Physiological properties such as RMP and RIN
Physiological properties such as RMP and RIN are of interest in terms of cochlear cell-based therapy, because these properties contribute to the overall responsiveness of replacement neurons to stimulation from a cochlear implant. Promisingly, our population of stem cell-derived neurons responded to low frequency pulsatile stimulation with high precision, however they were not able to match the efficiency of postnatal primary auditory neurons to entrain to high frequency input. Despite an extended period in culture, which we hypothesized may be sufficient to induce more electrically mature phenotypes, the present data suggest that other factors are required to reach this stage of maturation. Spontaneous activity and synapse formation are possible key steps in the production of mature electrical phenotypes from stem cell-derived neurons.
As indicated by the work of Marrs and Spirou (2012), one of the key factors in the maturation of auditory neurons is the formation of synapses with hair cells. Moreover, spontaneous firing may assist in directing the formation of specific synaptic connections and in activating auditory neurons during embryogenesis (Lippe, 1994). Spontaneous action potentials generated by immature inner hair thip are primarily mediated by Ca channels, which disappear around the onset of hearing (Kros et al., 1998). The release of neurotransmitter at the basal surface of the inner hair cell is likely to modulate the activity of the developing auditory nerve (Beutner and Moser, 2001; Beutner et al., 2001). In a similar way, electrical stimulation has been shown to promote survival of auditory neurons in vitro (Hegarty et al., 1997; Hansen et al., 2001) and in vivo (Shepherd et al., 2005), and normal synaptic activity can be recovered following chronic electrical stimulation in congenitally deaf cats (Ryugo et al., 2005). Taken collectively, these data support the idea that spontaneous activity or electrical depolarization is an important feature in the terminal differentiation of auditory neurons, and this may be essential in order for them to reach their mature electrophysiological phenotype.
While the described population of stem cell-derived neurons was not observed to achieve a mature electrical phenotype in vitro, they may be capable of doing so once transplanted in vivo. Recent evidence suggests that stem cell-derived neurons can incorporate into the deaf gerbil cochleae, and improve auditory-evoked response thresholds after 10weeks (Chen et al., 2012). Interestingly, the stem cell-derived neurons used by Chen et al. (2012) displayed similar basic firing properties prior to their transplantation, to the stem cell-derived neurons described in the present study. Moreover, the voltage-gated ion channels that control neural activity are sensitive to changes in their environment, including concentration of neurotrophic factors (Adamson et al., 2002a; Zhou et al., 2005; Needham et al., 2012; Purcell et al., 2013) and excitatory input (Leao et al., 2005; Holt et al., 2006; Hassfurth et al., 2009), both of which are likely to play a role following their transplantation into the deaf cochlea. Determining how to harness and balance both in vitro and in vivo factors to derive functionally appropriate stem cell-derived neurons will be an important task in developing a stem cell therapy for the deaf cochlea.
Introduction Embryonic stem cells have a great potential in regenerative medicine because they generate somatic cell types from all three germ layers. For example, insulin-producing pancreatic beta-cells derived from human embryonic stem cells (hESCs) can be applied in type-1 diabetes. One of the problems to overcome is that it has proven very difficult, if not impossible, to obtain fully differentiated and functional beta-cells in vitro. Currently, it is possible to generate pancreatic progenitors from hESCs, and they were shown to differentiate into functional beta-cells after a prolonged period of engraftment in mice (Kroon et al., 2008; Mfopou et al., 2010a; Nostro et al., 2011; Rezania et al., 2012). Further optimization is needed to establish suitable conditions required for beta-cell differentiation in vitro. To this end, the study of mouse embryonic stem cells (mESCs) can be useful for several reasons. First, in vitro differentiation protocols are intended to mimic the conditions in the developing embryo and most knowledge on embryonic development of the pancreas was accumulated from mouse studies. Furthermore, the timeline for the specification and maturation of a particular cell is theoretically shorter in mice than humans (respectively 3weeks and 40weeks). Second, mESCs are easier to maintain in culture as they grow faster, are more resistant to enzymatic dissociation during passaging, and have less tendency to spontaneously differentiate. mESCs can thus be used as a model to rapidly tweak protocols that thereafter could be implemented on hESCs, for example to obtain beta-cells. However, despite the similarities in their general properties, embryonic stem cells from mouse and human differ in their manipulation in vitro, which is related to the developmental origin of these cells. Indeed, in contrast to mESCs, pluripotent cells derived from mouse epiblast stage embryos (epiblast stem cells, EpiSCs) can be considered as the true developmental counterparts of hESCs. Interestingly, both EpiSCs and hESCs were shown to require the same culture conditions (Brons et al., 2007; Tesar et al., 2007). Human and mouse ESCs also differ in the conditions needed for definitive endoderm (DE) induction, the first step towards commitment into pancreatic and other gastrointestinal fates (Mfopou et al., 2010b). Whereas hESCs cultured as monolayers and stimulated with Activin A and Wnt3a in a basic medium efficiently generate DE progenitors, mESCs cultured under similar conditions usually fail to survive or they generate DE cells with low efficiency (<25%) (D\'Amour et al., 2005; Hansson et al., 2009; Morrison et al., 2008; Sulzbacher et al., 2009; Tada et al., 2005; Yasunaga et al., 2005). On the contrary, mESCs cultured as embryoid bodies generate DE cells in the presence of ActA with an efficiency that can reach 85% if Noggin is also supplemented (Gadue et al., 2006; Kubo et al., 2004; Li et al., 2011). We have previously shown that embryoid bodies do not constitute an optimal environment for efficient differentiation into pancreatic phenotypes (Mfopou et al., 2005, 2007). Interestingly, monolayer cultures are also technically more practical and simple to rapidly analyze microscopically; e.g. they don\'t need to be embedded and sectioned. They are thus optimal for high throughput screening of growth factor and small molecule combinations.