Optogenetic control of iPS cell‐derived neurons in 2D and 3D culture systems using channelrhodopsin‐2 expression driven by the synapsin‐1 and calcium‐calmodulin kinase II promoters

Abstract Development of an optogenetically controllable human neural network model in three‐dimensional (3D) cultures can provide an investigative system that is more physiologically relevant and better able to mimic aspects of human brain function. Light‐sensitive neurons were generated by transducing channelrhodopsin‐2 (ChR2) into human induced pluripotent stem cell (hiPSC) derived neural progenitor cells (Axol) using lentiviruses and cell‐type specific promoters. A mixed population of human iPSC‐derived cortical neurons, astrocytes and progenitor cells were obtained (Axol‐ChR2) upon neural differentiation. Pan‐neuronal promoter synapsin‐1 (SYN1) and excitatory neuron‐specific promoter calcium‐calmodulin kinase II (CaMKII) were used to drive reporter gene expression in order to assess the differentiation status of the targeted cells. Expression of ChR2 and characterisation of subpopulations in differentiated Axol‐ChR2 cells were evaluated using flow cytometry and immunofluorescent staining. These cells were transferred from 2D culture to 3D alginate hydrogel functionalised with arginine‐glycine‐aspartate (RGD) and small molecules (Y‐27632). Improved RGD‐alginate hydrogel was physically characterised and assessed for cell viability to serve as a generic 3D culture system for human pluripotent stem cells (hPSCs) and neuronal cells. Prior to cell encapsulation, neural network activities of Axol‐ChR2 cells and primary neurons were investigated using calcium imaging. Results demonstrate that functional activities were successfully achieved through expression of ChR2‐ by both the CaMKII and SYN1 promoters. The RGD‐alginate hydrogel system supports the growth of differentiated Axol‐ChR2 cells whilst allowing detection of ChR2 expression upon light stimulation. This allows precise and non‐invasive control of human neural networks in 3D.


| INTRODUCTION
The adult brain has very limited regenerative capacity, which is insufficient to prevent the progression of neurodegenerative diseases, such as Alzheimer's and Parkinson's disease or restore neuronal function following damage derived from stroke or traumatic brain injury (Vink & Bullock, 2010). Traditional treatments for neurodegenerative diseases and traumatic brain injury mainly rely on drugs to reduce the rate of degeneration (Kikuchi, Uchikado, Morioka, Murai, & Tanaka, 2012;Van der Schyf, 2011). In order to overcome neurological disorder and deficiency, generation of functional neuronal cells in three-dimension (3D) combined with optogenetic targeting is a cutting-edge strategy, not only aimed for treatment through tissue implantation and neuromodulation but also as an in vitro model for drug screening and disease modelling.  (Cho et al., 2008;Murphy, Laslett, O'Brien, & Cameron, 2017). However, excitatory glutamatergic neurons involved in synaptic functions have not been extensively investigated in tissue engineering and specific-cell targeting. Specific targeting of glutamatergic neurons in the mixed neuronal population is made possible by using an optogenetic approach.
Optogenetics is a novel technology that integrates genetic and optical engineering to achieve both high temporal and spatial precision within neuronal tissues, overcoming many of the limitations of conventional electrical stimulation and pharmacological methods. Optogenetics facilitates targeting and manipulation of specific neuronal sub-populations, such as excitatory and inhibitory neurons. Moreover, the brain tissues of conscious animals can be controlled by the expression of lightsensitive genes such as channelrhodopsin-2 (ChR2), which triggers membrane depolarisation upon illumination with 470 nm wavelength blue light. ChR2 has been applied to neuroscience research for imaging, targeting specific neurons within neural networks, monitoring neural developments, and regulating neural network activity (Steinbeck et al., 2015). In stem cell engineering, optogenetics has been used for tracking the differentiation of stem cells, for functional analysis of embryonic stem cell-derived grafts, as well as for testing the functional integration of induced pluripotent stem cell-derived neurons (Colasante et al., 2015).
Efficient gene delivery and expression of ChR2 in neuronal cells has been demonstrated using lentiviral vectors with neuron-specific promoters (Hioki et al., 2007). Lentiviral vectors with five different types of neuron-specific promoter: synapsin I (SYN1), calcium/calmodulin dependent protein kinase II (CaMKII), tubulin alpha I (Ta1), neuron-specific enolase (NSE), and platelet-derived growth factor-beta chain (PDGF) have been generated by fusing neuron specific elements with the cytomegalovirus enhancer element (Gloster et al., 1994;Lee et al., 2010;Sasahara et al., 1991). Among these promoters, CaMKII is specifically active in glutamatergic neurons, whereas pan-neuronal promoters are active in all neural types. In this study, CaMKII and SYN1 promoters were selected to establish optogenetic control of hPSC-derived neural networks. This led to the downstream induction of ChR2 expression facilitating the subsequent effect on the regulation of neural activity of hPSC-derived neural networks to be compared.
Furthermore, generation of neurons in 3D requires bioactive scaffolds combined with support cells to maintain hPSC viability and neural differentiation, leading to a functional neural network that imitates those in the human brain. It is a challenge to improve the microenvironment of 3D culture for optimal survival of optogenetically engineered cells.
Hydrogels have been reported as efficient cell carriers or growth factor delivery vehicles in brain tissue engineering (Cheng, Chen, Chang, Huang, & Wang, 2013). In addition, their mechanical properties are comparable with those of human brain tissue (Aurand, Wagner, Lanning, & Bjugstad, 2012), being soft and supportive for cell growth. Cell encapsulation in a hydrogel is, therefore, a promising therapeutic approach that can be applied for culture in vitro or transplanted in vivo.
In this study, alginate hydrogel was utilised for 3D culture of optogenetically modified neurons derived from hPSCs. Previous studies have demonstrated that human embryonic stem cells (hESCs) can be propagated and differentiated in 3D culture using alginate microcapsules (Chayosumrit, Tuch, & Sidhu, 2010;Dean, Yulyana, Williams, Sidhu, & Tuch, 2006). Alginate supports the survival of encapsulated cells in high-density cultures through allowing rapid exchange of nutrients, oxygen, and stimuli across the outer layer of the micro-capsule. Encapsulated cells are buffered in vivo from external factors such as antibodies from the host (Serra et al., 2011). Furthermore, it has been shown that the alginate encapsulation promotes cell growth, differentiation, maturation, and protein secretion of various cell types including mesenchymal stem cells, mouse ESCs, and human ESCs (Addae et al., 2012;Serra et al., 2011).
Despite success in the growth support of many cell types, studies utilising alginate encapsulation for hiPSC differentiation and 3D culture are limited (Addae et al., 2012;Chayosumrit et al., 2010).
In order to improve the ability of alginate hydrogel to support cell attachment and differentiation, alginate hydrogel has been functionalised with Arginylglycylaspartic acid (RGD) peptide. The RGD peptide sequence is found within many cell adhesive proteins within the extracellular matrix, acting as an integrin binding sites that mediate cell adhesion and increase cell-matrix interaction (Ruoslahti, 1996). The physical conformation of the RGD loop regulates the affinity of integrin binding, having a direct effect on cell adhesion (Hersel, Dahmen, & Kessler, 2003).
RGD and other peptide ligands such as YIGSR (Tyr-Ile-Gly-Ser-Arg) and IKVAV (Ile-Lys-Val-Ala-Val) have been reported as crucial in neural progenitor cell survival and differentiation (Li et al., 2014;Villard et al., 2006). Improving cell integrin within the alginate hydrogel alone is insufficient for optimal culture of hPSCs. Cell survival within alginate hydrogel, particularly hPSCs, can be further enhanced by inhibiting apoptosis using individual or multiple small molecules such as through the use of Rho-associated kinase inhibitor (ROCKi, Y-27632). The use of this small molecule was found to increase the viability of dissociated hESCs (10 μM) and several cell lines such as glioblastoma (U87-MG), patient-derived glioblastoma xenoline (JX12), primary glioblastoma (SMC448), and non-cancerous astrocytes (Lau, O'Shea, Broberg, Bischof, & Beart, 2011;Tilson et al., 2015;Watanabe et al., 2007). The anti-apoptotic effect of Y-27632 in feeder-free culture, and its effect in pro-expansion of most stem cell types has been successfully demonstrated (Watanabe et al., 2007), however, the incorporation of ROCKi into the alginate hydrogel had not been previously applied.
The aim of this study was to investigate the use of neural subpopulation specific promoters to drive ChR2 expression in hiPSCderived neurons (Axol-ChR2) after being encapsulated in the functionalised alginate hydrogel (RGD + ROCKi). We examined whether these cells survived and matured into active neurons/neural networks that respond to light stimulation, both in 2D and 3D culture. Results demonstrate that cells survive with optical excitability and show neural calcium-flux functionality in both 2D and 3D culture systems.

| Culture and differentiation of human pluripotent stem cells
Human embryonic stem cells, (HUES2 cell line) were obtained from Harvard University, Melton Lab, Massachusetts, United States. The use of HUES2 to conduct human stem cell research was approved by the ethical committee at the University of Oxford. The cells (1 × 10 6 cells/ml in hESCs medium) were co-cultured on a mouse embryonic fibroblast feeder layer and incubated at 37°C without disturbance. Cells were treated with ROCKi, Y-27632 (Calbiochem, California, United States) at 10 μM. Medium was changed daily (replaced with 50% of fresh medium) for 6 days before subculture or transfer to matrigel (feeder free culture) in mTeSR1 medium (Stemcell Technologies Inc., France). The mTeSR1 medium was changed daily until hESC cells reached 80% confluence, cells were then passaged using accutase to create single cell suspensions (Invitrogen, United States). Cells were expanded in culture prior to encapsulation with alginate hydrogel. In contrast, human induced pluripotent stem cellderived neural progenitor cells (hiPSC-NPC; AXOL13 and AXOL15 cell lines), purchased from Axol Bioscience, Cambridge, United Kingdom.
HiPSC-NPC cells (50,000 cells/cm 2 ) were plated onto laminin coated six-well plates (Sigma, United States) in Axol plating-XF medium for 24 hr. Cultures were maintained in Axol neural maintenance-XF medium (up to 50 days) until differentiation in Axol neural differentiation-XF medium for 72 hr to obtain mature neurons.

| Preparation of primary neurons
The dissociated hippocampal cells were gifts from Fabio Biachi (Oxford University), isolated from 14.5-day pregnant mice according to the local established ethical policies. The individual embryos were separated by cutting uterus between embryo locations and kept in cold neural basal media (NB media). An incision was made in embryo head to dissect the brain in half via the sagittal midline. The meninges were removed, and the hippocampus was removed into NB media on ice. Hippocampi in NB media were centrifuged at low speed (500 rcf) for 2 min following by enzymatic digestion with trypsin for 25 min at 37°C and stirred every 5 min to create a homogeneous mixture. Enzymatic activity was then neutralised, and the tissues were triturated using a fine Pasteur pipette coated with fetal bovine serum to obtain a single cell suspension. Cells were centrifuged and resuspended in complete NB media prior plated on PLO/laminin coated dishes for expansion.

| Cell modification using optogenetics
The channelrhodopsin-2 (ChR2) gene was fused to enhanced yellow fluorescent protein (eYFP) and cloned into a lentivirus expression plasmid with cell type-specific promoter (a) pan-neuronal promoter,

| Immunofluorescent staining
Optogenetically engineered hiPSC-NPCs were differentiated to neurons (Axol-ChR2 cells containing SYN1/CAMKII promoter) and fixed with 4% paraformaldehyde in PBS for 30 min at room temperature (RT). The cells were permeabilised and blocked (5% BSA, 0.2% Triton-X100, and 0.1% tween 20 in PBS) for 1 hr and incubated overnight in primary antibody solution at 4°C. Cells were washed three times with wash buffer (0.2% BSA, 0.2% Triton-X100, and 0.1% Tween 20 in PBS) and then blocked with 10% goat serum in wash buffer for 30 min. Secondary antibody solution was added, and cells were incubated at RT for 2 hr. Finally, the cells were washed with washing buffer and stained with 300 μl of 6-diamidino-2-phenylindole (DAPI). Pictures were taken using confocal microscopy (Zeiss-LSM 780, Germany) and analysed with ZEN light 2013 software.

| Cell viability assay
The hESCs (HUES2 derived from feeder-free layers) and hiPSC-NPC cells (Axol) were harvested as single cells using accutase, resuspended in 1.8% alginate and RGD-alginate solutions, prior to extrusion into 102 mM CaCl 2 at a flow rate of 3 ml/min. Beads encapsulated with hESCs or Axol cells were rinsed with PBS twice, transferred to a sixwell dish and cultured in a chemically-defined condition (mTeSR1 medium containing ROCKi was used for hESCs whereas Axol neural maintenance-XF medium containing ROCKi was used for Axol cell culture). The constructs were removed from culture medium ( and results were analysed using Prism5.

| Statistically analyses
The data in this study were analysed using SPSS 10 software (SPSS Inc., Chicago, United States). Student's t test was used for comparisons between two groups. ANOVA followed by Tukey's post hoc test was used for comparisons among the groups. Statistical significance was accepted at p < 0.05, unless otherwise stated. All experiments were performed in triplicate.

| Lentiviruses-mediated expression of ChR2-eYFP in neurons derived from human iPSCs (Axol cells)
The expression level of ChR2-eYFP was higher in Axol cells driven by the SYN1 promoter (6.5%) than the CaMKII promoter (2-5%) at MOI-2 from Days 7-28 after transduction (Figure 1a), indicating that the panneuronal promoter SYN1 was a stronger promoter in these cells than Factors such as neural maturity, stability of ChR2 expression, and network connectivity needed to be taken into consideration, which required the treated cell to mature in the culture for longer periods.

| Optogenetic engineered Axol cells exhibited characteristics of mature neurons
In order to better characterise cellular phenotype, immunofluorescent staining of Axol-ChR2 cells containing different promoters (between passages 8 and 10) was performed. The transgene ChR2-eYFP was expressed in both pSYN1-ChR2-eYFP and pCaMKII-ChR2-eYFP driven Axol cells. The cells were stained strongly positive for the mature neural markers ßIII-tubulin (TuJ1) and glial fibrillary acidic protein (GFAP/S100β; Figure 2). High expression of GABAergic cells

| Concentrations and flow rates regulate the diameter of alginate bead in 3D cell culture
Alginate beads were optimised to produce suitable bead diameter and morphology for 3D culture of hPSCs. Diameters ranging from 700-2,500 μm were obtained from alginate derived from high guluronic acid or G content (UP-MVG; Figure 3a). Bead diameter was adjusted by changing alginate concentrations and flow rates. A significant correlation between the two parameters was identified: The higher the flow rate, the smaller the bead diameter. The bead diameter increased constantly and proportionally to concentration at lower flow rates (2 and 2.5 ml/min). The data indicated that the highest flow rate, set at 3 ml/min in the study, played a critical role in reducing the bead diameter in all concentrations. However, the bead morphology and satellite fraction content were modulated by the concentration. The beads were produced with rounder shapes at higher concentrations than at lower concentrations, whereas the formation of satellites (tails) was only found at 1.2% (Figure 3b).
Small beads with diameters of less than 1,000 μm are favourable, decreasing the diffusion distance to the center of the bead and increasing the surface-to-volume ratio. Use of alginate derived from a 1.8% concentration solution and at a flow rate of 3 ml/min was found to produce beads with an average diameter 800 μm was chosen as optimal to satisfy the quality requirements of bead integrity, size, spherical morphology, and with no presence of satellite fraction.

| Cell viability increased in RGD functionalised alginate
Cells were initially encapsulated within alginate at a density of 2 × 10 6 cells/ml, and the total viable cells were measured by live-dead cell staining along the time course. The growth of HUES2 is very sensitive to the microenvironment and changes within the cell niche (Gattazzo, Urciuolo, & Bonaldo, 2014). Axol cells showed higher cell viability than HUES2 in the RGDalginate system. Cell viability increased and was maintained at between 45 and 50% at Day 7 and Day 14 in the culture (Figure 4b). The viability of Axol cells encapsulated in alginate without RGD was similar to that observed for HUES2 cells, dropping significantly by Day 14 (p < 0.01).

| ChR2-eYFP expression remained when culture in 3D RGD-alginate
Optogenetically modified neurons (Axol-ChR2 cells) were further investigated for their potential to establish a 3D culture network culture model by encapsulating these cells into RGD-alginate beads. Confocal microscopy revealed that expression of eYFP-ChR2 by transduced Axol cells was not affected by the culture environment, or by bead diameter (Figure 5a). The Axol-ChR2 cells were found clustered in spherical aggregates within the RGD-alginate beads, with approximately 50% (p < 0.5) when compared with Pax-6 and Nestin (Figure 5b).

| Neural network-forming capability on 2D and 3D culture
The neural network-forming capability of optically excitable iPSCderived neurons (Axol-ChR2 cells) in 2D cultures and 3D encapsulated in the RGD-alginate hydrogel was evaluated using calcium imaging.  AP. The type of calcium peak and AP is presented in Figure 7c whereas the classification of calcium event is described in Supplementary Figure 3.
Calcium imaging recordings revealed that neural activities of lightstimulated Axol-ChR2 cells in the RGD-alginate hydrogel appeared in mixed and burst calcium waves, whereas non-stimulated cells exhibited slow undefined waves (Figure 6d). Upon stimulation, the number of calcium spikes (single peak and multipeak) increased significantly, driven by the SYN1 and CaMKII promoters (Figure 8 (Rapti et al., 2015). Other research groups have reported that ChR2-ESC-derived neurons displayed strong ChR2expression, mature neuronal morphology, and positive expression of vGlut2 marker (Stroh et al., 2011), and this is in agreement with our findings from the use of lentivirus transduction on ChR2-iPSC-derived neurons (Axol-13 cell line). Other studies have also reported the robust expression of SYN1 promoter in various types of neuronal cells including hPSC-derived neurons (Steinbeck et al., 2015).
Following transduction, human iPSC derived neural progenitor cells were differentiated to distinct neuronal phenotypes with positive expression of neuron-specific tubulin (TuJ1) and astrocytes markers (S100B/GFAP). Mature glutamatergic and GABAergic neuronal subtypes, were observed, indicating the presence of excitatory and inhibitory neurons. Although optogenetic approaches have recently been used for in vivo and in vitro study in neuroscience (Steinbeck et al., 2015), it is novel to apply this strategy to generate an in vitro 3D neural culture model. Furthermore, the 3D culture system developed using modified alginate hydrogels (alginate functionalised with RGD and ROCKi showed potential in supporting cell survival and allowing neural networks to be light-stimulated in 3D culture. Prior to culture with cells, the physical properties of alginate hydrogel (bead size, sphericity and consistency of formation) were characterised. Results revealed that the physical properties of the hydrogel correlate to chemical composition, and specifically to the proportion of guluronic to mannuronic acid residues in alginate. Alginate consisting of a higher guluronic acid and purity (UP-MVG) forms with Calcein-AM (green); dead cells were stained with propidium iodide, PI (red). Images were captured using confocal microscopy (Zeiss-LSM 710) and processed with ZEN light 2011 software. The percentage of surviving cells was determined by the number of green signals divided by the total number of cells (green and red signals) and analysed using a sectional-views in ImageJ software. Significance was tested by ANOVA; ** = p < 0.01; n.s. denoted a non-significant value and error bars represent standard deviation (± SD) stiffer gels and rounder beads, and this enables the physical properties of alginate to be maintained for a longer period of culture. In line published reports, it was found that microspheres produced from highly purified alginate has less morphological imperfections, resulting in the production of more spherical beads with smaller diameters (Kendall, Darrabie, El-Shewy, & Opara, 2004). Although increasing the concentration of alginate correlated directly with an increase in bead size, flow rate was found to exert a significant influence on bead diameter. The bead diameter was optimised for cell encapsulation and culture, to enable rapid nutrient and gas diffusion through the alginate. Using the optimised parameters, the parameters alginate bead diameter (800 μm) falls within the range of 600-1,000 μm, where glucose, ammonia, and vitamin B have been found to rapidly diffuse across the outer layer of the bead (Gautier et al., 2011).
Other bead production parameters such as needle size also contributes to the shape and size of alginate beads. A larger diameter needle (25G, 5/8″) may lead to bead deformation (tailing), which is a very common phenomenon found with the use of viscous alginate solutions (Fundueanu, Nastruzzi, Carpov, Desbrieres, & Rinaudo, 1999). When a smaller diameter needle (30G, 1/2″) was used for extrusion of the beads, the beads with the smallest diameter could be formed with no tailing in combination with use of alginate formulations of high concentration (1.8%) with high extrusion flow rate (3 ml/min).
After encapsulation, HUES2 cells in RGD-modified alginate demonstrated higher viability than those encapsulated in unmodified alginate, as shown by live/dead analyses. This is thought to be because the RGD sequence acts as a site to bind cell-membrane integrin Graphs showing how the maturation of neurons increased during passaging (P6-P10) in neural differentiation medium. The heterogeneous Axol cells were harvested from different passages to determine cell phenotype spread within their cell populations using flow cytometry (N = 3). The cells were stained with progenitor stem cell markers (Pax-6 and Nestin) and neural marker (vGluT1) to assess neuronal maturation over passaging and culture. Unstained cells were used as a control and to set gate regions. ANOVA was used to assess significance. A *p-value below 0.05 was considered statistically significant. Error bars indicate standard deviation (± SD) [Colour figure can be viewed at wileyonlinelibrary.com] receptors, resulting in increased cell attachment and viability. It has previously been shown that bioadhesive sequences, such as RGD and IKVAV, can increase cell adhesion and growth through the integrin-mediated pathway (Li et al., 2014;Villard et al., 2006). One cause of cell death in the functionalised hydrogels may be apoptosis.
Our findings suggest that whilst alginate hydrogel may support the culture of other cell types, alginate alone is unable to provide an optimal microenvironment for the growth of hPSCs (HUES2). In this study, Axol cells survived better than HUES2 in RGD-alginate. Alginate provides a promising option for 3D cell cultures with cell viability at 40-50%. Alginate is a commonly chosen 3D culture substrate, having favorable characteristics and properties including its ability to make hydrogels at physiological conditions, transparency for microscopic evaluation, penetration of light for optical stimulation, gentle dissolution for cell retrieval, pore networks that allows diffusion of nutrient and waste as well as its nonanimal origin (Andersen, Auk-Emblem, & Dornish, 2015).
Interestingly, encapsulated Axol-ChR2 cells in RGD-alginate beads remained in aggregates with minimal neurite extension. Similar results have been observed elsewhere, where the incorporation of the IKVAV peptide sequence did not lead to an increase in neurite length in silk fibroid hydrogel after 1 week of differentiation (Sun et al., 2017). This could be due to structural limitations of the hydrogel, and pore size should be optimised for the migration and neurite outgrowth of cells  were also similar to previous studies reported on murine ESC-and hESC-derived neurons, which display basic neurophysiological activity in culture, such as AP firing and synaptic currents (Heikkila et al., 2009;Johnson, Weick, Pearce, & Zhang, 2007).  An increase in both the number of single peak and multipeak calcium spikes was observed in optogenetically modified primary neurons and Axol cells upon optical stimulation. Although the CaMKII promoter was found to drive less overall ChR2 expression than the SYN1 promoter, CaMKII driven expression was targeted at mature neurons that responded to light stimuli. (a) Calcium waves were analysed from the ROIs of nonstimulated and stimulated cells derived from transduced primary neurons (Neuron-ChR2-SYN1 and Neuron-ChR2-CaMKII) and Axol cells (Axol-ChR2-SYN1 and Axol-ChR2-CaMKII; ROIs = 40, n = 3). A higher number of burst waves and multipeak calcium spikes was found to be driven by cells expressing ChR2 containing CaMKII promoter upon stimulation, indicating the presence of mature neurons. Calcium spikes (single peak and multipeak) were identified from the selected 40 ROIs. Significance was assessed by two-way ANOVA * = p < 0.05; ns denoted a non-significant value and error bars represent standard deviation (± SD). (b) Comparison of single peak and multipeak calcium spikes, respectively in primary neurons and Axol cells (analysed from the same samples and ROIs used in (a)). (c) Examples of different trace categories, distinguished as (1) single peak and multipeak, calcium spikes and (2)

| CONCLUSIONS
In the current study, we show that hiPSC-derived neuronal cells are successfully generated and optogenetically engineered. The expression of ChR2 is efficiently driven by neuron-specific promoters CaMKII and SYN1, indicating the potential to target within the excitatory and inhibitory neural populations. The ChR2-positive neurons are not only viable but also detectable inside a transparent RGD-alginate hydrogel using calcium imaging. In conclusion, the cellbiomaterial construct created in this study combined with the optogenetic approach leads to the establishment of a functional 3D neural network model, which can be used for future drug screening purposes, non-invasive repetitive functionality analysis and neuromodulation, as well as providing new information for neural tissue engineering and stem cell research.

FIGURE 8
Upon light stimulation, an increased number of calcium spikes (single peak and multipeak) was observed in Axol-ChR2 cells driven by SYN1 and CaMKII promoter, indicating functional activity achieved in a 3D neural model using RGD-alginate. The optogenetically modified cells (Axol-ChR2-SYN1 and Axol-ChR2-CaMKII) and unmodified Axol cells were encapsulated in the alginate bead system (RGD-ALG), respectively. The cell constructs were stained with calcium dye and imaged using confocal microscopy (Zeiss-LSM 710). Total of 34 active cell aggregates were selected from the ROIs (N = 3) and stimulated with light before further analysed for the number of calcium spikes. Significance was tested by two-way ANOVA * = p < 0.05; error bars represent standard deviation (± SD)

SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section at the end of the article.