Bioelectric Potential in Next-Generation Organoids: Electrical Stimulation to Enhance 3D Structures of the Central Nervous System

Pluripotent stem cell-derived organoid models of the central nervous system represent one of the most exciting areas in in vitro tissue engineering. Classically, organoids of the brain, retina and spinal cord have been generated via recapitulation of in vivo developmental cues, including biochemical and biomechanical. However, a lesser studied cue, bioelectricity, has been shown to regulate central nervous system development and function. In particular, electrical stimulation of neural cells has generated some important phenotypes relating to development and differentiation. Emerging techniques in bioengineering and biomaterials utilise electrical stimulation using conductive polymers. However, state-of-the-art pluripotent stem cell technology has not yet merged with this exciting area of bioelectricity. Here, we discuss recent findings in the field of bioelectricity relating to the central nervous system, possible mechanisms, and how electrical stimulation may be utilised as a novel technique to engineer “next-generation” organoids.

The differentiation of pluripotent stem cells (PSCs) into multicellular tissue has been studied since the establishment of the first mouse embryonic stem cell lines in the 1980s (Evans and Kaufman, 1981; Martin, 1981). Stem cell technology has greatly benefited the study of the human central nervous system (CNS). Comprised of the brain, spinal cord, retina and olfactory nerve, the CNS is the most complex and critical system of the human body. Thus, as human pluripotent stem cells (hPSCs) emerged, including embryonic stem cells (ECS) and induced pluripotent stem cells (iPSCs), the CNS was a key target for differentiation.

Traditionally, the generation of different cell types in vitro aimed to recapitulate in vivo development by utilising our knowledge of developmental biology. As such, key biochemical cues, such as growth factors and signalling molecules that regulate gene expression, were employed to instruct PSCs to differentiate into various cell types. Two-dimensional (2D) monolayer cultures of neuronal and other CNS-tissues were quickly generated after the discovery of human ESCs (Thomson et al., 1998; Reubinoff et al., 2000; Reubinoff et al., 2001). However, 2D cultures lack the niche and cell-cell interactions that are crucial for CNS function. To address this, studies turned to an additional modality of differentiation: biomechanical (Eiraku et al., 2008; Eiraku et al., 2011; Eiraku et al., 2012). Biomechanical cues, such as extracellular matrix (ECM) scaffolds, alongside biochemical cues, powered the generation of three-dimensional (3D) mini-organs, or organoids, from PSCs.

Various protocols to derive organoid models of the CNS, particularly the brain and retina, have since been developed (Reviewed in Pacitti et al., 2019; Chiaradia and Lancaster, 2020; O’Hara-Wright and Gonzalez-Cordero, 2020). CNS organoids, including those derived from healthy and diseased patient-derived iPSCs, have been used to study disease and development in human tissue at an unprecedented level (Pacitti et al., 2019; Chhibber et al., 2020). As progress in our understanding of neurodevelopment is made, and the field of bioengineering evolves, differentiation protocols have been suitably optimised to generate more robust and advanced organoids. However, CNS organoids still lack many critical components and characteristics of their in vivo counterparts, such as long-term survival, maturing structures, and efficient oxygen transport. Moreover, differentiation protocols can be highly variable and produce heterogenous tissues (Lancaster et al., 2013). Therefore, as we look towards next-generation organoids, culture conditions must be optimised to address these shortcomings. Currently, headway is being made through fine-tuning of bioengineering approaches, such as 3D biomaterial compositions to support structural development, assembloid technology to generate wider CNS structures, or organ-on-a-chip technology to promote oxygenation (Achberger et al., 2019; Bierman-Duquette et al., 2021; Ao et al., 2022; Miura et al., 2022). However, to best replicate organogenesis in vitro, understanding native mechanisms in basic development is crucial.

An additional modality guiding development in vivo, alongside biochemical and biomechanical, is bioelectric communication (Levin, 2014; Newman, 2019). Yet, as a physiological cue, the inclusion of electrical fields in PSC-differentiation systems has been largely ignored. The role of bioelectricity in vivo was first elucidated in the 1800s, with studies on regeneration (Galvani, 1791; Bois-Reymond, 1884; Becker, 1961; Becker and Spadaro, 1972; Borgens et al., 1977; 1979a). Now, emerging evidence implicates bioelectricity in other key biological processes, including development, and differentiation. Low voltage endogenous electric fields (EFs) can be identified in the embryonic CNS during development and the importance of bioelectricity in guiding neural development has been demonstrated in animal models, primary neural cultures, and stem-cell derived cultures of the CNS (Metcalf and Borgens, 1994; Metcalf et al., 1994; Shi and Borgens, 1995). Control of ion flux, through manipulation of ion channels, or application of exogenous electrical stimulation, can be used to manipulate EFs both in vivo and in vitro. Meanwhile, in the field of regenerative medicine, exogenous application of low voltage electrical stimulation has been used to promote CNS regeneration, or prevent degeneration, in the clinic and experimental studies (Fu et al., 2015; Sehic et al., 2016; Marquez-Chin and Popovic, 2020; Hellenbrand et al., 2021; Krauss et al., 2021; Sanie-Jahromi et al., 2021; Sui et al., 2021; Karamian et al., 2022). Bioengineering has interfaced with bioelectricity to deliver exogenous electrical stimulation through conductive biomaterial polymers in basic 3D models of the CNS (Bierman-Duquette et al., 2021). These studies suggest promise in the use of EFs to improve organoid research and differentiation approaches.

This could have major implications in the utility of organoids for applications requiring precisely controlled differentiation or high-throughput, such as drug screening. Yet, the use of advanced conductive polymers extends beyond the realm of basic cell biology and detracts from the intrinsic self-organisation capacity of PSC-derived organoids. Many of the biological processes EFs are implicated to control in vivo, such as proliferation and survival, are similar to those that need to be controlled in vitro to improve CNS organoid differentiation. Therefore, enormous potential lies with the utilisation of exogenous electrical stimulation as a powerful, non-chemical and non-genetic tool to improve organoid cultures.

In this Review, we will firstly discuss the current state of PSC-derived CNS organoids, focusing particularly on the brain, and the approaches taken to guide differentiation of CNS tissue in vitro. We will then discuss the role of bioelectricity and endogenous EFs in vivo, and the lessons we have learnt about the role of bioelectricity in the CNS. Next, we will briefly review the importance of bioelectricity and endogenous EFs in key biological processes, before highlighting the key findings from studies which have manipulated EFs in vivo and in vitro. We will discuss the pertinence of these bioelectric-driven phenomena to improving 3D organoid cultures and describe the key few studies implementing bioelectricity and bioengineering in 3D CNS tissues. Finally, we will outline future studies that must be conducted to best utilise bioelectricity for organoid engineering. These include the proper evaluation of stimulation in various differentiation protocols and omics characterisation to understand mechanisms.

The vertebrate CNS encompasses the brain, spinal cord, neural retina and olfactory nerve (Northcutt, 1984). Distinct from the peripheral nervous system, the CNS acts as a processing centre for most necessary bodily functions. Conditions affecting the CNS present an increasing global disease burden. In recent years, neurological disorders have been the foremost cause of disability-adjusted life years, and the second leading cause of death (Feigin et al., 2019). The study of the human CNS has been greatly limited by the inability to study tissue anatomy in vivo and tissue scarcity ex vivo. Whilst model organisms provide invaluable insights into CNS biological processes and mechanisms of disease, the complexity of neural networks and retinal architecture in humans remains unrivalled in comparison to lower order animals (Gibson, 1938; Uga and Smelser, 1973; Zhao and Bhattacharyya, 2018; Desmoulin-Canselier and Moutaud, 2019). The advent of PSC technology has allowed generation of human CNS derivatives in the dish (Figure 1A).

Soon after the establishment of the first human ESC line in 1998, differentiation of neuronal cultures was achieved (Thomson et al., 1998; Reubinoff et al., 2000; Reubinoff et al., 2001). Despite the complexity of neurogenesis, PSC-derived monolayer cultures were shown to readily and faithfully differentiate into a neuroectoderm lineage with accordingly dynamic and stage-specific gene expression profiles (Yeo et al., 2013; Wu et al., 2010; Fathi et al., 2011). Moreover, the default differentiation path of PSCs was identified to be neural (Eiraku et al., 2008). Transcriptomic analysis confirmed hESC-derived neuronal cultures resembled in vivo human neurons, particularly of the foetal brain (Pai et al., 2012). Thus, even in conditions of minimal growth factors, the biochemical cues required to drive neuronal differentiation are somewhat intrinsically provided by endogenous signalling of PSCs. However, to produce alternative CNS-derivative cells from PSCs, suitable biochemical cues were required. For example, culturing in the presence of epidermal growth factor (EGF) and fibroblast growth factor (FGF)-2, known to promote proliferation of glial precursor cells, and timed addition of tri-iodothyronine (T3), which induces oligodendrocytes, was used to derive glial cell lineages (Murray and Dubois-Dalcq, 1997; Brüstle et al., 1999; Reubinoff et al., 2001).

Similarly, by mimicking the biochemical controls governing in vivo embryonic head development, antagonism of Wnt and BMP with DKK1 and Noggin promoted an anterior neural fate of PSCs (Glinka et al., 1997; Mukhopadhyay et al., 2001; Banin et al., 2006; Lamba et al., 2006). Later addition of IGF-1, known to induce ectopic eye formation in animal models, differentiated cells further to a retinal lineage (Pera et al., 2001; Richard-Parpaillon et al., 2002; Lamba et al., 2006). However, 2D monolayer cultures could never replicate the complexity of the in vivo CNS. The cytoarchitecture and interactions of networks of cells within their 3D environment is critical to the function of the brain, spinal cord, retina and olfactory structures, both as individual structures and an integrative system. Therefore, in attempting to better model CNS tissue in the dish, the development of 3D hPSC-derived organoids was pivotal. The pioneering work of Yoshiki Sasai demonstrated that PSCs grown in suspension, rather than 2D monolayers, form embryoid body aggregates which acquire a neural fate tissue-autonomously (Eiraku et al., 2008). The group cultured mouse PSCs in suspension without serum, forming embryoid bodies, before allowing them to re-aggregate in a 96-well plate to form 3D tissue of neuroectodermal origin. This neuroectodermal-like epithelium differentiated into multi-layered cortical tissue, containing progenitor cells and neurons (Eiraku et al., 2008). Serum-free culture of embryoid body-like aggregates with quick-reaggregation, or SFEBq, thereby formed the basis of early 3D neural differentiation protocols, and the predecessor to the CNS organoid. Notably, it was the consideration of the physical microenvironment, by transitioning the cultures to suspension, that mediated the formation of this 3D tissue.

Soon afterwards, modification of the SFEBq method with the addition of Matrigel to provide ECM, lead to the generation of neuroepithelial cysts which self-organised into optic cups resembling the in vivo retina (Eiraku et al., 2011). These may be considered the first true CNS organoid: 3D retinal structures with stratified neuroepithelia containing all major retinal cell types. The authors considered the complex forces controlling embryonic retina development and the sequential invagination and evagination folding processes (Eiraku et al., 2011; Eiraku et al., 2012). Matrigel, primarily containing laminin, aided the creation of a basement membrane structure for the aggregate to invaginate around. The inclusion of a biomechanical stimuli thereby facilitated organotypic-like development in vitro (Figure 1B) (Eiraku et al., 2011; Eiraku et al., 2012). Sasai’s group subsequently demonstrated differentiation of human PSC cultures to retinal and various neural 3D structures. With temporal addition of growth factors, agonists or inhibitors of FGF, Wnt, and BMP signalling networks, differentiations of forebrain, neocortex and hippocampus lineages were achieved (Kadoshima et al., 2013; Muguruma et al., 2015; Sakaguchi et al., 2015). Studies have therefore demonstrated competence of PSC-derived organoid cultures to respond to both biochemical and biomechanical stimuli, and through diversification of these cues, organotypic cultures of different CNS structures are derived, including the brain, spinal cord and retina (Figure 1B).

Brain organoid protocols can be broadly classified into two approaches- guided and unguided (Figure 1C). In guided approaches, building on the work of Sasai, exogenous patterning molecules are added to guide differentiation to the desired neural lineage (Eiraku et al., 2008; Eiraku et al., 2011). This has permitted brain-region specific organoid generation, including forebrain, midbrain, and hypothalamus (Qian et al., 2016; Xiang et al., 2017; Qian et al., 2018; Sloan et al., 2018; Cederquist et al., 2019; Kim et al., 2019). For example, manipulation of Sonic Hedgehog signalling patterns forebrain organoids into dorsal and ventral subdomains, better mimicking in vivo topology (Cederquist et al., 2019). Treatment with brain-derived neurotrophic factor (BDNF), glial cell line–derived neurotrophic factor (GDNF) and ascorbic acid guides differentiation to the midbrain and brainstem lineage (Eura et al., 2020). Sequential BMP and SMAD inhibition and activation promotes continued differentiation and maintenance of dopaminergic neurons within midbrain organoids (Jovanovic et al., 2018). Alternatively, at the neuroectodermal stage, addition of WNT3A and Hedgehog pathway agonist, Purmorphamine, induces the hypothalamic lineage (Qian et al., 2016).

Unguided approaches, however, depend on the spontaneous self-organisation capacity of PSC-derived aggregates. Lancaster and Knoblich generated the cerebral, or “whole-brain” organoid, self-patterned PSC-derived aggregates embedded in Matrigel (Figure 1C) (Lancaster and Knoblich, 2014). These are heterogenous structures, comprised of multiple neural cell types within discrete regions. The addition of Matrigel for ECM provides a scaffold for neuroepithelial budding, and these buds may expand and migrate and assume polarity to differentiate into various brain regions, including forebrain, midbrain, hindbrain, choroid plexus, and retina. Single cell RNAseq confirmed the identity of neural progenitor cells, astrocytes, oligodendrocyte-like precursors and excitatory neurons, along with photosensitive retinal cells (Quadrato et al., 2017). After 4 weeks in vitro, cerebral organoids show regional identities, such as FOXG1 demarcating the forebrain, whilst expression of TTR delineates the choroid plexus and FZD9 the hippocampal-like region (Lancaster and Knoblich, 2014; Quadrato et al., 2017).

In vertebrates, the retina and olfactory nerve are considered an extension of the CNS. The retina is a multi-layered tissue comprising of seven cell subtypes, each highly specialised to their respective role in the light-induced phototransduction cascade. After the initial demonstration of the first CNS organoid, using the SFEBq method, retinal organoid protocols have evolved and diversified (Eiraku et al., 2011). Similarly to the brain, organoid differentiation approaches can be guided, with temporal addition of small molecules and factors informed from studies of in vivo retinal development (Meyer et al., 2011; Kuwahara et al., 2015; Mellough et al., 2015; Singh et al., 2015). Other groups rely more on self-patterning of PSCs, employing minimal additional factors at early stages (Reichman et al., 2014; Zhong et al., 2014; Gonzalez-Cordero et al., 2017; Reichman et al., 2017).

Resulting organoids mimic in vivo developmental timeline and lamellar organisation and have informed knowledge of human retinal development (Reviewed in O’Hara-Wright and Gonzalez-Cordero, 2020). Retinal organoids are also amiable to longer-term maintenance than brain organoids, and electrophysiology and calcium-mediated functionality has been described (Zhong et al., 2014; Mellough et al., 2015; Reichman et al., 2017; Gagliardi et al., 2018; Cowan et al., 2020; Saha et al., 2022). Technology towards modelling the olfactory system in vitro, on the other hand, remains underdeveloped. Recently, three-dimensional culture of primary olfactory progenitor cells has been achieved with Matrigel encapsulation and guided differentiation (Ren et al., 2021). However, the generation of true olfactory organoids has not yet been demonstrated.

The other major component of the CNS, the spinal cord, is responsible for neurotransmission between the interconnecting motor cortex and medulla oblongata of the brain (Bican et al., 2013). The spinal cord co-ordinates highly organised neural circuits, and morphogenesis is accordingly complex. Study of the spinal cord in vivo in humans is greatly hampered by inaccessibility, making it an ideal target for in vitro recapitulation. Spinal cord organoids have been derived via adaptation of guided brain organoid protocols, implementing Matrigel encapsulation and spinning flasks (Hor et al., 2018). Patterning of spinal cord organoids mimics in vivo axes and morphogen gradients, which is not observed within 2D cultures (Hor et al., 2018; Winanto et al., 2019). Optimised guided protocols implementing small molecules, particularly fine-tuning of BMP4 to mimic in vivo development, also guides spinal cord organoid morphogenesis in vitro (Duval et al., 2019). Resultant spinal cord organoids generate a range of cell sub-types.

Albeit the remarkable process in the field of CNS organoids and some demonstrations of functionality, the generation of a multi-organ broader CNS system in the dish has not been possible and maturation is still limited. Therefore, we are still a very long way from truly recapitulating CNS development in vitro. Engineering the next-generation of organoids may require turning to an additional modality to direct differentiation. It is widely established that bioelectricity, along with biochemical and biomechanical signals, plays an important developmental role (Levin, 2012; Levin, 2014; Levin, 2021). Yet studies of PSC-derived CNS organoids have not considered the role of bioelectric cues in CNS morphogenesis and attempted to recapitulate these in the dish. Recently, there has been a renewed interest in the phenomena of bioelectric-driven processes, particularly relating to nervous system development and neural differentiation (Bertucci et al., 2019; Lancaster, 2019; Bierman-Duquette et al., 2021). It may now be pertinent to consider the body of research relating to intrinsic electric fields in development, and previous studies applying electric fields to stem-cell derived cultures. By reviewing the lessons learnt from these studies, alongside the specific shortfalls in current organoid models, bioelectricity could be exploited to engineer next-generation CNS organoids.

In vivo, cells exhibit endogenous electrical fields (EFs). Ions, and other charged biomolecules are segregated between the extra- and intracellular spaces by impermeable membranes. Ions can flow within the cell (trans-cellularly) in an apical-basal direction and be selectively transported through ion channels embedded in the membrane (Figure 2A). The asymmetric distribution of ion channels at the apical and basal membranes creates a specific pattern of ions and biomolecules and distinct domains of charges. This creates a difference in the electrical potential between the cell membrane and extracellular space, termed a membrane potential (Vmem) (Figure 2B).

In contrast to rapidly firing action potentials in excitable cells, endogenous EFs are steady or slow changing gradients present in all cell types. As low magnitude potentials, these are typically close to the resting neuron potential of ∼ −70 mV in differentiated cells, and lower yet in embryonic cells (McCaig et al., 2009). This Vmem of every cell in an epithelium amounts to a transepithelial potential difference (TEPD). Ionic current flows across the cell, transcellularly, to the extracellular space, and returns between the cells, paracellularly. The presence of tight junctions at the apical side of the intercellular space provides a barrier to ion flux. Thus, we can think of an epithelium like a battery (Figure 2B). A TEPD results from ionic segregation which is sustained by ionic pumps and channels, and these work against the resistance of tight junctions, and lesser so adherens junctions. The paracellular junctions therefore typically maintain a positive current basolaterally, and the level of resistance encountered at the junctions determine the TEPD, within a typical range of 15–60 mV (McCaig et al., 2009; Montalbetti and Fischbarg, 2009; Dubé et al., 2010). This intrinsic bioelectricity was early identified for involvement in development, regeneration and wound healing. Functional investigation established the existence of small endogenous EFs at the regeneration bud, or blastema- the mass of undifferentiated cells at amputated amphibian limbs (Becker, 1961). Limb regeneration was determined to be dependent on electrochemical activity, and enhancement or introduction of EFs promotes regeneration, even in typically non-regenerative amphibian and mammalian tissue (Becker, 1961; Becker and Spadaro, 1972; Borgens et al., 1977; Borgens et al., 1979a; Borgens et al., 1979b).

Remarkably similar and steady endogenous EFs to those present in the amphibian blastema were also identified in the blastema of the amputated digit of human children (Illingworth and Barker, 1980). Known as the injury potential, or demarcation current, a difference in electrical potential is created between intact epithelium and the site of a wound (Jia et al., 2021). The TEPD at the site of injury falls and current “leaks” out from the wound edge, being the pathway of least resistance, creating an electrical field (Figure 2C). Whilst regeneration can be defined as the recreation of tissues in response to damage or degeneration, development refers to the morphogenetic transformation of the embryo to differentiated tissue. However, biologically speaking, development and regeneration require the same processes: the division, migration, and differentiation of pluripotent cells into multicellular tissues and organs. It therefore follows that EFs have a pivotal role in embryogenesis. In vertebrates, gradients of electric field activity are identifiable in the very early two-cell stage embryo (Levin et al., 2002). The first description of endogenous electrical currents in embryology was given in day 2–4 chicken embryos (Hotary and Robinson, 1990). Using a vibrational probe, an EF of 20 mV/mm was measured at the site of the presumptive hindgut (Hotary and Robinson, 1990). This stage of development co-occurs with the degeneration of gut intestinal epithelium, thus, as in the case of a wounded epithelium, providing a pathway of less resistance for leakage of current. It was soon thereafter established that endogenous EFs play a crucial role in the morphogenesis of the CNS.

In the axolotl embryo, an endogenous EF of 10 mV/mm is generated from under the neural plate, polarising the embryo along the rostral/caudal axis during neurulation (Metcalf et al., 1994; Shi and Borgens, 1995). As the architecture of the neural tube takes forms, the neural folds provide a pathway of lesser resistance for current flow. Remarkably, in the Xenopus, artificial manipulation of transmembrane potential is sufficient to induce ectopic eye formation during early embryogenesis (Pai et al., 2012; Blackiston and Levin, 2013). Modification of endogenous EFs during neurulation by as little as 5–25 mV/mm, through application of artificial voltage, results in developmental abnormalities (Metcalf and Borgens, 1994). Of note, this effect is stage-dependent, and exposure to artificial EFs at the earlier gastrula stage does not impact embryo development. This suggests that timed EF exposure, similarly to spatiotemporal distribution of biochemical and biomechanical stimuli, has a powerful ability to guide morphogenesis.

Accordingly, differentiation of PSCs could benefit from the consideration of EFs. Whilst endogenous EFs exist in vivo, it is also possible to apply exogenous EFs, via electrical stimulation, or alter endogenous EFs via manipulation of ion channels. The parameters of electrical stimulation, namely waveform, intensity, frequency and duration can be varied according to purpose. Numerous studies demonstrate EF modulation to target the CNS both in vivo in animal models, clinical studies, or in vitro in cell cultures (Sehic et al., 2016; Bertucci et al., 2019; Lancaster, 2019; Mobini et al., 2019; Krauss et al., 2021; Sanie-Jahromi et al., 2021). By increasing or decreasing EFs, studies have demonstrated the importance of endogenous EFs in normal development (Levin, 2012; Smith et al., 2018; Iwasa et al., 2019; Sefton et al., 2020; Levin, 2021). Additionally, the utility of EF modulation to produce desired effects in vitro has been revealed, such as increased differentiation, survival or maturation of CNS cells (Figure 3) (Bertucci et al., 2019; Mobini et al., 2019; Bierman-Duquette et al., 2021; Hlavac et al., 2021). Whilst exact mechanisms of action are not understood, the beneficial effects of EF modulation have been accepted in vivo for multiple tissues and organs, including bone and cardiac (Leppik et al., 2018; Watson et al., 2019). Electrical stimulation-based treatments are now being implemented in the clinic, particularly for the brain (Sparing and Mottaghy, 2008; Fedorov et al., 2010; Kubis, 2016; Desmoulin-Canselier and Moutaud, 2019; Krauss et al., 2021; Sui et al., 2021).

Electrical stimulation of the brain is beneficial for neurological disease, psychiatric disease and brain tumours. Non-invasive methods, including transcranial direct current stimulation (tDCS) and transcranial magnetic stimulation (TMS) have been shown to facilitate motor recovery in stroke patients (Webster et al., 2006; Fedorov et al., 2010; Kubis, 2016). Non-invasive stimulation of glioblastoma patient brains, for 20 min at a desired field strength of 0.25 V/m, altered cerebral blood flow and aided necrosis of the tumours (Sprugnoli et al., 2019). Alternatively, epidural cortical stimulation (ECS) and deep-brain stimulation, involve direct implantation of electrodes into the brain, to provide more focal electrical stimulation (Nahas et al., 2010; Balossier et al., 2015; Sui et al., 2021) (Figure 4C). This exerts desirable effects on neuroplasticity, improving neural network connectivity and topological reorganisation in cases of neurodegeneration (van Hartevelt et al., 2014). Typically, patients receiving deep-brain stimulation, such as those suffering from neurodegenerative diseases, are adults with fully formed neural structures. However, the adult brain contains a small population of neural stem cells (NSCs) arising from the subventricular zone. NSCs are self-renewing and multipotent, giving rise to neural progenitor cells (NPCs). NPCs, which have a finite replication potential, form most, if not all, of the neural and glial cells of the CNS (Morest and Silver, 2003; Merkle et al., 2004). NSCs or NPCs can be isolated from primary brain and cultured in vitro, both in 2D, or 3D aggregates deemed neurospheres (Figure 4A). In contrast to organoids, neurospheres lack cytoarchitecture or complex organotypic-like organisation.

Nonetheless, when engineering neural organoids, we aim to replicate neurogenesis in vitro via guiding PSCs into NPCs, before forming more mature neural and glial subtypes. It is therefore interesting to consider the effect of EF modulation on neural stem cell, progenitor cell populations or neurospheres.

Effects of brain stimulation from the clinic have been translated to NPCs and NSCs in vitro. Many studies demonstrate neural cells are electrosensitive, responding with behavioural or functional changes to EFs. Applied exogenous electrical stimulation promotes survival of NSCs both in vivo and in vitro (Du et al., 2018; Sefton et al., 2020). Specifically, Sefton et al. (2020) demonstrate a significant increase in the number NPCs in neurospheres stimulated at 250 mV/mm for 3 h, formed from mouse brain-derived NPCs in vitro. NPCs in stimulated neurospheres were identified to have an increased capacity for survival, with an increased number of cells entering S-phase of the cell cycle. Interestingly, addition of stimulated conditioned media to non-stimulated neurospheres was not sufficient to induce this increased neurogenesis phenotype. Thus, niche-released factors were not responsible for the increase in neurogenesis, but rather intrinsic cell cycle modulation (Figure 3) (Sefton et al., 2020). In foetal NPCs derived from the mouse embryo, brief (50 μs or 200 μs) low-voltage biphasic electrical stimulation pulses significantly promoted cell proliferation (Chang et al., 2011). Similar phenotypes have been identified in vivo, where stimulation of the adult mammalian cortical brain via electrode implantation, leads to an increase in the proliferative pool of NPCs (Stone et al., 2011; Sefton et al., 2020). Targeted stimulation of the Entorhinal Cortex (EC), a region of the mouse hippocampus, resulted in increased proliferative cells in a projecting region, the denate gyrus. Remarkably, once these cells matured and integrated into hippocampal neural circuits, an increased spatial memory response, measured by water maze, was observed (Stone et al., 2011).

In humans, the brains of neurodegenerative patients who received deep-brain stimulation have been examined by immunohistochemistry (Vedam-Mai et al., 2014). Compared to brains of unstimulated patients, the levels of proliferative cells in the subventricular zone, the birthplace of NPCs, were two-six fold greater (Vedam-Mai et al., 2014). From the subventricular zone in vivo, NPCs must also migrate long-distance, particularly to give rise to the olfactory bulb. Migration occurs along a route deemed the rostral migratory stream (Lois and Alvarez-Buylla, 1994; Curtis et al., 2007). Endogenous electrical fields, with a field strength of ∼3 mV/mm, help to guide this migratory behaviour (Cao et al., 2013). Further studies have identified endogenous EFs surrounding the cerebral cortex capable of inducing migration of transplanted NPCs (Iwasa et al., 2019). In vitro, application of physiologically relevant EFs to rat embryonic brain-derived NPCs facilitates migration towards the cathode (Li et al., 2008). Thus, it is postulated that NPCs respond to electrical stimulation, both in vivo and in vitro with directed migratory behaviour (Babona-Pilipos et al., 2011; Babona-Pilipos et al., 2015; Feng et al., 2017; Iwasa et al., 2019; Sefton et al., 2020). Application of exogenous EFs can alter endogenous EF-guided migrations but may not be sufficient to reverse endogenous EF effects entirely (Iwasa et al., 2019).

Collectively, studies described in this review indicate the viability of cells in vivo and in vitro in mild electrical fields conditions, typically within a range of 25-250mV/mm (Wang et al., 2013; Sefton et al., 2020). However, as with all cell-cycle related effects, manipulation of electrical fields may either promote, inhibit, or exert no effect on cell death mechanisms. This likely depends on stimulation parameters, and tissue of interest. Electrical stimulation of cells in vitro suggests a proportional relationship between cell death and the duration and number of electrical stimulation pulses (Matsuki et al., 2010). Furthermore, electrical stimulation at higher voltages induces formation of pores in cell membranes (electroporation) which may result in induction of apoptosis or necrosis with prolonged pulse lengths beyond the nano-second scale (Oshima et al., 1997, Oshima et al., 1998; Hofmann et al., 1999; Al-Sakere et al., 2007). Researchers have appropriately manipulated the parameters of electrical stimulation in vivo and in vitro to target aberrantly dividing malignant cells, and promote tumour ablation (Wang et al., 2012; Chen et al., 2014). However, the application of low voltage biomimetic electrical fields, below those capable of breaking the membrane, may also induce apoptosis of carcinogenic cells, specifically through increased Ca2+ transport (Matsuki et al., 2010). Mild (15 V, 7.5 V/mm) EF-induced Ca2+ flow, promotes apoptotic signalling through the activation of caspase-8 and caspase-9, upstream of caspase-3 (Matsuki et al., 2010). Promisingly, most reports of cell death-related phenotypes in response to electrical stimulation of the CNS describe a reduction in apoptosis.

In vivo application of electrical stimulation to the cerebral cortex exerted anti-apoptotic effects in an ischemic stroke rat model, whereby ischemia-induced cell death is a major aetiology (Baba et al., 2009). The phosphoinositide 3-kinase (PI3K) pathway, a major player in cell cycle regulation, survival and apoptosis, is activated in the electrically stimulated cerebral cortex, leading to a reduction in apoptosis and ischemia-induced pathology. These findings were similarly confirmed in vitro using primary rat-derived NPCs from the olfactory bulb (Wang et al., 2013). Specifically, the authors examined the effects of biphasic pulsed mild EFs on growth factor-deprivation activated apoptosis. Electrical stimulation at 25 mV/mm and 50 mV/mm decreased rates of apoptosis and necrosis in NPCs cultured in serum-free conditions. Specifically, these neuroprotective effects were inhibited via blocking of PI3K/Akt (Wang et al., 2013). Electrical stimulation upregulated phosphorylation of Akt, and despite the absence of growth factors in culture conditions, BDNF was upregulated after continued (12 h) stimulation (Wang et al., 2013). Since PI3K signalling is a major cell cycle regulator, EF-induced mechanisms may control multiple fundamental processes, including survival, proliferation and differentiation, along with apoptosis. Similarly, growth factor BDNF contributes to neural cell survival, growth, differentiation and maturation. Therefore, multiple studies demonstrated that EF modulation could mediate distinct control of cell cycle and survival. These changes may also translate to changes in neurogenic differentiation capacity and the profile of stem cell-derived neural populations (Figure 4).

Electrical stimulation is reported in multiple studies to increase differentiation of NSCs and progenitor cells into neurons and oligodendrocytes both in vivo and in vitro (Ariza et al., 2010; Stewart et al., 2015; Chang et al., 2016; Du et al., 2018; Dong et al., 2019; Sefton et al., 2020). EFs of ∼300–400 mV/mm are reported to increase neurogenesis of Beta tubulin III + neurons, whilst field strengths of lower magnitudes may promote NPC proliferation and expansion in favour of differentiation (Ariza et al., 2010; Dong et al., 2019; Sefton et al., 2020; Hlavac et al., 2021). In vivo, hyperpolarisation of cerebral progenitor cells (radial glial) via over-expression of potassium channels, induced pre-emptive differentiation of typically later-born, mature neural subtypes (Vitali et al., 2018). Lineage tracing determined EF modulation via ion channel manipulation specifically regulated the division mode of NSCs. Notably, effects on neuronal subtype fate determination could be reversed with de-polarisation (Vitali et al., 2018). Thus, in vivo, modified EFs can function as fate determinants for NPCs. The increased maturation of NPCs in response to increased TEPD is a cue that could be mimicked in vitro to promote neurogenesis and maturation. Studies examining EF modulation of human NSC lines also identified altered fate bias. Exogenous electrical stimulation of human NSCs increased induction of neuron differentiation, as defined by a larger population of cells expressing neuron marker MAP2 (Stewart et al., 2015; Tomaskovic-Crook et al., 2019). Within these stimulated MAP2+ neurons, an increase in neurite outgrowth was also observed (Stewart et al., 2015; Tomaskovic-Crook et al., 2019).

Increased EF-induced neurite branching and length is similarly found in studies applying exogenous electrical stimulation to foetal mouse-derived and immortalised human brain derived-NPCs (Chang et al., 2011; Pires et al., 2015). Sufficient neurite branching is important for formation of functional neuronal networks, suggesting stimulation could also enhance the structure and functionality of in vitro CNS models. Taken together, these findings indicate a potential for EF modulation to promote cell cycle-related effects in neural cells (proliferation, survival and differentiation) to aid recapitulation of neurogenesis from PSCs.