transient proliferation cells and directional progenitor cells have a more rapid, but short-term proliferation. Stem cells have capacity to proliferate infinitely, whereas the proliferative ability of transient proliferation cells and directional progenitor cells is finite, and the progenitor cells may give preference to proliferation (Krishnan et al., 2021). Figure 2 demonstrates the differentiation of stem cells into unipotent stem cells, and finally into terminally differentiated cells (Clevers and Watt, 2018).

Schematic of the proliferation and differentiation processes of ASCs. ASC, adult stem cell; D, differentiated cell; PC, progenitor cell or precursor cell; U, unipotent stem cell. Thick arrows indicate cell self-renewal and thin arrows indicate cell differentiation.
To study ASCs, their identification is crucial. ASC shape and location can be used to accurately identify some cells, such as Drosophila gonadal cells and peripheral nervous system and mammalian muscle stem cells. However, in most cases, stem cell location is unclear, and molecular markers must be used to identify stem cells (Laplane and Solary, 2019), for example, integrin β1 is a marker of epidermal stem cells (Zhang et al., 2010). However, there are many stem cells that do not have specific markers, rendering it difficult to isolate and study them (Mobasseri et al., 2019). Although many stem cells have been isolated, the methods of purification are limited; to achieve good-quality cells, a combination of methods is used for purification (Shintani and Higuchi, 2022).
Classification of Transdifferentiation
Previous studies have reported that many adult cells can change their phenotype and transdifferentiate into other cell lineage (Masip et al., 2010). According to the transdifferentiation of cell types, it divided into transdifferentiation of ASCs and differentiated cells.
Transdifferentiation of nerve cells
In 2010, the researchers determined that only Ascl1, Brn2 (also known as Pou3f2), and Myt1l were needed, enough to quickly and efficiently convert mouse embryos and postnatal fibroblasts into functional neurons in vitro. These induced neuronal cells express multiple neuron-specific proteins, generate action potentials, and form functional synapses (Vierbuchen et al., 2010). In 2014, Chen Gong and colleagues (Guo et al., 2017) reported the use of a single neural TF, NeuroD1, to achieve in situ transdifferentiation of glial cells in the brain into neurons (Guo et al., 2014). The other study found that NeuroD1 expression induces cell death in microglia (Rao et al., 2021). In 2022, the authors further pointed out the reasons why other teams failed to achieve direct transdifferentiation of astrocytes into neurons and provided strong evidence, including two-photon in vivo imaging data demonstrating the direct conversion of astrocytes into neurons (Xu et al., 2022).
Transdifferentiation of epidermal stem cells
Toma et al. identified and isolated a new stem cell type from the dermis, named skin-derived precursor cells. On polylysine-coated culture flasks, depending on the serum concentration of the medium, skin-derived precursor cells can be induced to differentiate into various cells, including smooth muscle cells, adipocytes, and neuronal cells (Toma et al., 2001). Further studies confirmed that epidermal stem cells can be amplified in in vitro culture and differentiated into neurons, glial cells, smooth muscle cells, and adipocytes (Kawase et al., 2004).
Transdifferentiation of muscle stem cells
Numerous studies have confirmed the existence of skeletal muscle pluripotent stem cells, which can differentiate into muscle, fat, bone, and cartilage cells, as well as into hematopoietic stem cells. For example, Wada et al. induced the transdifferentiation of stem cells from adult mouse skeletal muscle in vitro, showing that muscle stem cells can differentiate into muscle, bone, and fat cells (Wada et al., 2002).
Transdifferentiation of adipose stem cells
Guo et al. (2018) found that adipose tissue is rich in ASCs, and these can be induced to differentiate into nerve, muscle, bone, and other cell types. Further studies have demonstrated that adipose and hematopoietic stem cells can be cultured together in vitro (Cousin et al., 2016; Gavin et al., 2021). Morizono et al. also demonstrated that adipose stem cells can differentiate into adipocytes and osteoblasts after viral transfection-mediated forced expression of exogenous genes (Morizono et al., 2003).
Transdifferentiation of pancreatic and liver cells
The pancreas and liver are formed from the mesoderm during embryonic development, and under certain conditions, the two cell types can be transformed into each other. Indeed, pancreatic duct ligation and partial pancreatectomy promote β-cell transdifferentiation not only from α-cells but also from duct and acinar cells in murine models (Wang and Zhang, 2021). In addition, deletion of the Aristaless-related homeobox gene or Pdx1 overexpression promotes the differentiation of β-cells from α-cells (Chakravarthy et al., 2017; Xiao et al., 2018).
The transdifferentiation of liver cells is related to a permissive epigenome, which can be extended to liver transdifferentiation-resistant cells by specific soluble factors. Restoration of the pancreatic niche and vasculature promotes the maturation of liver cell transdifferentiation to mimic β-cell function (Meivar-Levy and Ferber, 2019). Xie Xin and colleagues (Guo et al., 2014) reported that a combination of small-molecule compounds can make mouse fibroblasts transdifferentiate into hepatocytes under the action of a TF (Guo et al., 2017). These transdifferentiated hepatocytes have been demonstrated to have hepatocyte characteristics both in vitro and in vivo, thereby representing a potential therapeutic strategy for liver failure (Guo et al., 2017).
Transdifferentiation of bone marrow cells
Bone marrow stem cells can be classified as hematopoietic or mesenchymal stem cells. Hematopoietic stem cells produce blood and immune cells under normal circumstances, whereas mesenchymal stem cells generate chondrocytes and osteoblasts. Mesenchymal stem cells are the most studied and one of the most representative models of stem cells, as they hold the potential to differentiate into most cells. By activating the EPO-R signaling pathway, EPO induces osteogenesis and the endothelial transdifferentiation of multifunctional mesenchymal stem cells, leading to bone remodeling, angiogenesis, and the secretion of macrotrophic factors (Tsiftsoglou, 2021). Reyes et al. demonstrated that after a long period of in vitro culture, bone marrow mesenchymal stem cells can differentiate into bone cells, chondrocytes, myoblasts, adipocytes, or endothelial cells under appropriate conditions (Rietze et al., 2001). Purified bone marrow hematopoietic stem cells were shown to differentiate into liver, lung, gastrointestinal tract, skin, and other epithelial cells (Krause et al., 2001).
Advantages of cell transdifferentiation
ESC research has played an important role in regenerative medicine, as well as for disease treatment and the establishment of disease models. However, it has many relevant limitations. First, ESC isolation represents important ethical challenges as the establishment of ESC lines requires the destruction of an embryo. Second, when ESC algebra is high, there is a high chance for their genome to harbor mutations and/or chromosomal abnormalities, which can seriously affect the quality of the cells. This can also exert detrimental effects on the cell types generated and the experimental results obtained.
Moreover, the current criteria to evaluate ESC quality are insufficient (Chen and Shao, 2022). The third and most critical issue is the transformation of ESCs into tumor cells. It is necessary to ensure that ESCs used in cell therapy completely differentiate only into cells with physiological functions; otherwise, the risk of cancer development will increase (Nussbaum et al., 2007).
In contrast to ESCs, iPSCs can be differentiated into phenotypic cells, such as cardiac and neuronal cells, without ethical concerns (Kwok et al., 2022; Nicholson et al., 2022). Although research on iPSCs has substantially advanced in recent years, many challenges still remain, such as tumorigenicity, immune rejection, heterogeneity of iPSC lineage phenotypic cells and cell line variations, genetic instability, and lack of maturity in iPSC-derived phenotypic cells (Doss and Sachinidis, 2019). Since a method for generating iPSCs was described (Takahashi and Yamanaka, 2006), iPSC technology will become a prominent choice for the treatment of human genetic diseases and screening of drugs (Nicholson et al., 2022). iPSCs have better breadth and flexibility than ASCs, as they can differentiate into various cells associated with a particular trait.
Nevertheless, iPSC use is still limited by their low efficiency, possible immunogenicity, and tumorigenicity (Yamashita et al., 2011). Although iPSCs avoid the ethical problems of ESCs, the reliability of their source and technology security are still the biggest obstacles for their application (Wang et al., 2020). Therefore, research is ongoing on some key cell reprogramming factors to ensure that adult differentiated cells can be directly modulated without requirement to return them to a pluripotent state, a process known as direct reprogramming (Wang et al., 2021).
Indeed, direct transdifferentiation represents an attractive treatment strategy to modulate cell fate in situ. In 2008, Zhou et al. reported that TFs could be used to transform adult mouse pancreatic exocrine cells into an insulin-producing endocrine cell line (Zhou et al., 2008). Directly transdifferentiated cells (DTCs) represent a potentially new therapeutic strategy that requires relatively simple and safe procedures. More importantly, DTCs undergo differentiation directly in the body, thus representing a possible method for the early therapeutic intervention and prevention of disease progression. The advantages and disadvantages of ESCs, iPSCs, ASCs, and DTCs are summarized in Table 1.
Advantages and Disadvantages of Embryonic Stem Cells, Induced Pluripotent Stem Cells, Adult Stem Cells, and Directly Transdifferentiated Cells
ASC, adult stem cell; DTC, direct transdifferentiated cell; ESC, embryonic stem cell; iPSC, induced pluripotent stem cell.
Challenges of Cell Transdifferentiation
From the above table, we can see that some challenges exist when studying transdifferentiation, including the following:
How similar are transdifferentiated and target cells?
In the last decade, transdifferentiation has become a new method to obtain mature functional dopamine neurons from somatic cells. This approach could overcome the problems associated with the use of human pluripotent stem cells, such as ethical issues and possible safety concerns arising from the overproliferation of undifferentiated cells after transplantation (Aversano, 2022). A study showed that totipotent iPSCs still differed from human ESCs, implying that iPSCs could be at a totipotent state, but still retain molecular traces of their initial cells (Ohi et al., 2011). Three-dimensional (3D) culture may be one way to reduce the heterogeneity of transdifferentiated cells and thus become more similar to target cells.
In regenerative medicine, the ability to use these reprogrammed cells without the risk of causing cancer or other cellular alteration remains to be fully elucidated. 3D culture platforms are being increasingly used by researchers studying stem cells owing to their ability to achieve high cell output and increase culture volume. In addition, this culture method may reduce the heterogeneity of transdifferentiated cells and make them more phenotypically similar to the target cell. Thus, benchtop bioreactors and 3D suspension incubators hold potential for the successful production of high-quality iPSCs (Kwok et al., 2022).
Do transdifferentiated cells maintain all their properties when the inducible factor is removed?
Hanna et al. suggested that upon the removal of reprogramming factors (either by chemical or genetic inhibition), iPSCs would continue to express pluripotency markers, thereby maintaining their differentiation potential and genetic status (Hanna et al., 2009). This stability is likely to be controlled by inherent pluripotent factors through regulated positive feedback networks (Dumasia et al., 2021). Hence, it is very important to further explore the impact of removing reprogramming factors during transdifferentiation.
What is the transdifferentiated cell survival pattern both in vivo and in vitro, and how does it integrate with the physiological response?
In vivo transdifferentiation is a strategy to achieve cell fate transformation within the natural physiological niche. This technology can provide a time- and cost-effective alternative to applications in regenerative medicine, and can also minimize the problems associated with in vitro culture and cell transplantation. It has been developing rapidly because of increased identification of TFs that induce cell reprogramming (Fu et al., 2014). In vitro transdifferentiation to determine the fate of somatic cells can be divided into two operations, ‘Cis-reprogramming’ by pioneer TFs or epigenetic regulators and ‘Transreprogramming’ by specifier TFs (Egawa et al., 2020). The issue must be further explored. For example, can induced neurons transmit and receive signals at synapses after transplantation?
Therapeutic Application of and Perspectives for Transdifferentiated Cells
As the mechanism of cell transdifferentiation remains poorly understood, it is difficult to achieve this in vitro. High efficiency of cell transdifferentiation has become the main research direction in the field, which is not only a key step in transdifferentiation research but will also provide valuable data for the practical application of genetic engineering approaches. Autograft or allograft of bone marrow or peripheral blood can promote the functional migration and homing of hematopoietic, mesenchymal, and endothelial stem cells and may improve tissue remodeling for and the treatment of severe injuries by recruiting stem cells to the area (Zengin et al., 2006). More specifically, hematopoietic stem cell transplantation may be used for the treatment of patients with autoimmune diseases, refractory and aplastic anemia, congenital thrombocytopenia, osteoporosis, cardiovascular disease, Crohn's disease, diabetes, Hodgkin's and non-Hodgkin's lymphoma, and invasive tumors (Balassa et al., 2019).
Mesenchymal and endothelial stem cells can be used for regenerating several tissues, including bones, cartilage, tendons, muscles, fat, brain, lungs, heart, pancreas, kidneys, and eyes. These immature cells may be used to treat osteogenesis insufficiency, atherosclerotic lesions, ischemia, and muscle cell-related diseases (Kajstura et al., 2005). Mesenchymal stem cells have low immunogenicity (Yagi et al., 2016) and therefore have the therapeutic potential to prevent tissue and organ allograft rejection and severe acute graft versus host disease. They may also be used for treating of autoimmune diseases such as inflammatory bowel disease and autoimmune cardiomyopathy (Penack et al., 2020).
Cell transdifferentiation has opened avenues for the advancement of the clinical approach of regenerative medicine for incurable central nervous system diseases. Blood–brain barrier breakdown after the disruption of the neurovascular unit is a pathological feature common to both acute central nervous system injury and some chronic neurodegenerative diseases, creating an attractive delivery site of ectopic TFs for direct reprogramming in vivo (Egawa et al., 2020).
Adipose stem cells not only can differentiate into mesodermal functional cells (including cardiomyocytes, adipocytes, and cartilage, bone, muscle, and endothelial cells) but also into cells of the endoderm (such as liver and endocrine pancreatic cells) and ectoderm (such as neurons) (Prunet-Marcassus et al., 2006). Although adipose stem cell research started late, it is currently an important research topic owing to its advantages, such as ubiquitous source, easy access, rapid expansion, genetic stability, low proportion of cell aging and death, and continuous subculture of up to 130 generations (Vieira et al., 2010; Jeon et al., 2016).
Adipose stem cells can be easily acquired from surgically resected adipose tissues; therefore, they represent another promising, easily available immature cell type for the treatment of different diseases, including various types of bone, cartilage, and musculoskeletal disorders, cardiovascular and liver disorders, neurological diseases, and diabetes, as well as for adipose and skeletal muscle tissue reconstruction (Fraser et al., 2006).
To design new treatment strategies, it is necessary to consider the local microenvironment of extracellular signals that affect cell biology. Some questions require further investigation, such as the amplification of ASCs in vitro, daughter cell differentiation in different animal models for long-term treatment effects, and expansion of cells, while maintaining their function. Such additional knowledge will eventually lead to the development of new effective treatments for degenerative diseases, which still lack effective therapies. For example, targeting cancer stem cells, which are known to contribute to cancer development and growth, is currently considered a very promising new treatment approach (Benmelouka et al., 2021).
Conclusion
Given that the number of cell types that can be generated through direct reprogramming is rapidly increasing, the production of functional cells for therapeutic purposes has emerged as a promising strategy (Wang et al., 2021). For example, bone diseases are prevalent worldwide and not only cause public health problems but also account for a considerable portion of health care expenditures. In particular, hypertrophic chondrocytes can transdifferentiate into bone cells during endochondral bone formation, fracture repair, and some bone diseases. Moreover, tendon cells, beyond their conventional role in joint movement, directly participate in normal bone and cartilage formation and ectopic ossification (Wang et al., 2022). The underlying genetic mechanism of transdifferentiation remains unclear, because it is very challenging to fully understand its microenvironment in vivo. This limited knowledge hampers the technical development of directional differentiation and clinical application of transdifferentiation.
Although the in situ conversion of fibroblasts into cardiomyocytes or nerve cells after heart and brain injury is a scientific hypothesis, it has aroused substantial attention in the field of stem cell research and regenerative medicine. Currently, we know that cells can change phenotypes through differentiation, dedifferentiation, transdifferentiation, and reprogramming, with cells regulating their fate through change and unchanged homeostasis. For example, the transdifferentiation of glial cells into functional neurons is a potential therapeutic approach to reverse neuronal loss associated with neurodegenerative diseases and brain injury.
Conversion diagram of neural stem cells and other cells is shown in Figure 3 (Dallas and Bonewald, 2010; Tang and Lane, 2012; Edri et al., 2015; Saera-Vila et al., 2015; Komori, 2016; Vieira et al., 2018; Bielczyk-Maczynska, 2019; Tresguerres et al., 2020; Ababneh et al., 2022; Hoang et al., 2022; Millay, 2022). Downregulation of the expression of the RNA-binding protein Ptbp1 using in vivo viral delivery of a recently developed RNA-targeting CRISPR system CasRx was reported to result in the efficient conversion of Muller glia into retinal ganglion cells, leading to the remission of disease symptoms associated with retinal ganglion cell loss (Zhou et al., 2020).

Conversion diagram of NSCs and other cell types. Full, dotted, dashed/dotted, and dashed arrows represent transdifferentiation, dedifferentiation, differentiation, and programming, respectively. AE, adipocyte; CE, chondrocyte; CT, chondroblast; iPSC, induced pluripotent stem cell; ME, myocyte; MSC, mesenchymal stem cell; MT, myoblast; NCSC, neural crest stem cell; NSC, neural stem cell; OE, osteocyte; OT, osteoblast; PE, preadipocyte; VENTC, ventral transference nerve cell.
Although what determines the fate of stem or differentiated progeny cells is not entirely clear, it is known that these cells form a connected network. Nevertheless, the successful use of cell transdifferentiation for the treatment of major diseases still requires considerable research. We cannot ensure that transdifferentiation research will help us treat all diseases, but it will certainly help many people affected by disorders that remain untreatable to date (Yao et al., 2020).
Footnotes
Authors' Contributions
F.W.: Conceptualization, methodology, and writing–original draft. R.L.: Writing–review and editing. L.Z.: Resources and data curation. X.N.: Resources and writing–review and editing. W.L.: Resources and writing–review and editing. L.C.: Writing–review and editing and funding acquisition.
Author Disclosure Statement
The authors declare they have no conflicting financial interests.
Funding Information
This research was supported by the Program for Science and Technology Innovation Talents in Universities of Henan Province (No. 22HASTIT041), National Natural Science Foundation of China (No. 32071447), and National Program for Key Scientific and Technological Projects in Henan Province (No. 222102110421).
