An official website of the United States government
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Noebels JL, Avoli M, Rogawski MA, et al., editors. Jasper's Basic Mechanisms of the Epilepsies. 5th edition. New York: Oxford University Press; 2024. doi: 10.1093/med/9780197549469.003.0078
Cell transplantation could offer a strategy to overcome pharmacoresistance in epilepsy if transplanted cells can survive long term within the host brain, migrate into appropriate brain regions, differentiate into interneurons (INs) that increase inhibitory tone within the epileptic network, and integrate into the circuitry by forming synapses with host neurons. Several studies have demonstrated all of the above-mentioned prerequisites of a successful IN transplantation using rodent INs derived from the medial ganglionic eminence (MGE). They showed that INs are effective in decreasing seizure frequency consistently and in the long term across a variety of animal models. Behavioral comorbidities of epilepsy could be alleviated, and even hippocampal neuropathology of the disease was reduced. While use of rodent INs is feasible for preclinical research, a human cell product is needed for clinical translation. Protocols for IN in vitro differentiation from human pluripotent stem cells (PSCs) have been developed. First, data from transplantation studies in models of epilepsy confirm the promising disease-modifying activity that was observed with rodent MGE-derived INs: human IN transplantation was effective in reducing seizure burden, while no obvious adverse effects were associated with it. Further research on characterization of cells prior to transplant, process improvements, and good manufacturing practice production will be needed to ensure safety of an IN cell product; however, the recent advances in our ability to generate MGE-like cells from PSCs, in-depth characterization of cells with modern techniques and the combinatorial use of electrophysiology, optogenetics, and translational immunodeficient animal models to evaluate disease-modifying activity have brought human IN transplantation significantly closer to clinical application.
The vast majority of approved antiseizure medications have a solitary aim: to increase the inhibitory tone in the hyperexcitable epileptic brain. Despite the introduction of novel targets and mechanisms, the proportion of patients who do not respond to these treatments has remained stagnant at around 30% (Löscher and Schmidt, 2011). Systemic delivery of antiseizure drugs leads to exposure of multiple organ systems and indiscriminate access to the whole brain, which in turn contributes to unwanted side effects and loss of efficacy over time. Temporal lobe resection surgery offers a better outcome regarding seizure freedom, but it is not suitable for every patient and carries the risks of invasive brain surgery. Progress has been made in other targeted approaches such as cell transplantation as a restorative approach that could offer more than just symptomatically balancing inhibition and excitation in the epileptic brain. Such cell-based regenerative medicine combines the advantages of targeted modulation of hyperexcitability in distinct brain regions without the need for resection of brain tissue.
Various cell types have been evaluated for transplantation; however, the most promising body of experimental data is available on GABAergic neurons and consequently will be the focus of this review and is summarized in Table 78–1. In 2005, the term “interneuronopathy” was introduced to characterize intractable epilepsy that is caused by migration disorders of interneurons (INs) (Kato and Dobyns, 2005). Experimental animals, in which such developmental disorders are recapitulated, commonly display seizures (Powell et al., 2003; Marsh et al., 2009). Furthermore, tissue specimens from patients with epilepsy and brains from experimental animals show a loss and/or dysfunction of INs (de Lanerolle et al., 1989; Tuunanen et al., 1997; Kobayashi and Buckmaster, 2003; Swartz et al., 2006; Choi et al., 2007; Zhang et al., 2009). Further evidence for the role of INs in inhibiting brain excitation is derived from IN ablation studies, which produced seizures (Spampanato and Dudek, 2017). On the contrary, activation of inhibitory INs by chemo- and optogenetic methods or increasing inhibition via systemic or targeted drug delivery in animal models of epilepsy results in suppression of seizure activity (Krook-Magnuson et al., 2013; Hofmann et al., 2016). These findings support the hypothesis that transplanted INs or IN precursor cells could replace the degenerated or dysfunctional endogenous host INs and restore the inhibitory network, ultimately leading to seizure freedom and overcoming intractable epilepsy. Additionally, it is hypothesized that disease progression, including cognitive comorbidities such as learning and memory deficits, as well as hippocampal pathology, may be slowed down or halted by cell therapy (Shetty and Upadhya, 2016).
Transplantation Studies with Fetal Cells and PSC in Models of Seizures/Epilepsy.
Proof-of-concept studies with IN precursor transplantation have utilized medial ganglionic eminence (MGE) derived rodent cells. The MGE is the fetal source of somatostatin (SST) and parvalbumin (PV) expressing INs in the cortex and hippocampus (Anderson et al., 1997; Wonders and Anderson, 2006). Across several models of experimental epilepsy, the disease-modifying effects of MGE-derived IN transplantation have been demonstrated (Baraban et al., 2009; Calcagnotto et al., 2010; Waldau et al., 2010; Hunt et al., 2013; Henderson et al., 2014; Casalia et al., 2017; Romariz et al., 2017). While the use of rodent fetal cells is feasible for preclinical research, utilization of human fetal cells for transplantation would pose challenges in clinical applications due to ethical and safety concerns. To circumvent the need for fetal human cells, protocols for IN generation from pluripotent stem cells (PSCs) are needed. Indeed, in vitro initial differentiation protocols of MGE-like cells from human PSC have been developed, and the cells were extensively characterized in culture as well as following transplantation into rodent host brains (Maroof et al., 2013; Nicholas et al., 2013). Subsequently, some groups have successfully transplanted such human INs into experimental animal models of epilepsy (Cunningham et al., 2014; Upadhya et al., 2019; Waloschková et al., 2021; Zhu et al., 2023; Bershteyn et al., 2023). For clinical translation, however, it will be mandatory to produce and characterize human INs following good manufacturing practice (GMP) principles. The field has made enormous progress in developing protocols to reproducibly generate MGE-like cells from human stem cell lines and in developing technologies to scale up production of these cells for clinical trials (Sheu et al., 2014), and the first-in-human phase I/II trial to evaluate a human IN cell therapy in mesial temporal lobe epilepsy started in 2022.11 https://www.clinicaltrials.gov/study/NCT05135091.
The following sections track the process of the development of a cell therapy product. First, the hypothesis that increasing GABAergic inhibition can reduce seizures needs to be proven. This also includes evaluation of brain target regions for modulating inhibition. One of the following steps in developing a cell therapy for epilepsy is to identify sources for GABAergic INs, which will be covered later. Later clinical translation challenges include the following:
Distinctive and consistently reproducible production of a pure population of GABAergic INs versus projection neurons or cholinergic neurons or glia, which also arise from fetal MGE
Production of a consistent product from cell lines versus patient-derived cells
Extensive scale-up of cell production for clinical use with GMP (no serum, no feeder cultures, sterility)
Ability to cryopreserve cells for distribution and logistics of thawing and preparing cells for transplantation.
The cells need to be characterized pre- and post-transplantation in experimental animals to determine their differentiation and functional properties as well as survival and integration into the host brain. Reproducible efficacy for reducing or preventing seizures of IN therapy is evaluated in preclinical rodent epilepsy models. Additionally, other disease-modifying effects are measured, such as behavioral comorbidities or neuroprotective effects.
Considerations for later clinical application should include the following:
Endogenous influences on cell survival, integration by inflammation, glial scarring in the epileptic host’s brain
Potential lag time because of IN maturation before effects can be expected
Potential loss of efficacy over time or risk of adverse effects due to increased inhibition.
Adverse effects are studied in safety and toxicology studies and are performed under good laboratory principles (GLPs) that regulate uniform quality standards in the development of human health products. Among remaining challenges are the following considerations:
Safe delivery of cells into the patient, and monitoring of cell dispersion and differentiation over time with appropriate tracers and imaging
Immune incompatibility with host if allogeneic (donor of the same species) or xenogeneic (donor of another species) cell transplants are used
Tumorgenicity: tracing and removing transplanted cells, or building in suicide genes
The effectiveness of focal therapy by directly targeting brain regions that are involved in seizure initiation and propagation was assessed with intracerebral drug infusions, which were pioneered by Karen Gale in the 1980s (Gale and Iadarola, 1980; Gale et al., 1983; Gale, 1985; Piredda and Gale, 1985). Intracerebral infusion of GABAergic drugs across multiple animal models of epilepsy and seizures was successful in reducing seizures (for review, see Bröer, 2020). In the most common form of epilepsy, mesial temporal lobe epilepsy (mTLE), 80% of seizures start in the hippocampus (Tatum, 2012). Patients with epilepsy and experimental animals display a loss of hilar INs in the hippocampus (de Lanerolle et al., 1989; Tuunanen et al., 1997; Swartz et al., 2006; Choi et al., 2007); thus, most studies investigating focal therapy have focused on the hippocampus as a target area. Another option is targeting brain regions involved in seizure propagation such as the basal ganglia, which are already established surgical targets for deep brain stimulation in epilepsy and movement disorders (for review, see Bröer, 2020). The rationale for choosing downstream propagation regions of the seizure focus is as follows: (1) some patients have multiple or difficult to localize foci and thus could benefit from a less targeted approach; (2) experimental animal models often do not present with a clear seizure-onset zone; (3) animal models may not resemble the neuropathology that is comparable to the clinical human population.
Drug infusions into the brain pose some challenges: (1) The drug most likely will need to be delivered multiple times or constantly, for example, by a pump. Multiple brain surgeries or a permanent catheter into the brain generate an increased risk of infection and require a high amount of maintenance and compliance by the patients. (2) GABAergic drugs can induce tolerance and consequently lose their efficacy. This has been shown with systemic delivery of benzodiazepines (for review, see Riss et al., 2008), and also experimentally with systemic (Löscher, 1986; Löscher et al., 1996) and focal delivery of GABAergic drugs (Gey et al., 2016). Cell therapy could offer a way out of this dilemma if transplanted cells survived long term and functionally integrated into the network, which should result in a physiological release of GABA when required by the brain instead of a tonic release by a pump, and would optimally require only one injection of cells into the target brain region. On the other hand, cell transplantation could pose safety risks since the transplantation is irreversible. One option to reduce potential adverse effects would be to silence, cells for example, by implementing a “kill switch,” for example, “tet-off,” which leads to the cessation of neurotransmitter release, or by resecting the transplanted brain region.
To achieve potential cell-based therapies for epilepsy, reliable sources of biological/other material will be needed to deliver focal inhibition once grafted into a prospective patient’s brain. Ideally, each material would find a balance of ethical standards met, efficacy in each patient, and long-term reliability to warrant their use in humans. Notably, several fronts have already begun research into possibilities that could meet these milestones and the potential for humans afflicted by various diseases and disorders to receive these developing cell-based therapies. In addition to the therapeutics that currently exist, these new cell-based approaches could help decrease or ablate seizures in individuals as well as work in combination with current therapeutics to achieve a better quality of life. Below, we will discuss the approaches being developed to generate GABA/GABAergic cells for potential cell-based therapies and their potential risks and benefits for human’s impacted by a disease or disorder.
Some of the first attempts to generate a biological source to deliver GABA introduced the GABA synthesizing enzyme genes, Gad1 or Gad2, independently into fibroblasts and a neural immortalized cell line (Ruppert et al., 1993; Conejero-Goldberg et al., 2000). In turn, these cells were able to release elevated amounts of GABA. The advantage of this approach is that cell lines such as these are scalable and could potentially be stored and easily shipped for future use. However, whether these cells could be used therapeutically to locally secrete GABA in distinct areas of the brain still needed to be assessed. Subsequent experiments used engineered cells harboring the Gad genes for transplantation in rodents that were induced to have seizures. Indeed, these cells altered kindling rate, elevated the seizure threshold, and could survive in the host brain for at least 3–5 weeks after transplant (Thompson et al., 2000; Gernert et al., 2002). Overall, these were pioneering studies to demonstrate that increasing local GABA secretion in brain regions that contribute to seizures could be a potential therapeutic option. However, many obstacles still needed to be assessed. First, immortalized cell lines were transplanted in the host brains and while they were present several weeks later, whether these cells lines might continue to divide and cause tumor like masses was not known. Moreover, the scalable feature of these cells was also a positive factor, but as of today the introduction of genetic material into tissue before transplantation is still an unresolved ethical issue for more widespread use in populations impacted by epilepsy and other ailments.
Cell-based therapies are not new; nor are the uses of derivatives acquired from animals in humans to treat disease. A prominent example of this, and exemplary timeline of scientific advances, comes from the research into and treatment of type I diabetes mellitus (TIDM). TIDM is a disease caused by the loss of pancreatic islet/beta cells, which produce insulin. Without normal insulin production, those affected have elevated blood glucose levels and can suffer from organ damage over time. Notably, it was recognized early that insulin was key to alleviating the elevated blood glucose levels and preventing comorbidities later in life. To this end, porcine insulin and beta cell transplantation have been conducted in humans, with success and with signs of what we could expect as inhibitory neuron transplants move forward (for a recent review, see Coe et al. 2020). These are the two major lessons learned: (1) the ability of delivering key reagents to alleviate symptoms is critical (insulin for diabetes and potentially GABA for epilepsy); (2) cell-based therapies are possible, but demand is much more likely to outpace sources of tissues/cells that can be rare.
One of the animals most similar to human physiology is the pig. To this end, some research groups have made significant strides into generating a potential supply chain of tissue/cells that could be used in humans. While porcine tissue has been used in the medical research field for decades, it came with some pitfalls, including the existence of porcine endogenous retrovirus (PERV) genes, of which some subtypes could produce viruses which can transduce human cells (Denner, 2008). Since retroviruses can generate genetic material that can integrate randomly into the host genome, this could potentially be damaging. Over the years some studies have conducted porcine pancreatic beta cell transplants to treat diabetes and while these retroviruses have shown porcine to human cell transduction in vitro, no evidence exists to show that this transmission happens in vivo (Elliott et al., 2000; Wang et al., 2011). Despite these data, researchers have sought to make porcine models that have a very low probability of transducing human cells with their endogenous retroviruses.
To reduce the probability that PERVs and other incompatible porcine genes can be transmitted to human cells, pigs have been progressively genetically engineered to remove these genes from the porcine genome. One example comes from the lab of George Church, where a porcine cell line was targeted using CRISPR-CAS9 technology to delete the multiple copies of PERVs in the porcine genome. After this modification, the engineered nucleus was transferred to a porcine enucleated oocyte to generate an embryo that lacked the PERV genetic material (Niu et al., 2017), which resulted in healthy offspring. In addition, immune rejection is also a major problem with allogeneic organ transplants but could be exaggerated in xenotransplanted cells/organs. One of the main drivers of this rejection is the Gal antigen, which was found to be involved in the most immediate form of transplant rejection that occurs within hours to weeks of receiving a transplant (Cooper et al., 1993). To remedy this problem, groups began engineering pigs in the early 2000s to eliminate the gene for the Gal antigen (Dai et al., 2002; Lai et al., 2002), resulting in healthy offspring that would later be used to assess their potential in the transplant of various organs. While pigs continue to be modified for optimal use in organ transplantation, a recent study characterized the properties of porcine MGE and MGE-derived cells and showed that they have the same migratory potential and ability to generate functional GABAergic INs once transplanted into rodent brains (Casalia et al., 2021). Several years earlier, a small trial with transplantation of porcine GABAergic cells into the seizure focus was initiated, but despite some anticonvulsant effects, the trial was halted due to the potential risk of porcine retrovirus infection in the patients (Schachter et al., 1998). Overall, porcine cells could potentially overcome the limited sources of human donor tissue and be a means to acquire INs for cell-based therapeutics, with the caveats of potentially needing more genetic engineering to create the safest porcine donor cell.
The ability to reprogram human cells into a more pluripotent state and the promise of generating/harvesting PSCs for use in human disease have been progressing for some time. Importantly, the ability to have a source of human cells with the potential to be programmed into multiple cell types has long been a goal of the scientific community, as it has the potential to alleviate symptoms associated with diseases and disorders where known cell types are damaged, lost, or not produced. With regard to epilepsy, generating sources of human IN has been a major goal with several advancements in generating these human cell sources; for a recent review, see Fitzgerald et al. (2020). Since the first successful harvest and in vitro culture of embryonic human stem cells in 1998 by Thompson and colleagues, the field has made major advances in culturing, programming, and utilizing these unique cells.
Early approaches to generate forebrain neuron stem cells utilized rodent cells to assess which factors may yield the most efficient cultures that could give rise to IN progenitors. Since INs are derived from more ventral regions of the forebrain, these initial studies sought to generate neural directed cells that exhibited molecular signatures of the ventral forebrain. One such study found that a combination of Wnt inhibition and sonic hedgehog (Shh) induction yielded differentiated stem cells with greater ventral forebrain identity (Watanabe et al., 2005). The important role of Shh in patterning the stem cells to acquire ventral forebrain properties is still an important factor in the differentiation of IN progenitors (Brady and Vaccarino, 2021) and has even been used to bias INs to produce different ratios of SST versus PV INs (Tyson et al., 2015). One early observation was that while these methods did produce cells with ventral forebrain signatures, the efficiency of producing INs from these cultures was very low. To overcome these limitations and enhance the ability to isolate and utilize these differentiated INs, several groups generated distinct stem cell lines and molecular tools to track and isolate INs at different stages of development and in unique populations.
One approach generated a genetic knock-in human stem cell line that drove the expression of GFP from the Nkx2.1 locus (Goulburn et al., 2011), the initial transcription factor that patterns the MGE (Sussel et al., 1999). This line allowed labs a fluorescent tool to identify when the MGE regional marker, Nkx2.1, is expressed to track and manipulate these cells. Moreover, conserved DNA enhancer elements were used to make stable rodent stem cell lines that drove expression of a fluorescent reporter in cells that could activate these reporters (Chen et al., 2013). Even though the enhancer labeled a small number of cells in the differentiated cultures, the labeled cells had properties of the distinct MGE lineage cell types, demonstrating that specific populations of INs and other cells could be isolated and studied. Subsequent studies using human stem cell lines utilized similar approaches to inhibit Wnt and stimulate with Shh, and generated tissue with ventral forebrain properties, with the caveat that human differentiation and maturation was longer than in rodent cells (Maroof et al., 2013; Nicholas et al., 2013). While each of these reports was also able to generate IN-like cells with molecular markers associated with distinct subtypes, the proportion of subtypes differed, suggesting that manipulations to future protocols could be used to direct and/or bias toward certain IN types.
There has been a protracted effort to generate cell/tissue sources that could deliver GABA to impacted brain regions with a loss of inhibition/elevated excitation. From genetically engineering immortalized cell lines to primary animal and/or human cell sources, each generation has been able to build upon the successes and lessons that came before to get closer to clinical use. One possible route to clinical application is autologous cell transplantation, in which the cells would be derived from the patients themselves. Since the reprogramming of somatic cells to induced PSC is possible (Takahashi et al., 2007), this would enable differentiation of human patient-derived cells into MGE-like INs. While graft rejection would not be a concern with this strategy, an off-the-shelf cell product would have some advantages over an individualized cell product, since large-scale production of allogeneic transplants, for example, from human PSC lines, is more cost-effective and readily available. In general, cells have to be produced according to GMPs to ensure quality and safety of the product. This includes consistent purity of the desired cell product, being free from other cell types as well as microbial contaminations. One of the remaining challenges for clinical translation is large-scale production of cell products without any supplements that would pose a safety concern such as animal-derived serum or feeder cultures (Sheu et al., 2014). Another challenge is cryopreservation of the cell product that does not impact viability after thawing and preparing cells at the hospital before transplantation. With the advent of more streamlined genetic molecular tools and improvements in human stem cell derivation of cell types, the next generation will bring forth new insights and advancements that continue to build and improve upon the current advances into how to focally target, deliver, and control inhibitory tone in the brain.
The transplantation of GABAergic INs into the brains of those impacted by a disease/disorder would predictably elevate GABAergic tone and, in turn, inhibition onto other neurons. Regardless of IN type, an elevation in GABA is likely to occur; however, each IN subtype exerts distinct molecular, cellular, and electrophysiological properties that could result in either elevated inhibition or potentially excitation (Kessaris et al., 2014; Tremblay et al., 2016; Lim et al., 2018; Huang and Paul, 2019). Moreover, the relative proportions of different IN types could drive unique changes in circuitry in the microcircuit in which they incorporate. This could impede heterogeneous IN transplants but may be overcome by using an enriched and/or pure population of IN subtypes that would best correct the diagnosed symptoms in the altered microcircuit of a patient. Below, we discuss some of the unique properties of IN subtypes.
An earlier chapter (see Chapter 47, this volume) reviewed the early developmental sources of cortical INs that are derived from the MGE and caudal ganglionic eminence (CGE), progenitor regions that generate the majority of INs. Once they are mature and have incorporated into their local microcircuit, each group of INs performs distinct tasks in the circuit that are largely governed by their different molecular makeups (transcription factors, ion channels, etc.), morphologies (axon and dendrite targets and innervation), and electrophysiological properties. Together, these diverse properties contribute to action potential amplitude and frequency, laminar control of electrical information, and the overall balanced code in the microcircuit that underlies how our brain processes sensory input and the resulting output to other nuclei to execute human behavior.
MGE-lineage INs are primarily composed of those expressing the peptide SST or the calcium binding protein PV, which are found in roughly equal numbers and together comprise about 70% of all cortical INs in rodent (Wonders and Anderson, 2006). CGE lineages can be divided into two broad groups of either vasoactive intestinal peptide (VIP) or Reelin/neuron-derived neutrophic factor (Ndnf) expressing INs (Wonders and Anderson, 2006; Miyoshi et al., 2010; Abs et al., 2018); the latter will be referred to as neurogliaform cells (NGFCs). Each of these four broad groups can be further divided into more distinct types by the presence/absence of other proteins, including transcription factors, ion channels, and other unique proteins. Each group of INs in the neocortex exhibits a unique laminar pattern and axon/dendrite morphologies, which was compared earlier in a seminal review (Huang et al., 2007). Briefly, both PV+ and SST+ INs are enriched in deeper layers of the cortex, while VIP+ INs are biased to reside in layers 2/3 and NGFCs in layer 1. Figure 78–1 shows some of the distinct laminar, molecular, and axon projections for different IN groups and subgroups. In addition to these differential enrichments of IN subtypes in cortical layers, the ratios of subtypes can also vary in different cortical regions, with areas that reside closer to where the MGE developed harboring greater MGE-lineage INs and vice versa for areas closer to where the CGE developed (Fazzari et al., 2020). Within the microcircuit, additional differential properties can be found. Some of these unique molecular signatures as well as their distinct innervation into the circuit and properties may be exploited in future therapeutic cell transplant studies.
Molecular and morphological features of cortical interneurons. Schema depicting some of the unique features associated with cortical INs. In addition to their different laminar profiles, there are unique features to specific groups and subgroups of interneurons, (more...)
PV+ INs: PV+ INs have elevated expression of the Kv3.1 fast-inactivating potassium channel, allowing these cells to fire rapid action potentials without a decrease in amplitude, that is, accommodation (Rudy and McBain, 2001). In contrast to other IN groups, PV+ INs lack many unique molecular markers, making this group more difficult to study at earlier ages before PV and Kv3.1 expression appear. However, recent studies have identified the gene Mef2c to be enriched in PV lineages at earlier ages (Mayer et al., 2018; Pai et al., 2020). PV+ INs target different regions of a pyramidal neuron depending on their subtype; basket cells innervate the soma, while chandelier cells can target the axon initial segment (Somogyi et al., 1998; DeFelipe et al., 2013). PV+ INs have also been implicated in underlying gamma rhythms that can be measured in the brain and which are thought to underlie mechanisms of learning and memory (Pesaran et al., 2002; Bauer et al., 2007; Cardin et al., 2009; DeFelipe et al., 2013).
SST INs: Recent assessments of SST+ INs have revealed several subtypes based on the differential expression of molecular markers, making this one of the most diverse IN groups (for a review on the different properties of SST+ INs, see Riedemann, 2019). This diversity was observed from earlier transgenic lines that revealed distinct SST subgroups that could be distinguished by the presence or absence of calbindin expression as well as associated morphologies and electrical properties (Riedemann, 2019). This diversity could also be part of the reason for diverse electrical properties as well, including SST+ INs that have slower/regular action potential firing or bursting properties (Kawaguchi and Kubota, 1997). Within the local microcircuit, one class of SST+ INs, Martinotti cells, synapse onto the distal dendrites of pyramidal cells (Wang et al., 2004). While the Martinotti cell has been well studied, other SST+ IN subgroups have been examined with differing electrical and morphological properties, including those with more locally projections and multipolar or bipolar morphologies (McGarry et al., 2010). If these distinct subtypes could be selectively programmed or isolated, it would be an advantage in more targeted cell transplants.
VIP INs: VIP+ INs can synapse onto other INs that normally inhibit excitatory pyramidal cells, leading to a situation whereby VIP+ IN activity leads to a disinhibition of pyramidal cells and an overall increase in local excitability (Lee et al., 2013; Pfeffer et al., 2013; Pi et al., 2013). This unique property may allow VIP+ INs to have the opposite effect on an epileptic brain than intended in cell transplantations. Many VIP+ INs also have a bipolar morphology, with processes extending in both dorsal and ventral directions (Connor and Peters, 1984; Morrison et al., 1984). Another unique feature of these INs is the presence of a subgroup that expresses choline acetyltransferase (Houser et al., 1983), allowing co-release of both acetylcholine and GABA onto synaptic targets.
NGFCs: NGFCs can produce slow, nonionotropic, GABA inhibition onto pyramidal neurons (Tamás et al., 2003). This may have to do with their unique property to bulk/volume release GABA locally to inhibit other neurons in the vicinity (Overstreet-Wadiche and McBain, 2015). In addition, NGFCs have a late action potential firing response to a stimulus, referred to as late spiking (Miyoshi et al., 2010). At the molecular level, NGFCS express some unique markers, including alpha-actinin-2 and Ndnf (Price et al., 2005; Abs et al., 2018). Finally, as noted above, their ability to integrate into cortical layer 1, bulk release GABA, and other distinct features may be utilized in targeted transplant approaches in the future.
Similar to targeted drug infusion studies, cell transplantation studies have also focused on seizure generation and propagation in regions such as the hippocampus and basal ganglia. It has been described that cell survival is higher in brain regions that are affected by the disease, such as the hippocampus in epilepsy, because they might be more permissive for new cells due to endogenous cell death and degeneration (Zaman and Shetty, 2003).
Another factor to consider is the donor cell type for transplantation. While a decrease in the susceptibility to elicited seizures and a reduction in spontaneous seizures have been reported after transplantation of genetically engineered mouse cells that produce GABA, these effects were mainly transient (Gernert et al., 2002; Thompson and Suchomelova, 2004). Transplanted cells displayed a low survival, which would limit their applicability as a long-term therapeutic for chronic epilepsy, and in some cases strong tissue reactions such as graft rejection, massive infiltration of inflammatory cells, and gliosis have been reported (Nolte et al., 2008; Handreck et al., 2014). Other studies that have used fetal GABAergic precursor cells from other sources than MGE (Holmes et al., 1991; Holmes et al., 1992; Löscher et al., 1998), or that did not confirm MGE identity by immunohistochemistry (Fine et al., 1990; Backofen-Wehrhahn et al., 2018) reported no, moderate, or short-term moderate effects on seizures. Some potential reasons are likely the lack of migratory potential of non-MGE GABAergic cells, including those from the lateral ganglionic eminence (LGE) (Wichterle et al., 1999), or early-born MGE-lineage neurons like projection neurons from the globus pallidus, which could lead to cell clusters at the site of injection and impaired functional integration into the host brain. In addition, it is possible that other GE neurons could have been part of the transplants, including VIP+ INs derived from the CGE, which can drive increases in excitation due to their inhibition of other INs and subsequent disinhibition of pyramidal neurons (Acsády et al., 1996; Pfeffer et al., 2013). Another reason for lack of efficacy could be attributed to low cell survival depending on the cell source and target region in the host’s brain. There is evidence that grafted cells persist preferably in regions that they were developmentally supposed to migrate to, for example, for MGE-derived INs to hippocampus and cortex.
The main cell type that has been shown to successfully elicit disease-modifying activity in models of epilepsy are INs, which are primarily generated by the MGE, the fetal source of INs of the cortex and hippocampus. Many groups have shown long-lasting seizure suppression by using MGE-derived rodent cells in rodent models of epilepsy. Results from cell transplantation studies with MGE-derived cells or cells from other fetal sources in preclinical rodent models of epilepsy are summarized in Table 78–1. A complete review of most of the studies using GABAergic progenitor cells was provided in the last edition of this book by Anderson and Baraban (2012). The MGE was chosen for cell isolation by many researchers because of their long-distance migration, differentiation, and integration as mature INs comprising several IN subtypes, increase of GABAergic inhibition in the epileptic brain, and their efficiency to reduce seizures in several experimental rodent models (cf. Anderson and Baraban, 2012).
While use of rodent fetal cells is feasible for preclinical research, the required utilization of human fetal cells for transplantation could pose challenges in clinical applications due to ethical, safety, reproducibility, and supply concerns. Subsequently, in vitro differentiation protocols of MGE-like cells from human PSCs have been developed (Nicholas et al., 2013; Maroof et al., 2013). Human cells were extensively characterized in culture and in organoids as well as following transplantation into rodent host brains regarding their maturation, subtype differentiation, and electrophysiological properties (Nicholas et al., 2013; Xiang et al., 2017). It has been confirmed that these cells developed into functional GABAergic INs; however, the intrinsic duration of the maturation process followed the longer timeline of human development (Nicholas et al., 2013; Maroof et al., 2013). The protracted maturation makes preclinical studies in rodents challenging because a duration of several months is necessary to allow transplanted cells to mature within the host. However, it is critically important that cells are transplanted in a more immature state, in which newly committed INs are still migratory, which enables dispersion throughout the desired target region and functional integration into the circuitry (Anderson and Baraban, 2012). More recently, researchers started to study human IN transplantation with non-clinical-grade cells in experimental models of epilepsy. Seizure suppression effects were comparable to data from rodent cell transplantation, which reported 72%–90% seizure reduction. Transplantation of 300k human cells per hemisphere reduced spontaneous seizures by 72% at 5 months post transplant (MPT) in the systemic kainate rat model (Upadhya et al., 2019). The average duration of seizures was unchanged. This was the first report of a potential path to patient-specific autologous MGE cell transplantation for epilepsy since the cells were generated from induced human PSCs (Upadhya et al., 2019). They were transplanted 4 weeks after neural induction directly from culture into immunosuppressed rats 7 days post epilepsy induction. While such an early intervention within days after the initial epileptic insult, and before animals reliably display spontaneous seizures, is not going to be a treatment strategy for clinical patients, it offers a proof of concept. The seizure suppression in the cell-treated group was consistent across the 3-week electroencephalogram (EEG) recording period. The authors confirmed that the effect was partly reversible with a graft cell silencing approach via designer receptors exclusively activated by designer drugs (DREADDs; Upadhya et al., 2019). For clinical translation, however, eligible patients for cell therapy would most likely be patients with intractable chronic epilepsy.
Human cell transplantation in a chronic seizure stage was performed by Cunningham and colleagues in 2014. They transplanted cells in the pilocarpine mouse model and found that about 50% of their mice were seizure-free at 3 MPT with 200k transplanted cells/hemisphere. In line with the previous report, the average duration of seizures was not altered. The decision to monitor seizures by EEG for a rather short period of 5–10 days on one occasion over the course of the study presents a shortcoming of this study as clustering of seizures can lead to silent phases without seizures of several days in this model (Henderson et al., 2014). Cells were generated from H7 human embryonic stem cells (ESCs) and transplanted fresh after sorting for neural cell adhesion molecule 1 (Cunningham et al., 2014). Waloschková et al. (2021) and Zhu et al. (2023) similarly found antiseizure effects after transplantation of human ESC-derived GABAergic progenitors that were induced in vitro. Transplantation of fresh cells from culture is feasible for research purposes, but clinical applications would require a cryopreserved GMP-manufactured product. A clinical-grade human IN cell therapy derived from a human PSC line has recently been developed for chronic focal epilepsy (Bershteyn et al. 2023), and is now benig evaluated in a clinical trial.1 It has been reported that the transplanted INs have shown a greater than 75% electrographic seizure reduction in 72% of animals in the intrahippocampal kainate mouse model of temporal lobe epilepsy. Similarly, the cumulative duration of seizures was reduced. These antiseizure effects persisted long term 4–9 MPT Bershteyn et al., 2023. Two thirds of cell-treated animals were seizure-free by 6–7 MPT versus none of the animals in the vehicle control group (Bershteyn et al., 2023). Animals receiving high doses of the human IN cell therapy were evaluated in a battery of behavioral assays, and no behavioral abnormalities were reported.
The described cell-mediated effects are comparable to the performance of systemically applied GABAergic drugs in experimental epilepsy models (see Fig. 78–2); for example, 3 mg/kg diazepam i.p. reduced the number of electrographic seizures by 73% in the hour after injection compared to vehicle-injected mice in this model (Duveau et al., 2016). Cell therapy can potentially increase GABAergic inhibition in a more spatially restricted and synaptically regulated fashion than systemic GABA enhancing drugs, and consequently present with fewer systemic side effects. This would be advantageous, because drug adverse effects are one of the major reasons for discontinued treatment in clinical patients (Giussani et al., 2017); another issue is loss of efficacy over time due to tolerance, which has also been shown for GABAergic drugs in experimental models (Löscher, 1986; Gey et al., 2016). The potential advantage of cell therapy is that it is a targeted and long-lasting approach.
Comparison of reduction in seizure frequency across exemplary studies with the GABAergic drug diazepam, mouse MGE-derived cells, and several studies with human IN transplantation. For a direct comparison percentages in seizure reduction were estimated (more...)
Studies on rodent MGE-derived INs have shown that these progenitors are ideal candidates for transplantation studies since they are still migratory at the time of transplant (for review, see Anderson and Baraban, 2012), while other cell types display a less favorable migratory potential (Wichterle et al., 1999, 2001). This allows the grafted cells to disperse within the target structure and to potentially integrate where endogenous INs have degenerated or increased excitation has been generated in the epileptic brain. Importantly, transplanted INs follow the intrinsic cues in the rodent brain and spinal cord to stay within boundaries at neonatal ages and in the adult nervous system (Bráz et al., 2012; Hunt et al., 2013; Vogt et al., 2014), making it possible to deliver and restrict INs exclusively into the structures of interest. In 2017, Birey and colleagues were able to model human neural spheroid cultures of forebrain subdomains and recapitulated unidirectional fetal human IN migration from ventral forebrain toward the cortex after spheroids fused. In transplantation studies with human cells in epilepsy models human INs extensively migrated within the epileptic hippocampus up to 1.6 mm from the injection site (Cunningham et al., 2014), which is comparable to migration of transplanted rodent MGE cells (Hunt et al., 2013). However, some nonmigratory cells clustering close to the injection sites were detected (Cunningham et al., 2014; Upadhya et al., 2019), which could point to a different identity and fate of grafted cells.
In order to increase local inhibition, which is the suspected mechanism of action of IN transplantation, transplanted INs should functionally integrate into the circuitry by forming synapses onto and receiving input from host cells. This has been shown with electrophysiological characterization and optogenetics for rodent cells by Henderson et al. (2014) and for human INs by several groups. Even though the human transplants were not fully mature by earlier time points of recording (4-5 MPT), functional synaptic integration was evidenced by excitatory input from host glutamatergic neurons, as well as by induced inhibitory synaptic responses in host neurons by GABA from transplanted human INs that could be blocked by bicuculline, an inhibitor of ionotropic GABA-A receptors. (Cunningham et al. 2014; Zhu et al. 2023; Bershteyn et al. 2023) At later recording time points transplanted cells showed continued maturation (Zhu et al. 2023; Bershteyn et al., 2023). Similarly, immunohistochemical evidence for synapse formation involving rodent or human INs and host neurons has been provided (Henderson et al., 2014; Cunningham et al., 2014; Upadhya et al., 2019; Zhu et al., 2023; Bershteyn et al., 2023).
It has been speculated that transient effects of cell transplantation could in part be allocated to a low persistence of transplanted cells (Backofen-Wehrhahn et al., 2018). Persistence of rodent MGE transplants has been described to range between 15% and 30% of initially transplanted dose depending on the time of evaluation post transplant (Waldau et al., 2010; Zipancic et al., 2010; Hunt et al., 2013; Casalia et al., 2017). While allotransplants such as mouse MGE transplanted into mice from proof-of-concept studies did not pose significant challenges regarding cell survival, xenotransplantation typically requires immunosuppression in order for the grafts to survive. The groups using hPSC-derived GABAergic derivatives (Cunningham et al., 2014; Upadhya et al., 2019, Waloschková et al. 2021; Zhu et al. 2023; Bershetyn et al. 2023) were using either immunodeficient animals or immunosuppressive drugs to ensure xenograft survival. Persisting human cells in a mouse epileptic hippocampus 10 months after transplantation are shown in Figure 78–3. The persistence of 20% in Cunningham’s study (2014) was comparable to rodent MGE data; however, the high persistence of 129% in Upadhya’s study (2019) reflected cell expansion due to proliferation of transplanted cells, which poses a safety concern that would need to be addressed before clinical transplantation. Both studies were not able to confirm depletion of proliferating cells before transplantation and found some grafted cells that were still positive for the proliferation marker Ki67 several MPT (>1%). While neural stem cell markers (Nestin, Sox2) were still present, a marker for pluripotency (Oct4) was not detected, and no teratoma formation was observed in any of the transplanted rats (Upadhya et al., 2019). Waloschková et al. (2021) reported that their transplanted cells were devoid of markers for proliferation or neural stem cell markers. Recent reports (Zhu et al. 2023; Bershteyn et al. 2023) observed lower human cell persistence of about 6-10% of the initial dose, however they did not observe any remaining cycling cells (KI67) or other stem cell markers several months after transplantation. No teratomas, uncontrolled or ectopic growth of human-derived tissue was reported (Zhu et al. 2023; Bershteyn et al. 2023).
Human cells persist, disperse, and express markers of INs up to 10 months post transplant in the kainate-treated hippocampus of a mouse. DAPI, 4,6-diamidino-2-phenylindole; LHX6, marker for medial ganglionic eminence cells; SST, somatostatin. From Bialer (more...)
Apart from the tumorigenic potential associated with impurities of remaining stem cells, another concern is off-target cell populations that could have other effects depending on their fate. Based on current research, astrocytes and oligodendrocytes were sporadically observed in human cell transplants (Cunningham et al., 2014; Upadhya et al., 2019). Astrocytic secretion factors such as -derived neurotrophic factor (BDNF) could have a favorable effect on seizure suppression, as evidenced by earlier reports by the same group (Waldau et al., 2010). However, in other reports increased astrocyte proliferation was associated with an increase in seizure activity (Amiri et al., 2012).
Another example for an off-target cell population would be other neuronal cells. GABAergic INs from CGE, projection neurons and cholinergic neurons are closely related to GABAergic INs since some of these populations are also MGE-derived and were also present in the described human transplantation studies by Cunningham et al. (2014), Upadhya et al. (2019), and Waloschková et al. (2021). They may lead to nonspecific innervation and circuit modulation within the host brain and thus could cause safety concerns.
One option to confirm MGE-derived cortical IN identity could be to label for the expression of Lhx6 and MafB, as well as the loss of Nkx2.1 expression. The continuous expression of Lhx6 and reduction of Nkx2.1 is a well-documented event in cortical IN maturation (Wonders and Anderson, 2006). However, even if Nkx2.1 decrease is verified, Lhx6 expression is not specific for cortical or hippocampal INs but could also be evidence of globus pallidus projection neurons or distinct hypothalamic neurons (Kim et al., 2021). Recently, Mafb was shown to be enriched in MGE cortical INs but not in other Lhx6+ cells, making it a potential new tool in the validation of cortical and hippocampal INs (Pai et al., 2019). Bershteyn et al. (2023) showed that the majority of transplanted human cells (>80%) labeled positive for Lhx6 (see Fig. 78–3). In other studies, Lhx6 or Mafb genes were not assessed (Upadhya et al. 2019) or only 28–40% of human cells were positive for Lhx6, while about 30% of cells still expressed Nkx2-1 (Cunningham et al., 2014; Zhu et al., 2023). This could indicate the presence of GABAergic projection neurons, striatal INs, or MGE progenitors. Moreover, a small population of VIP-positive neurons has been detected in single-cell RT-PCR of grafted human cells (Cunningham et al., 2014), as well as by immunohistochemistry after transplantation (Zhu et al. 2023). Inhibiting VIP neurons can delay seizure onset and reduce seizure duration experimentally (Khoshkhoo et al., 2017). However, VIP INs in a graft could pose a potential concern due to their inhibition of other INs as mentioned earlier.
Even if mixed transplants could potentially have synergistic effects, it would be hard to demonstrate safety, reproducibility, and standardization of graft composition for regulatory approval; thus, a cell therapy product will need to prove a high purity of the desired cell type. Despite some issues with purity, human cell transplants consisted of a high proportion of GABAergic neurons (>76%) and markers of several IN subtypes, including those expressing SST (see Fig. 78–3), PV, calretinin, and Neuropeptide Y (Cunningham et al., 2014; Upadhya et al., 2010; Zhu et al., 2023; Bershteyn et al., 2023), which are likely responsible for the antiseizure effect of the transplants. While these findings seem promising, some of the GABAergic cells are still undefined, which future studies will assess. In addition, many of the cells express somatostatin but there were few to no parvalbumin expressing cells found. All cited studies used similar doses of cells and found comparable seizure suppression.
The epileptic hippocampus displays a typical pathology, including dispersion of the granule cell layer, neurodegeneration, mossy fiber sprouting, and hippocampal shrinkage in human patients (Sutula et al., 1989; Blümcke et al., 2017), which is to some extent recapitulated in experimental models (Bouilleret et al., 1999; Depaulis and Hamelin, 2015). Novel therapies should also aim to slow or prevent the progressive neuropathology of epilepsy. Data from some studies did not reveal any changes in pathology and suggest that an effect on hippocampal damage is not necessary for modulating seizure activity (Hunt et al., 2013; Cunningham et al., 2014). However, a pathological hallmark of epilepsy in patients and experimental animals, granule cell dispersion, was significantly reduced in some studies (Zhu et al. 2023; Bershteyn et al. 2023), with one study even reporting a dose-dependent effect of IN transplantation (Bershetyn et al. 2023). Then start new sentence with Upadhya et al. (2019) found that mossy fiber sprouting as well aberrant neurogenesis in the dentate hilus, both of which are suspected to contribute to hyperexcitability of the dentate gyrus, were reduced in human IN transplanted rats, while other reports declared no changes to mossy fiber sprouting after transplantation (Zhu et al. 2023). Furthermore, host INs in the hilus were preserved compared to nontreated epileptic rats. These effects could be attributed to the early intervention just days after status induction, since the grafts might have modified the insult severity itself or had an effect on the development of epilepsy. This is also supported by the fact that silencing grafted cells at a late timepoint did not fully reverse seizure suppression (Upadhya et al., 2019). Further studies are needed to dissect the role of improving hippocampal pathology or preventing progressive damage associated with seizures in the context of cell therapy.
Similar to exploring neuroprotective effects of treatments, “strategies that suppress [spontaneous recurrent seizures] (e.g. increased inhibition) may not necessarily improve cognitive function and mood in chronic epilepsy” (Shetty, 2012); however, it would certainly be beneficial to explore the disease-modifying efficacy of novel treatments in this respect as well. Patients with seizures suffer from a range of comorbidities such as anxiety, depression, learning and memory deficits, and hyperactivity or aggression (Keezer et al., 2016), and some studies on rodent MGE transplantation have given the first evidence for an alleviation of such behavioral correlates in models of epilepsy (cf. Table 78–1; Hunt et al., 2013; Casalia et al., 2017). For human IN transplantation, Cunningham et al. (2014) showed that deficits which are displayed by epileptic animals in short-term working memory (Y maze), recognition and exploration of novel objects, hyperactivity, and aggression were reversed in mice that received INs. Comparable results were seen by Upadhya et al. (2019), Zhu et al. (2023) and Bershteyn et al. (2023), who reported that deficits in cognitive, pattern separation, and recognition memory tasks were also reversed by IN transplantation. Furthermore, animals were tested in a paradigm for eating-related depression, in which epilepsy leads to decreased motivation for food, which was not observed in transplanted animals (Upadhya et al., 2019). Similarly, transplanted animals did prefer sweet water like naive animals, while epileptic nontreated animals displayed anhedonia and did not show sucrose preference (Upadhya et al., 2019; Zhu et al., 2023). Furthermore, cell transplanted mice did show reduced mortality compared to vehicle-injected mice that partly experienced sudden, unexpected death or required euthanasia due to deteriorating health in long-term studies in the intrahippocampal kainate model (Bershteyn et al. 2023), however this effect was not observed in other studies (Zhu et al. 2023).
The potential for IN transplantation to alleviate symptoms of various diseases and disorders is underscored by the observations of either IN loss or dysfunction in both humans impacted by these and the potential for alleviation of symptoms in recent animal model studies. Notably, in addition to the obvious role of IN transplantation in epilepsy, other disorders may benefit from the advancement of IN transplantation, with much attention on neuropsychiatric disorders. Herein, we will discuss the various studies and observations involving IN loss and dysfunction in human populations and animal models underlying these conditions. While not discussed in this chapter, it should be noted that INs are also impacted in neurodegenerative diseases such as Alzheimer, Parkinson, and Huntington disease (Verret et al., 2012; Schmid et al., 2016; Shetty and Bates, 2016; Lozovaya et al., 2018; Holley et al., 2019; Hijazi et al., 2020). However, the role of IN loss and dysfunction in these diseases is still poorly understood and how IN transplantation could alleviate symptoms is just beginning.
In addition to a role in epilepsy, IN dysfunction is also often implicated in neuropsychiatric disorders, including autism spectrum disorder (ASD) and schizophrenia (SCZ). An early hypothesis for an altered inhibitory tone in ASD was proposed in the early 2000s (Rubenstein and Merzenich, 2003). Since then, a great deal of work has been done to show that, in many ways, an unbalanced inhibitory to excitatory balance can be associated with phenotypes associated with ASD, with much of the data coming from animal models assessing monogenic syndromes with high rates of ASD (Sohal and Rubenstein, 2019). Many studies highlight that certain IN subgroups, notably those expressing PV, are commonly changed in ASD gene models. However, it should be noted that other INs are also impacted in different models and the experiments examining CGE-lineage INs are well behind MGE-derived due to a lack of tools that have just become available in recent years but with promising results. Some recent studies have noted the role of the syndromic genes MeCP2 and ErbB4 in VIP IN function (Batista-Brito et al., 2017; Mossner et al., 2020).
Overall, there is ample evidence for IN dysfunction in human conditions other than epilepsy, and there is a need to explore whether IN cell transplantation could also be a viable option for those diagnosed with these conditions. Furthermore, it should be reemphasized that there is not a one-size-fits-all approach, as each of these models clearly demonstrate that while PV+ INs are often impacted, they can be altered in different ways. Moreover, other IN cell types may be affected in distinct syndromes, suggesting that a more precision medicine approach, tailored to each disorder/diagnosed patient, needs to be taken as the field moves forward and builds upon our growing knowledge of unique and beneficial IN properties that could be targeted for therapies.
One of the main safety considerations in progenitor cell transplantation is tumorigenicity. As discussed before, no PSCs have been detected in aforementioned studies on human IN transplantation (Upadhya et al., 2019); however, cells that are already committed to the neurogenic lineage, but are still proliferative, could pose significant safety concerns. Optimally other neurons such as cholinergic neurons and globus pallidus projection neurons that also arise from fetal MGE should be eliminated from the cells to be transplanted. Despite some reported impurities, no ectopic tissues or teratoma formation has been reported for up to 12 MPT (Cunningham et al., 2014; Upadhya et al., 2019; Bershteyn et al., 2023). Nevertheless, rigorous tumorigenicity studies at high doses must be investigated further to ensure safety of cell therapy candidates, and biodistribution and toxicology studies need to be performed as well.
Another safety issue could be posed by sedative effects via increased inhibition from transplanted cells. Electrophysiological analysis revealed no over-inhibition of host neurons by transplanted INs (Zhu et al., 2023). Additionally, a modified Irwin Screen has been performed on mice up to 9 MPT and suggested that transplantation of MGE-type human IN is well tolerated and does not cause obvious signs of sedation or motor impairment (Bershteyn et al., 2023), however this should be monitores as different labs begin to develop their own cells for transplantation. The behavioral testing battery included paradigms that are used to evaluate the safety and tolerability of antiseizure drugs or drug combinations in animal models (Klee et al., 2015). Other paradigms such as Open Field test and Learning and Memory tests have not revealed increased anxiety or cognitive deficits compared to untreated epileptic mice (Zhu et al., 2023; Bershteyn et al., 2023).
A trial with local GABAergic drug infusion into the hippocampus has been published recently (Heiss et al., 2019), in which two out of three patients became seizure-free 2 years post drug infusion, and one had less frequent seizures, with no adverse effects reported. However, these results could not be directly attributed to prior drug infusion and thus are inconclusive. An earlier small trial with transplantation of porcine GABAergic cells into the seizure focus reported promising anticonvulsant effects, but the trial was halted due to the potential risk of porcine retrovirus infection in the patients (Schachter et al., 1998). Currently, the first-in-human clinical trial with human IN transplantation in epilepsy is recruiting patients in phase I/II1.
Other neurological disease indications have prompted clinical development of cell therapies. A rather similar approach to the described efforts with human MGE-type GABA IN (Bialer et al. 2020) is the cell therapeutic candidate MSK-DA01 midbrain-type dopamine projection neurons derived from hESCs for treatment of Parkinson disease. Apart from improvements of the behavioral phenotype in their preclinical Parkinson disease animal model, 6-hydroxydopamine lesioned rats, the cell product candidate was subjected to an extensive set of biodistribution, toxicity, and tumorigenicity assessments in mice, which are described by Piao et al. (2021). No adverse effects, cells outside of the target region, ectopic tissue, or tumor formation was observed. The clinical trial is recruiting patients to replace the characteristic loss of substantia nigra dopaminergic neurons by cell transplantation into the striatum.2 The progression of this next wave of human neural cell therapy candidates from preclinical studies to clinical investigation will provide key insights regarding production, delivery, dosing, trial design, imaging, immune system reactivity, safety, and, ultimately, potential efficacy considerations for future treatment of multiple neurological disorders.
Despite available treatment strategies such as antiseizure drugs, focal stimulation, and surgical resection, about one-third of patients suffering from epilepsy are not satisfyingly treatable. New therapeutics for epilepsy are desperately needed. Cell transplantation could offer a way to overcome pharmacoresistance if cells can survive long term within the host brain, migrate into appropriate brain regions, differentiate into INs that increase inhibitory tone within the epileptic network, and integrate into the circuitry and form synapses with host neurons.
Several studies have demonstrated all of the above-mentioned prerequisites of a successful IN transplantation. They showed that INs are effective in decreasing seizure frequency consistently and long term across a variety of animal models. Behavioral comorbidities of epilepsy could be alleviated, and even hippocampal neuropathology of the disease was reduced. Human IN transplantation was effective in reducing seizure burden, while no obvious adverse effects were associated with it. No ectopic tissue formation or teratomas were recorded in transplanted animals, but some impurities with off-target cell populations were still observed within the grafts. Further research on characterization of cells prior to transplant, process improvements, and GMP manufacturing is ongoing.
While there are many challenges that needed to be addressed before clinical trials could be launched, the recent advances in our ability to generate specific neural cell sublineages from PSCs, in-depth characterization of cells with modern techniques, and the combinatorial use of electrophysiology, optogenetics, and translational immunodeficient animal models to evaluate disease-modifying activity have brought human cell therapy significantly closer to clinical application for the treatment of epilepsy.
We thank Jill Helms for critical review of the manuscript. Daniel Vogt is supported by the Spectrum Health-Michigan State University Alliance Corporation and the Autism Research Institute.
Sonja Bröer is a former employee and current shareholder of Neurona Therapeutics, Inc. Daniel Vogt has no disclosures.
This is an open access publication, available online and distributed under the terms of a Creative Commons Attribution-Non Commercial-No Derivatives 4.0 International licence (CC BY-NC-ND 4.0), a copy of which is available at https://creativecommons.org/licenses/by-nc-nd/4.0/. Subject to this license, all rights are reserved.
Your browsing activity is empty.
Activity recording is turned off.
See more...
