Improved Cryopreservation of Human Induced Pluripotent Stem Cell (iPSC) and iPSC-derived Neurons Using Ice-Recrystallization Inhibitors

Abstract

Human induced pluripotent stem cells (iPSCs) and iPSC-derived neurons (iPSC-Ns) represent a differentiated modality toward developing novel cell-based therapies for regenerative medicine. However, the successful application of iPSC-Ns in cell-replacement therapies relies on effective cryopreservation. In this study, we investigated the role of ice recrystallization inhibitors (IRIs) as novel cryoprotectants for iPSCs and terminally differentiated iPSC-Ns. We found that one class of IRIs, N-aryl-D-aldonamides (specifically 2FA), increased iPSC post-thaw viability and recovery with no adverse effect on iPSC pluripotency. While 2FA supplementation did not significantly improve iPSC-N cell post-thaw viability, we observed that 2FA cryopreserved iPSC-Ns re-established robust neuronal network activity and synaptic function much earlier compared to CS10 cryopreserved controls. The 2FA cryopreserved iPSC-Ns retained expression of key neuronal specific and terminally differentiated markers and displayed functional electrophysiological and neuropharmacological responses following treatment with neuroactive agonists and antagonists. We demonstrate how optimizing cryopreservation media formulations with IRIs represents a promising strategy to improve functional cryopreservation of iPSCs and post-mitotic iPSC-Ns, the latter of which have been challenging to achieve. Developing IRI enabling technologies to support an effective cryopreservation and an efficiently managed cryo-chain is fundamental to support the delivery of successful iPSC-derived therapies to the clinic.

Graphical Abstract

Graphical Abstract.

Significance Statement.

Cryopreservation of terminally differentiated post-mitotic neurons, with high cell viability and functional recovery of electrophysiological properties post-thaw, remains challenging and limits translational applications to the clinic. Novel cryopreservation formulations, such as those incorporating ice recrystallization inhibitors, that improve cell viability, recovery and functionality post-thaw is a promising strategy to bring high quality cell therapy products to patients.

Introduction

Human induced pluripotent stem cells (iPSCs) are well known for their ability to perpetually self-renew and differentiate into specialized cell types from all three germ layers.¹ Due to these unique properties, iPSCs are important cellular products with applications in disease modeling, drug discovery, and development of engineering cell-based therapies for regenerative medicine and cancer therapies.^2,3 More specifically, iPSCs are being explored as an alternate and renewable source of human neurons for the treatment of various neurodegenerative diseases as emerging evidence highlights that iPSC-derived neurons (iPSC-Ns) recapitulate many properties of mature neurons in vivo such as synaptic transmission, generation of action potentials, and formation of spontaneously active neuronal networks.^4-10 However, despite these exciting advances, the successful cryopreservation of iPSCs and iPSC-Ns, while maintaining high viability, recovery, and functionality, remains a challenge for clinical translation of iPSC-derived therapeutics¹¹ and the establishment of reproducible and scalable neuronal cell models for drug screening platforms.

Human iPSCs are known to be highly sensitive to cryoinjury from freezing and thawing procedures, resulting in low viability, poor re-attachment, and expansion rates as well as altered differentiation capacity.^12-15 “Optimized” cryopreservation protocols and cryoprotective agents (CPAs) dramatically reduce cell viability and recovery and introduce selective genetic/epigenetic pressures on the iPSCs enhancing the likelihood of phenotypic variation and/or alterations in potency. In the case of iPSC-Ns, optimization of enhanced cryopreservation strategies is quite important to retain high cell viability, recovery, and functionality given the post-mitotic nature of these terminally differentiated cell types. This is further compounded by the length of time it takes to generate mature and functional iPSC-Ns, requiring anywhere from 60 days to 6 months, depending on the type of neuron and time required to reach maturity.^4-10 This poses a significant challenge for the clinical translation of iPSC-Ns since large-scale manufacturing will necessitate downstream cryopreservation that can retain high viability, differentiation fidelity, and functionality with minimal manipulation post-thaw to deliver high-quality cell products to the patient.

There are main approaches used in cryopreservation: slow freezing and vitrification. Slow freezing (1-2 °C/minute) is a traditional method of cryopreservation that involves gradually reducing the temperature of the cells over an extended period. The process typically includes controlled cooling rates and the use of permeating CPAs, namely DMSO, to mitigate the cellular injuries caused by the freezing process.^16,17 The majority of cellular damage during cryopreservation results from the uncontrolled growth of ice crystals, a process known as ice recrystallization. Unfortunately, conventional CPAs are unable to prevent this type of cryoinjury.^18-22 Vitrification is a cryopreservation method that involves the ultra-fast freezing of cells (from room temperature to −196°C) in less than two seconds resulting in very fast cooling rates of −10 000 °C/minute. During the vitrification process, the formation of ice crystals is prevented by the ultra-fast freezing and dehydration of the cells.²³ This is achieved by using high concentrations of CPAs to increase the viscosity of the solution and hence prevent the formation of ice crystals.²⁴ Although these high CPA concentrations can be toxic for sensitive cells such as stem cells,^14,25 several vitrification methods have been developed for iPSCs wherein high cell survival rates after thawing could be achieved by adherent vitrification, the CryoLogic vitrification method or by using droplet-based vitrification of adherent iPSCs on alginate.^26-28 While vitrification has been successfully used for the cryopreservation of iPSC, it requires specialized technical expertise, is more labor intensive, requires very small sample volumes and high CPA concentrations, which can be challenging particularly when handling large number of iPSCs and or iPSC-Ns samples or scaling up for clinical applications.

Given the complexities of vitrification, conventional cryopreservation protocols for iPSCs and iPSC-Ns employ a slow cooling rate. As such, the rational design of small molecules capable of controlling ice recrystallization and mitigating cellular injury/damage is a promising strategy to improve cryopreservation outcomes. Toward this end, the Ben laboratory has designed, synthesized, and characterized a number of different classes of small-molecule ice recrystallization inhibitors (IRIs).^29-33 One class of IRIs, N-aryl-D-aldonamides, have demonstrated improved cryoprotectant properties enhancing the post-thaw viability, recovery, and functionality of stem cells, such as human hematopoietic stem and progenitor cells (HSCs) derived from human umbilical cord blood (UCB).^34,35 Since human iPSCs are more vulnerable to intracellular ice formation than many other human or animal cell types,¹¹ in this study we chose to assess whether the N-aryl-D-aldonamides IRIs can improve the post-thaw recovery and functional capacity of human iPSCs and iPSC-Ns. Given that application of iPSC-derived cell therapy products will rely on high-quality products, this approach will enable the development of a biomanufacturing paradigm for these unique and therapeutically relevant cell products.

Materials and Methods

IRI Assay and Formulations

Four IRIs from the N-aryl-d-aldonamide class were chosen to be tested in iPSCs and iPSC-Ns for their ability to inhibit ice recrystallization. These include 2-fluorophenyl gluconamide (2FA, IC₅₀ = 4 mM), 4-chlorophenyl gluconamide (4ClA, IC₅₀ = 12 mM), 4-methoxyphenyl gluconamide (PMA, IC₅₀ = 3 mM) and N-(2,6-difluorobenzyl)-D-gluconamide (2,6 DFB, IC₅₀ = 11 mM; Supplementary Fig. S1a). These IRIs were used as additives in commercially available cryomediums, specifically mFreSR (STEMCELL Technologies) for iPSCs and CryoStor10 (CS10, STEMCELL Technologies) for iPSC-Ns. The IRIs were formulated at different concentrations and warmed in a 37 °C water bath until fully dissolved in each respective medium. The IRI-formulated cryo-solutions were then cooled to room temperature and stored at 4 °C until use.

To quantify ice recrystallization inhibition activity and thereby obtaining an IC₅₀ curve (Supplementary Fig. 1b), a modified splat-cooling assay was used, as previously described.²⁹ In brief, a minimum of five concentrations were investigated for each compound, and a 10 µL droplet of IRI solution was frozen at −78 °C by dropping from a height of 2 m onto a polished aluminum block. The wafer was carefully and quickly removed and transferred to a precooled Peltier stage and left to anneal at −6.4 °C for 5 minutes. One image was selected from each wafer for further analysis using ImageJ software. Ice crystals, with well-defined boundaries within the selected image, were circled in the software and the area of each circled ice crystal was calculated. A binning approach based on ice crystal size was implemented to obtain an initial rate (v) of ice recrystallization. Using Excel, the ice crystal areas were then sorted into discrete bins based on size (bin size increases in increments of 0.001 mm²). As ice crystals grew due to recrystallization, their resulting areas moved from bin 1 to higher bins. The proportionate area of each bin was then calculated for each wafer by dividing the sum of areas within a bin by the sum of the areas of all crystals in the image. Rates were determined for each test concentration and normalized based on the rate determined for the PBS control (zero inhibitor concentration). The generated dose-response curve was plotted in GraphPad using the normalized rate constants, v_norm, for each inhibitor concentration and the corresponding log values of the concentration. A value for the half-maximal inhibition concentration (IC₅₀) was obtained from the 4-parameter sigmoidal curve to fit the data.

iPSC Maintenance Culture

All experimental protocols using human iPSCs were performed following the guidelines established and approved by the National Research Council Canada Research Ethics Board. Two human iPSC lines were used in this study. One line is derived from human amniotic fluid (HAF) cells and the other from human brain microvascular endothelial cells (HBMEC, Cell Systems). Both cell lines were reprogrammed using non-integrating oriP/EBNA1 episomal vectors encoding OCT4, SOX2, c-Myc, KLF4, NANOG, and LIN28 and cultured in feeder free conditions, as previously described.³⁶ For routine maintenance of iPSC, HAF-iPSC, and HBMEC-iPSC were cultured on human embryonic stem cell (hESC) qualified Matrigel (Corning) coated plates and fed daily with mTeSR1 (STEMCELL Technologies). The iPSCs were split as clumps at a 1:6 ratio every 2-3 days upon reaching 70%-80% confluence by dissociating with ReLeSR according to manufacturer’s instructions (STEMCELL Technologies). The iPSCs were routinely frozen at approximately 1 × 10⁶ cells per 1 mL cryogenic vial (Thermo Fisher Scientific) using mFreSR (STEMCELL Technologies) as part of maintenance culture every 3-4 days after passaging.

Differentiation of iPSC Toward Mixed Forebrain Neurons (iPSC-Ns)

HMBEC-iPSCs were thawed and cultured in mTeSR1 medium (STEMCELL Technologies) on 6 well Matrigel-coated plates (Corning) for 5 days. Neural progenitors were derived using a directed monolayer SMAD inhibition-mediated differentiation protocol adapted from Chambers et al³⁷ by switching the medium to STEMdiff Neural Induction Medium and STEMDiff Neural Induction Supplement (STEMCELL Technologies) for 14 days, as per manufacturer’s instructions. Neural progenitor cells (iNPCs) were subsequently expanded in STEMDiff Neural Progenitor Medium (STEMCELL Technologies) for 1-2 passages and differentiated into forebrain neurons by transitioning to BrainPhys Neuronal Medium supplemented with SM1 (STEMCELL Technologies), as previously described.¹⁰ Approximately half medium changes were performed every 2 days as routine maintenance of iPSC-Ns. The differentiating neurons (iPCS-Ns) were maintained in BrainPhys Neuronal Medium supplemented with SM1 for a minimum of 3 weeks and beyond, followed by addition of maturation growth factors (GF: 20 ng/mL GDNF, 20 ng/mL BDNF, 250 µM db cAMP, and 200 nM ascorbic acid; all from STEMCELL Technologies) to obtain fully mature iPSC-Ns.

Freezing and Thawing of iPSC and iPSC-Ns

For freezing, iPSCs were dissociated at 70%-80% confluence using ReLeSR (STEMCELL Technologies), an enzyme-free reagent for dissociation and passaging of iPSCs. In brief, ReLeSR was added to the iPSC and incubated at room temperature for approximately 1 minute. ReLeSR were subsequently aspirated and the iPSCs were incubated at 37 °C for an additional 5-7 minutes to generate small cell clumps (50-200 µm), as per manufacturer’s instructions. The iPSCs were resuspended in ice-cold mFreSR (STEMCELL Technologies), a serum-free cryopreservation medium optimized for iPSCs, containing different concentrations of IRIs in 1 mL cryovials (Thermo Fisher Scientific). The iPSC aggregates from one well of a 6 well-plate were frozen in two 1 mL cryovials. mFreSR medium alone was used as a control. iPSC-Ns were dissociated with Accutase (STEMCELL Technologies) at approximately 63 days in neuronal maturation medium into a single-cell suspension and 1.0 × 10⁶ iPSC-Ns/mL were frozen in a control cryo-solution (CryoStor10, CS10) or CS10 supplemented with 2FA at a concentration of 2.5, 5, and 10 mM in 1 mL cryovials (Thermo Fisher Scientific). The iPSC and iPSC-N samples were frozen at −1 °C/minute in 1 mL cryovials using an isopropyl alcohol-based device, according to manufacturer instructions (Mr. Frosty, Thermo Fisher Scientific). After being cooled to −80 °C for 24-48 hours, the frozen cryovials were then transferred and maintained in a liquid nitrogen storage tank.

For thawing, iPSC and iPSC-N cryovials were warmed in a 37 °C water bath for approximately 1-2 minutes. The cell suspension was then transferred to a 15 mL conical tube and mixed with 5 mL of warmed mTeSR1 or basal BrainPhys Neuronal Medium, respectively (both from STEMCELL Technologies). The supernatant solution was removed after a 5 minutes centrifugation at 300 × g. The iPSC and iPSC-N were re-suspended in their respective culture medium, with or without 10 µM Y27632 (STEMCELL Technologies), and plated for post-thaw assessment assays; 10 ɥM Y27632 was supplemented in iPSC and iPSC-N post-thaw cultures for the first 24 hours and subsequently washed out. Of note, supplementation of the IRIs during HAF-iPSC maintenance culture resulted in high cytotoxicity (data not shown). As such, it was key to ensure the IRI was washed out prior to replating the HAF-iPSCs post-thaw.

Flow Cytometry

iPSCs were dissociated with Accutase into a single-cell suspension (STEMCELL Technologies) for approximately 5-7 minutes, harvested, and centrifuged at 300 x g. The cells were blocked with 3% BSA in PBS and incubated with human Anti-TRA-1-81-VIO Live stain conjugated antibody (1:100 dilution, STEMCELL Technologies) in 100 µL of 1% BSA and 2 mM EDTA in PBS for 30 minutes at room temperature. The iPSCs were washed 3 times with 1% BSA in PBS and re-suspended in 500 µL 1% BSA in PBS and assessed using the BD Accuri C6 Plus instrument. Analysis was performed using FlowJo software.

Assessment of Post-Thaw Viability and Recovery

The iPSC and iPSC-Ns were re-suspended in their respective culture medium post-thaw, with or without 10 µM Y27632 (STEMCELL Technologies), and the cells were counted using Trypan blue exclusion assay. Briefly, a sample of the cells was mixed 1:1 in 0.4% trypan blue and counted using a hemocytometer. Cell recovery and viability were calculated, as previously described³⁸:

$${\displaystyle \begin{array}{l}{\displaystyle \begin{array}{lll}& {\displaystyle \mathrm{R}\mathrm{e}\mathrm{c}\mathrm{o}\mathrm{v}\mathrm{e}\mathrm{r}\mathrm{y}(\mathrm{\%})=\frac{\mathrm{C}\mathrm{e}\mathrm{l}\mathrm{l}{\mathrm{s}}_{\mathrm{u}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{e}\mathrm{d}}}{\mathrm{C}\mathrm{e}\mathrm{l}\mathrm{l}{\mathrm{s}}_{\mathrm{f}\mathrm{r}\mathrm{o}\mathrm{z}\mathrm{e}\mathrm{n}}}\times 100}& {\displaystyle \mathrm{V}\mathrm{i}\mathrm{a}\mathrm{b}\mathrm{i}\mathrm{l}\mathrm{i}\mathrm{t}\mathrm{y}(\mathrm{\%})=\frac{\mathrm{C}\mathrm{e}\mathrm{l}\mathrm{l}{\mathrm{s}}_{\mathrm{u}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{e}\mathrm{d}}}{\mathrm{C}\mathrm{e}\mathrm{l}\mathrm{l}{\mathrm{s}}_{\mathrm{u}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{e}\mathrm{d}}+\mathrm{C}\mathrm{e}\mathrm{l}\mathrm{l}{\mathrm{s}}_{\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{i}\mathrm{n}\mathrm{e}\mathrm{d}}}\times 100}\end{array}}\end{array}}$$

Multielectrode Array (MEA) Recordings

MEA systems were obtained from MultiChannel Systems (Reutlingen, Germany), with the headstage accommodating a 60-electrode MEA dish. The thawed iPSC-Ns were plated at a density of 12 × 10³ cells/mL in a 10-15 µL droplet onto 0.005% PLO-coated MEA plates, as previously described.¹⁰ The iPSC-Ns were fed 2 times per week by replacing 50% of the medium with fresh BrainPhys + SM1 + growth factor medium. The plated iPSC-Ns were grown on the MEAs for 19 days before starting recording. To record spontaneous activity in neurons, the conditioned medium was replaced with fresh BrainPhys medium plus SM1 supplement and growth factors. For each recording session, a 10-15 minutes equilibration period was performed after the transfer of the MEA dish to the headstage housed within a 37 °C incubator, followed by a 20 minutes recording. Raw voltage recordings were analyzed offline to yield multi-unit activity using custom software written in MATLAB. Some electrodes were usually inactive due to lack of neuron coverage and were eliminated from analysis as determined using an empirically determined threshold method.³⁹

MEA recordings were obtained using the following settings: unitless amplifier gain (1100), input voltage range (±2048 mV), and acquisition rate (5 kHz). Some electrodes were usually inactive due to lack of neuron coverage and were eliminated from analysis as determined using an empirically determined threshold method.^10,39 Spike and burst detection (MaxInterval program) was performed using the NeuroExplorer (Nex Technologies) software, since Cotterill et al⁴⁰ showed that the method of burst detection compared quite favorably in an unbiased quantitative assessment with 8 other techniques, we adopted the same parameters: the maximum beginning of interspike intervals (ISI), maximum end of ISI, maximum interburst interval, maximum burst duration, and minimum spikes in a burst were set to 0.17, 0.3, 0.2, 0.01 and 3 seconds, respectively. Due to a wide range of spontaneous activity frequencies and bursting between electrodes in an MEA observed in pre-drug recordings, ratios of post-drug:pre-drug, MFRs, and bursting parameters were evaluated on an electrode-by-electrode basis. Electrodes displaying less than 1 burst/minute were eliminated from the analysis of drug effects.

Immunocytochemistry

iPSCs were plated onto 15 mm round coverslips coated with Matrigel (Corning) in mTeSR1 (STEMCELL Technologies) medium. The iPSCs were fixed using 4% PFA (Sigma) and permeabilized with 0.2% Triton X-100 (Sigma) in PBS (without Ca/Mg) for 20 minutes, washed and blocked using DAKO protein block serum free (Agilent) for 20 minutes at room temperature. iPSC-Ns were plated onto 15 mm round coverslips that were coated with Matrigel (Corning) in a neuronal maturation medium. The cells were fixed using 10% formalin (Sigma) and permeabilized with 0.2% Tween-20 for 10 minutes at room temperature. The cells were washed with PBS and blocked with protein block serum-free solution (Agilent) for 1 hour at room temperature. The coverslips were incubated with primary antibodies for 1 hour, then washed 2 times with PBS and incubated with secondary antibodies for 1 hour at room temperature (the antibodies used are summarized in Supplementary Table S1). The coverslips were then washed twice with PBS and mounted in DAKO fluorescent Mounting Medium (agilent) containing 5 µg/mL of Hoechst 33258 (Sigma). Images were captured using the Axiovert 200M microscope (ZEISS). Imaged cells using 20×/ 0.4LD Archroplan Korr (DICII) objective.

Cell Proliferation and Caspase 3/7 Assays Using Incucyte

After thawing, approximately 200 000 iPSCs were plated onto 12 well plates pre-coated with Matrigel in mTeSR1 medium with or without 10 µM Y27632 ROCK inhibitor. After 4 hours, both culture conditions were switched to mTeSR1 (STEMCELL Technologies). To assess iPSC proliferation, the plates were placed into the Incucyte S3 live cell imaging system (Sartorius). Continuous live-cell imaging was used to assess iPSC proliferation over 48 hours. The Incucyte integrated analysis software allows phase object counts and area assessments that enabled confluence assessment in the visual field as a measure of proliferation. To assess Caspase 3/7 activity, 5 µM of Incucyte Caspase-3/7 red dye (Sartorius) was added to the cultures to assess apoptosis. The percentage of Caspase 3/7 positive cells was assessed using continuous live cell-imaging over a 24 hour period. The incucyte integrated analysis software allows phase and red (excitation (ex), 565-605 nm; emission (em), 625-705 nm) fluorescent object counts that enabled real-time evaluation of cell apoptosis post-thaw. For both assays, 9 images were taken from each well and the confluence percentage data, based on phase object confluence, was recorded per time point. The value reported, per time point and treatment condition, was the mean of 9 images per well.

Teratoma Generation

The teratoma studies were approved by the Animal Care Committee at the National Research Council of Canada. Prior to injection, HAF-iPSCs were pre-treated with mTeSR1 with 10 µM Y27632 ROCK inhibitor (STEMCELL Technologies) for 1 hour and were subsequently dissociated into a single-cell suspension using accutase for approximately 5-7 minutes (STEMCELL Technologies). Approximately 1 × 10⁶ iPSCs were gently resuspended in 50 µL hESC qualified Matrigel on ice and injected subcutaneously into the tibialis anterior muscles of immunocompromised SCID mice (Charles River). The mice were monitored for 6-8 weeks post-injection. When the teratomas reached 1 cm³ in size, the mice were euthanized in accordance with the regulations of the National Research Council of Canada Animal Care Committee. The teratomas were explanted, fixed, and sectioned for H&E staining.

Results

2FA Improves Post-thaw Viability of iPSCs

In this study, we assessed whether 4 IRIs from the N-aryl-D-aldonamide class could improve HAF-iPSC viability post-thaw. To test the effect of 4ClA, 2FA, PMA, and 2,6 DFB on iPSC post-thaw viability, HAF-iPSCs were frozen in mFreSR alone or mFreSR supplemented with IRIs. The IRIs were tested at concentrations ranging from 0.5 to 25 mM based on the IRI; the range was chosen based on the IC₅₀ values (Supplementary Fig. S1) and solubility in the respective cryomedium. The HAF-iPSCs vials were frozen in mFreSR or mFreSR + IRI for a period of 1 week prior to thawing and performing cell viability and recovery assays. The immediate post-thaw viability of the iPSCs was assessed using the trypan blue exclusion assay. Although most of the IRIs showed improved post-thaw viability over mFreSR controls (46%), the highest percent viability was observed for 0.5 mM 4ClA (71%), 10 mM 2FA (71.5%), and 15 mM PMA (74.5%) (Fig. 1a). Interestingly, 25 mM PMA resulted in low HAF-iPSC post-thaw viability, highlighting the importance of performing dose-response assays to assess high IRI potency with low cytotoxicity. Furthermore, 2,6 DFB was also cytotoxic to the HAF-iPSCs, with negligible viability and recovery post-thaw, and hence was excluded from the experimental panel. Collectively, supplementation with 0.5 mM 4CLA, 10 mM 2FA, and 15 mM PMA IRIs resulted in an average increase in viability of approximately 20% relative to the mFreSR controls.

Figure 1.

Viability in HAF-iPSCs using mFreSR supplemented with IRIs. (a) Percentage (%) of post-thaw viability of HAF-iPSCs in different cryomedias: mFreSR alone (control) and mFreSR supplemented with different concentrations of 4ClA, 2FA, and PMA (mean + SEM). Statistical significance marked by asterisks assessed by one-way analysis of variance (ANOVA) for comparison to control (mFreSR), where ns = P > .05, *P ≤ .05, **P ≤ .01 and ***P ≤ .001 (n = 3). Dashed line delineates the mFreSR control viability. (b) Representative phase-contrast images of post-thaw HAF-iPSCs cryopreserved in mFreSR supplemented with 10 mM 2FA, 0.5 mM 4ClA, and 10 mM PMA. Phase-contrast images acquired using incucyte live cell imaging showing differences in iPSC cell morphology, density, and colony size at 48 hours post-thaw. Scale bar = 400 µm. (c) HAF-iPSC confluence (mean + SEM) at 48 hours post-thaw obtained by averaging the cytolight red objective counts of 9 images, captured for each well, of triplicate conditions (incucyte). Statistical significance marked by asterisks assessed by one-way analysis of variance (ANOVA) for comparison to control (mFreSR), where ns = P > .05, *P ≤ .05, ***P ≤ .001, and ****P ≤ .0001. (d) Flow cytometry histogram analysis comparing TRA-1-81 expression for HAF-iPSCs immediately post-thaw for all cryomedia formulations (mFreSR, 10 mM 2FA, 0.5 mM 4ClA, and 15 mM PMA). Unstained HAF-iPSC (controls) are shown in red. (e) Percentage (%) of Caspase 3/7 positive cells (mean ± SEM) over the 24 hours post-thaw period obtained by averaging the red objective counts of 9 images captured for each well of the triplicate condition normalized to confluence using incucyte live cell imaging. Statistical significance marked by asterisks assessed by Student’s t-test for comparison to control (mFreSR), where ns = P > .05 and *P ≤ .05 (n = 3).

2FA Improves Post-thaw Recovery

We subsequently assessed HAF-iPSC proliferation as well as any secondary IRI-related cytotoxicity, over a 48 hour period. Vials from each IRI test group were thawed and plated immediately in mTeSR1 maintenance culture supplemented with 10 µM Rho-associated kinase (ROCK) inhibitor (Y-27632, RI).^41-43 Routine supplementation with 10 µM ROCK inhibitor, in post-thaw iPSC cultures, has also been shown to reduce cryopreservation-induced apoptosis and improve the recovery of cryopreserved iPSCs and neural stem cells.^41,42,44,45 Over the 48 hours post-thaw culture period, no significant cytotoxicity was observed for HAF-iPSCs thawed from 10 mM 2FA and 0.5 mM 4ClA, which formed compact colonies with characteristic high nuclear to cytoplasmic ratios, similar to mFreSR control colonies. However, HAF-iPSC thawed from 10 mM PMA had fewer and smaller colonies and decreased proliferation rate, indicative of a potential delayed recovery response (Fig. 1b). The proliferation of attached HAF-iPCS colonies was assessed using an incucyte live-cell imaging system after 48 hours post-plating. Of the IRI tested, 10 mM 2FA frozen HAF-iPSC showed the highest proliferation as assessed by HAF-iPSC colony confluence (65.0%) compared to 0.5 mM 4ClA (41.5%), 15 mM PMA (36.5%), and mFreSR (47.6%) 48 hours post-thaw (Fig. 1c; Supplementary Fig. S2a). No alteration to iPSC pluripotency was observed for any of the IRIs, as assessed by TRA-1-81 expression (Fig. 1d). Based on these collective observations, we chose to focus on 10 mM 2FA as the lead IRI for all subsequent downstream characterization experiments.

2FA Decreases Apoptosis-Induced Cell Death

Cryopreservation-induced apoptosis has been reported for iPSC aggregates and single cells post-thaw.^46,47 To evaluate whether the increased HAF-iPSC viability, recovery, and proliferation was due to decreased post-thaw apoptosis, we used a quantitative real-time Caspase 3/7 live cell imaging assay. To assess Caspase 3/7 activity, a Caspase-3/7 fluorescent dye was added to the HAF-iPSC cultures immediately post-thaw and the percentage of Caspase 3/7 positive cells was assessed using continuous live cell-imaging over a 24 hour period. The HAF-iPSCs frozen in 10 mM 2FA showed a lower percentage of Caspase 3/7 positive cells immediately post-thaw (1.9%) compared to the mFreSR control (4.6%), respectively (Fig. 1e). However, the highest percentage of apoptosis occurred between 8 and 14 days post-thaw, significantly higher for mFreSR (P ≤ .05) compared to 2FA cryopreserved HAF-iPSCs (Fig. 1e). These results suggest that 2FA improved post-thaw recovery of the HAF-iPSCs, in part, by decreasing post-thaw apoptosis. Importantly, Caspase3/7 activity was not observed in non-frozen control cells (data not shown), suggesting that cell processing (dissociation, centrifugation, and re-plating) does not induce significant apoptosis of iPSCs.

2FA Improves Post-thaw Confluence Recovery of iPSCs

Given the increased HAF-iPSC viability, colony size, and proliferation in mFreSR + 2FA conditions, we subsequently assessed the additive and individual effects of 10 mM 2FA and 10 µM ROCK inhibitor on HAF-iPSC post-thaw attachment and proliferation using Incucyte real-time live cell imaging. HAF-iPSCs were frozen in 4 conditions: mFreSR, mFreSR + ROCK inhibitor, mFreSR + 2FA, and mFreSR + 2FA + ROCK inhibitor. ROCK inhibitor in these treatment conditions was added only post-thaw for the first 24 hours. HAF-iPSC attachment and proliferation were assessed immediately post-thaw (0 hours) and after 42 hours; respectively, using percentage phase object confluence. A higher post-thaw attachment was observed for mFreSR + 2FA (58.9%) and mFreSR + 2FA + ROCK inhibitor (58.2%) compared to mFreSR (29.7%) and mFreSR + ROCK inhibitor (32.7%) immediately post-thaw (Fig. 2a). After 42 hours, the HAF-iPSC frozen in mFreSR + 2FA + ROCK inhibitor showed the highest proliferation (84.8%), significantly higher (P ≤ .001) than HAF-iPSC frozen in mFreSR + ROCK inhibitor (58.2%), mFreSR + 2FA (56.8%), and mFreSR controls (34.1%; Fig. 2b and 2c). Interestingly, HAF-iPSC frozen in mFreSR + 2FA showed similar attachment and proliferation rates to that of mFreSR + ROCK inhibitor; whereas HAF-iPSC frozen in mFreSR alone had the lowest post-thaw attachment and proliferation confluence (Fig. 2b–2c, Supplementary Fig. S2a; Supplementary Video S1). In addition, the aggregate size of HAF-iPSCs colonies 6 hours post-thaw was significantly higher (P ≤ .001) in mFreSR + 2FA + ROCK inhibitor (28 688.3 µm²) compared to mFreSR controls (3 659.6 µm²). The iPSC aggregate size was similar for mFreSR + ROCK inhibitor (17 138.5 µm²) and mFreSR + 2FA (15 607.3 µm²; Fig. 2d–2e). These findings suggest that HAF-iPSCs from the mFreSR + 2FA conditions had a similar attachment and recovery profile as mFreSR + ROCK inhibitor without the ROCK inhibitor-induced morphological (Supplementary Fig. S2b) and potential metabolic deviations.⁴⁸ These findings suggest that optimized IRI-based cryoformulations can potentially be used as an alternative to post-thaw ROCK inhibitor treatment. However, it is also important to highlight that when 2FA was used in combination with ROCK inhibitor, it did have an additive effect significantly (P ≤ .0001) increasing post-thaw HAF-iPSC viability and proliferation.

Figure 2.

Proliferation of HAF-iPSCs frozen in mFreSR + 2FA. Percentage (%) confluence of HAF-iPSCs (a) immediately post-thaw (0 hours), (b-c) 42 hours post-thaw in the presence and absence of 10 mM 2FA and/or 10 µM Rock inhibitor (RI; mean ± SEM) assessed by trypan blue exclusion assay and incucyte live cell imaging. Confluence was determined by averaging the phase object confluence of 9 images captured for each well, of triplicate conditions. Statistical significance marked by asterisks assessed by one-way analysis of variance (ANOVA) for comparison to control (mFreSR), where ns = P > .05, ***P ≤ .001 and ****P ≤ .0001 (n = 3). (d) Average iPSC aggregate colony size (µm²; mean + SEM) 6 hours post-thaw analyzed using ImageJ. A total of 20 random HAF-iPSC colonies were measured per treatment. Statistical significance marked by asterisks assessed by one-way analysis of variance (ANOVA) for comparison to control (mFreSR), where ns = P > .05, ***P ≤ .001 and ****P ≤ .0001. (e) Representative phase-contrast images of HAF-iPSC confluence taken at 6, 24, and 48 hours using incucyte live cell imaging. Scale bar = 80 µm. See also Supplementary Video S1.

Long-Term Cryopreservation With 2FA

To ensure no adverse effects of long-term mFreSR + 2FA cryostorage, we assessed HAF-iPSC viability and pluripotency after 6 months storage in the liquid nitrogen tank. The immediate post-thaw viability of the HAF-iPSCs frozen in mFreSR + 2FA was 84% compared to 64% in mFreSR alone (Fig. 3a) with no alterations to pluripotency as assessed by TRA-1-81 (Fig. 3b) and NANOG and SOX2 expression (Fig. 3c). To ensure mFreSR + 2FA cryopreserved HAF-iPSCs retained their tri-lineage differentiation potential, we performed a teratoma assay confirming differentiation toward all 3 germ layers, as shown in Fig. 3d.

Figure 3.

Long-term cryopreservation of HAF-iPSC in mFreSR + 2FA. (a) Post-thaw viability after 6-month cryopreservation of HAF-iPSCs frozen with 10 mM 2FA (mFreSR + 2FA) and mFreSR (mean + SEM) assessed by trypan blue exclusion assay. Statistical significance marked by asterisks assessed by Student’s t-test, where ns = P > .05 and **P ≤ .01 (n = 3). (b). Expression of pluripotency marker TRA-1-81 post-thaw assessed by flow cytometry. (c) Immunofluorescence staining confirming expression of pluripotency markers SOX2 and OCT4 post-long-term cryopreservation in mFreSR + 2FA (2FA) and mFreSR alone (control). Hoechst counter stain. Scale bar = 40 µm. (d) Histological analysis of teratomas from HAF-iPSC frozen in mFreSR + 2FA (2FA) and mFreSR formed in immunodeficient SCID mice. Representative H&E images confirming the presence of all 3 embryonic germ layers are shown by: (1) Ectodermal neural tissues (black arrows indicate neuroepithelial rosettes. (2) Endodermal squamous epithelial cells (black arrow indicates glycogen storage). (3) Mesodermal extramedullary hematopoiesis (black arrow indicates bone marrow cells. (4) Ectodermal pigmented epithelial cells. (5) Endodermal gut-like epithelial cells. (6) Mesodermal bone formation with peripheral calcification. Scale bar = 50 µm.

2FA Improves Recovery of Electrophysiologic Function of iPSC-Ns

A great deal of progress has been made to optimize iPSC differentiation protocols toward neuronal subtypes, with mature functional electrophysiological profiles.^4-10 However, a cryopreservation process for terminally differentiated, post-mitotic iPSC-Ns, particularly optimized for maintaining high viability/recovery and functionality, has not been extensively studied. To address this critical translational gap, we sought to investigate whether 2FA could also improve post-thaw viability, recovery, and functionality of post-mitotic iPSC-derived forebrain neurons. For these studies, we used a second iPSC line (HBMEC-iPSC) to generate terminally differentiated forebrain neurons (iPSC-Ns) and assessed cryopreservation outcomes. In brief, HBMEC-iPSCs were differentiated into a mixed population of excitatory and inhibitory forebrain neurons using the StemDiff Forebrain neuronal differentiation strategy, as previously described.¹⁰ At day 63 of maturation, iPSC-Ns were dissociated and frozen in CS10 cryomedium in the presence or absence of 2.5 mM, 5 mM, or 10 mM 2FA. Since neurite branching is regulated, in part, by the Rho kinase pathway,^49,50 addition of ROCK inhibitor has been shown to increase neurite formation (within 24 hours) without any adverse effects on neuronal maturity or electrophysiological properties of iPSC neurons⁵¹ as well as other neuronal cell types.^52-54 As such, 10 µM ROCK inhibitor was added for the first 24 hours post-thaw in all CS10 + 2FA formulations.

The iPSC-Ns were thawed approximately 1 month after cryopreservation and cell viability and recovery was assessed using the Trypan blue exclusion assay immediately post-thaw. We observed a slight increase in iPSC-N viability for 5 mM 2FA (67.2%) and 2.5 mM 2FA (65.3%) compared to CS10 (54%) and 10 mM 2FA (45%); however, this increase was not statistically significant (Fig. 4b). We subsequently assessed post-thaw recovery and observed that 10 mM (45.0%) and 5 mM 2FA (50.3%) frozen iPSC-Ns showed a statistically significant (P ≤ .05) 1.5-fold increase compared to 2.5 mM 2FA (35.0%) and CS10 (32.2%) controls (Fig. 4c). The iPSC-Ns frozen in CS10 + 2FA and CS10 displayed characteristic neuronal morphology of defined cell bodies and neurite extensions accompanied by the expression of differentiated neuronal markers βIII-TUBULIN, NeuN, Nestin, MAP2, and GAD65 + 67 (GAD; Fig. 4d). The presence of synapses was validated by the expression and localization of synaptic protein synaptotagmin (SYT) and presence of astrocytes was assessed by the expression of GFAP (Fig. 4d), as previously described.¹⁰

Figure 4.

Differentiation and cryopreservation of HBMEC-iPSC-derived forebrain neurons (iPSC-Ns). (a) Schematic diagram of the iPSC-Ns differentiation protocol and timeline (days) for differentiation, cryopreservation, and MEA analysis. DIV: days in vitro (b) Percentage (%) of post-thaw viability and c) recovery (mean + SEM) of iPSC-N in CS10 and CS10 supplemented with different concentrations of 2FA (2.5, 5, and 10 mM) and 10 µM Rock Inhibitor. Statistical significance marked by asterisks assessed by one-way analysis of variance (ANOVA) for comparison to control (mFreSR), where ns = P > .05 and *P ≤ .05 (n = 3). (d) Immunofluorescence staining with neuronal markers MAP-2, NeuN, βIII-tubulin, synaptotagmin (SYT), GAD65 + 67 (GAD), and astrocyte marker GFAP. Hoechst counterstain. Scale bar = 20 µm. Data representative of 2 independent differentiations.

Functional Recovery of iPSC-Ns Post-Thaw

Mature and functional neuronal networks generate spontaneous electrical activity as evidenced via spontaneous action potential spikes and synchronized patterns of action potential bursts. To assess functional recovery post-thaw, we assessed the electrophysiological properties of the cryopreserved neurons using multielectrode arrays (MEAs). The thawed iPSC-Ns were plated on PLO coated MEAs and cultured for 20 days prior to recording to enable neuronal adherence to the MEA surface and facilitate neuronal migration and establishment of synaptic connections, as previously described¹⁰ (Supplementary Fig. 3). After 20 days in vitro, the iPSC-Ns started to generate some sporadic electrical activity in a few electrodes but minor activity overall for all cryomedia conditions. After 27 days, the iPSC-Ns cryopreserved in 2.5 mM and 5 mM 2FA started to show robust synchronous electrical activity, in increasing number of electrodes, with robust bursting behavior developing by 48 days (Fig. 5a). Over subsequent weeks, more electrodes were increasingly recruited and bursting behavior developed. (Fig. 5a). Both the median number of bursts (Fig. 5b) and spikes (Fig. 5d) per electrode increased on a weekly basis with several electrodes consistently displaying robust near-constant activity throughout the lifetime of the culture. These iPSC-Ns maintained high mean firing rates and robust spontaneous and synchronous spiking and bursting activity up to 236 days (Fig. 5b–5d). By contrast, iPSC-Ns frozen in CS10 and 10 mM 2FA showed a delay in establishing robust network activity, with only minor activity observed at 27 and 48 days and robust synchronous activity only developing after 130 days (Fig. 5a). Thereafter, both spontaneous and synchronous spike and burst activity plateaued (Fig. 5d–5c), at overall lower thresholds than was observed for 2.5 and 5 mM 2FA conditions. Of note, the iPSC-Ns frozen in CS10 alone lifted off the MEAs by 204 days, which is why no functional data was recorded beyond this time point (Fig. 5b–Figure 5c). The loss of the iPSC-Ns was also reflected in the gradual decrease of the number of active electrodes (Fig. 5d). In conclusion, 5 mM 2FA showed the most robust post-thaw recovery as well as re-establishment and long-term maintenance of functional electrical firing properties.

Figure 5.

Pharmacological responses of iPSC-Ns in CS10 + 2FA. (a) Representative MEA spike raster plots showing the developmental profile of spontaneous activity of cryopreserved iPSC-Ns in CS10, 10 mM 2FA + CS10, 5 mM 2FA + CS10, and 2.5 mM 2FA + CS10 acquired using NeuroExplorer. The y-axis in each plot represents electrode number (1-59) and the x-axis represents recording time (20 minutes). (b) Median number of burst and (c) spikes, per electrode, from 20-236 days showing gradual increase in MFR over time. The red dashed line refers to day 204, the last day where CS10 MEAs were recorded. A generalized linear model (GLM) analysis suggests that there is a significant difference in the median number of spikes (days ≥ 48) and bursts (days ≥ 48) between the treatments (***P < .001, n = 3-4 MEAs). (d) Scatter plot representing the mean percentage (mean ± SEM) of active electrodes at day 236 acquired using NeuroExplorer. Statistical significance marked by asterisks assessed by one-way analysis of variance (ANOVA) for comparison to control (mFreSR), where ns = P > .05 and ***P ≤ .001 (n = 3). Data representative of 2 independent differentiations.

Functional Pharmacological Responses of 2FA Cryopreserved Neurons

Once spontaneous and synchronous neuronal network activity was established, we subsequently confirmed the presence of functional channels/receptors in the 5 mM cryopreserved iPSC-Ns by assessing changes in the mean firing rates (MFR) following treatment with a panel of neuroactive drugs. One concentration per drug was chosen, typically above known IC₅₀ values (to ensure any effect would be observed) and/or previously shown to alter spiking MFR in studies elsewhere.¹⁰ Evaluations of all neuropharmacological drug recordings for iPSC-Ns were performed at 47 days, once the majority iPSC-Ns achieved maturity and displayed synchronized spiking and burst activity. A panel of 8 neuroactive drugs, including 4-amino-pyridine (K⁺ channel agonist), bicuculline (GABA_A receptor agonist), memantine + NQBX (NMDAR & AMPAR antagonist), nicotine (Ach receptor agonist), phenytoin (Na⁺ channel agonist), verapamil (C-type voltage- of spiking within bursts gated Ca channel antagonist), and TTX (Na⁺ channel antagonist) were assessed. Prior to treatment, a pre-drug recording was performed to acquire the MFR for each active electrode, which was then normalized to 100% to allow relative comparisons of the effect of a drug on the MFR of each electrode. Thus, a value over or under 100% in Table 1 represents an increase or decrease; respectively, in the MFR of each electrode. The neuropharmacological profiling of these excitatory and inhibitory neurotransmitter agonists and antagonists was assessed based on the changes in a number of different bursting parameters, such as burst duration, intervals between bursts, frequency of spiking within bursts, and interspike intervals (ISI) in bursts as quantified in Table 1. For each drug, a change in MFRs was almost always reflected by a change in the number of bursts, which generally inversely correlated with interburst intervals (Table 1). Overall, the MFR and bursting data support the presence of functional excitatory glutamatergic and inhibitory GABAergic receptors as well as functional sodium, potassium, and calcium ion channels, as previously described,¹⁰ validating that cryopreservation with 2FA does not alter the functional activity of key neuronal receptors and channels. Supplementary Fig. 2 depicts the treatment regimen and shows representative raster plots with spontaneous recordings obtained before and following the addition of GABA (10 µM) for all 3 cryomedia conditions. GABA treatment resulted in a marked suppression of spontaneous activity of most electrodes for all 3 neuron cultures.

Discussion

A critical requirement for iPSC/ESCs clinical applications is the supply of a large quantity of cryopreserved iPSC master cell banks.⁵⁵ The quality of these cryopreserved iPSC cell banks impacts post-thaw expansion as well as the efficiency and reproducibility of subsequent lineage-specific differentiation.¹⁴ A robust cryopreservation process in generating iPSC working and master cell banks thus becomes a crucial first step to ensuring high quality and reproducibility of iPSCs and iPSC-derived cell products. However, optimization of iPSC cryopreservation remains largely overlooked with suboptimal post-thaw viability tolerated, due to the capacity of the remaining viable iPSC to expand rapidly. However, multiple passages during routine culture and maintenance induced “aging” of iPSC can lead to reduced cell differentiation potential and genetic alterations necessitating an efficacious cryopreservation regimen to prevent these deleterious effects.⁵⁶ By contrast, for terminally differentiated post-mitotic cells such as neurons, suboptimal cryopreservation impacts not only the number of available cells post-thaw but also the functional recovery of electrophysiological properties.^57-60 In fact, post-thaw functional recovery is critical to the successful translational application of cell-based therapies in the clinic. For instance, multipotent mesenchymal stromal/stem cells (MSC) are known to harbor great therapeutic potential for numerous diseases and have been used extensively in clinical trials for their regenerative properties and ability to promote tissue homeostasis.⁶¹ However, MSCs appear to have compromised immunomodulatory, blood regulatory, and engraftment properties directly post-thaw.^62-64 This may provide a possible explanation for the lack of therapeutic efficacy and failures of early clinical trials. Although it is not clear whether and how quickly MSCs recover their full therapeutic activity post-thaw, these findings highlight the importance of an optimized cryopreservation process for clinical success. As a result, tremendous effort has been focused on identifying the regulatory pathways regulating cellular responses during cryopreservation and the development of small-molecule interventions that can effectively improve the efficiency of cryopreservation.⁵⁵ In this study, we assessed a panel of IRIs as novel CPAs for improving cryopreservation outcomes for iPSC and iPSC-Ns using a conventional isopropyl alcohol-based device (such as Mr. Frosty) that can be easily adapted to routine cryopreservation protocols.

Conventional slow-rate freezing in suspension is the current gold standard for cryopreservation of large quantities of iPSCs and neural derivatives. Nevertheless, despite progress in optimizing cryopreservation formulations, iPSCs/ESCs and terminally differentiation iPSC-Ns remain highly sensitivity to cryoinjury and low survival/recovery rates during cryopreservation.^{13,27,55,65-70} Of the panel of IRIs tested, we found the 2FA supplementation of mFreSR cryomedium increased post-thaw recovery of HAF-iPSC, approximately 20%, compared to mFreSR controls and reduced the time required to reach confluency. This was likely due, in part, to the preserved iPSC colony aggregate size and intact cell-cell interactions facilitating more efficient re-attachment to the ECM. Hence, maintaining intact iPSC aggregates post-thaw improves subsequent recovery and proliferation. A similar outcome was also reported for ESC cryopreserved using ultrafast cooling adherent vitrification.⁶⁹ Similarly, preserving endogenous ECM during cryopreservation has also been shown to improve cell survival by maintaining cell-ECM and cell-cell adhesion.⁷¹

The loss of E-cadherin-dependent cell-cell adhesion during iPSC colony dissociation triggers cell death by Rho/Rho-associated protein kinase (ROCK) and myosin hyperactivation.^41,45,72,73 As a result, ROCK inhibitor is routinely added to post-thaw cultures for the first 24 hours to inhibit cell death^42,43,66 by disrupting the extracellular cues that would normally induce apoptosis and increases the cell-cell interactions through modulating cadherins and gap junctions to enable cell re-aggregation.^74-76 Intriguingly, 2FA had a similar effect to ROCK inhibitor supplementation post-thaw suggesting that inhibiting ice crystallization during cryopreservation might be improving overall iPSC integrity and viability thereby decreasing the dependence on inhibiting apoptotic pathways post-thaw. In fact, ROCK inhibitor is known to stabilize the cellular skeleton which is an intracellular effect. Similarly, IRIs have been shown to be readily internalized in cells and control intracellular ice recrystallization and thereby mitigate cellular damage. It is feasible that controlling intracellular ice recrystallization would have a similar effect on the cytoskeleton but additional studies would be required to verify this. However, ROCK inhibitor also had an additive effect with 2FA significantly improving HAF-iPSC viability and proliferation enabling confluence to be reached much faster than with 2FA or ROCK inhibitor-treated cells alone. These observations suggest that combining IRI with post-thaw stress-induced apoptosis inhibition (ROCK inhibitor) represents a promising strategy for improving iPSC cryopreservation outcomes. However, it has been recently shown that ROCK inhibitor treatment alters the metabolism of iPSCs by reducing glycolysis, glutaminolysis, and the citric acid cycle⁴⁸ in addition to observed phenotypic effects such as actin bundling, disruption of colony formation, and altered differentiation bias after prolong use.⁷⁷ While the mechanisms of iPSC cellular damage during cryopreservation are diverse and complex, improving post-thaw viabilities with 2FA, to similar levels observed with ROCK inhibitor treatment, represents a very promising strategy for developing improved cryopreservation solutions limiting dependence of ROCK inhibitor treatment post-thaw.

Significant progress has been made in recent years to develop neuronal differentiation strategies toward a number of different neuronal subtypes of the central nervous system (CNS). These advances are paving the way toward developing the next generation of iPSC-derived cell-based therapeutic approaches as a differentiated modality toward the treatment of a number of neurodegenerative diseases. In fact, recent evidence has shown that transplantation of fetal neural tissues and human embryonic CNS tissues into lesions of the rat spinal cord has restored axonal regeneration and neuronal functionality at the injury site.^78-87 Grafting human stem cells derived from neural stem cells have been shown to enhance the cognition in Alzheimer’s modeled mice.⁸⁸ ESC-derived neurons have also been used in treating neurodegenerative diseases in mice such as Parkinson’s disease.⁸⁹ The cryopreservation of mature neurons to enable immediate application for cell replacement therapies or disease modeling, without requiring further complex differentiation or maturation processes post-thaw, is critical to advance cell therapy development and the establishment of reproducible and scalable cell models for drug screening platforms. However, few studies have reported protocols for cryopreserving mature, terminally differentiated iPSC derived neurons optimized to maintain high cell viability, recovery, and long-term electrophysiological functionality. The majority of iPSC-derived neuronal cryopreservation strategies have focused on neural precursor/progenitors cells.^27,90-92 Most of these studies have demonstrated neural precursor cell cryopreservation, which still requires an additional 2-3 weeks of further neuronal differentiation and maturation steps.^59,93 Similar to iPSCs, these precursor/progenitor cells are highly proliferative and self-renewable cellular intermediates that can be subsequently differentiated into neurons. More recently, Ting et al demonstrated the successful cryopreservation of mature iPSC-derived motor neuron embryoid bodies that retain high cell viability, motor neuron marker expression, and classical electrophysiological properties post-thaw.⁶⁰ Cryopreserved iPSC-derived dopaminergic neurons have also been shown to exhibit high viability and maintain biochemical and electrophysiological signatures of human midbrain dopamine neurons that were able to rescue Parkinsonian phenotypes in vivo.⁵⁸ The limited evidence of robust cryopreservation protocols for terminally differentiated and post-mitotic neurons, indicates that cryoformulations optimized for these specialized cell types are still in their infancy.

Given the success of 2FA in our HAF-iPSC studies, we sought to assess whether 2FA would have similar positive cryopreservation outcomes on iPSC-Ns when supplemented with CS10 cryomedium. Although we observed a trend toward higher iPSC-N viability immediately post-thaw in 2FA compared to CS10 alone, the differences were not statistically significant. However, we observed that the post-thaw recovery of iPSC-Ns was significantly higher (P ≤ .05) for 5 mM and 10 mM 2FA compared to CS10. This emphasizes the importance of assessing post-thaw recovery, in addition to immediate post-thaw viability, to truly assess the efficacy of cryopreservation.³⁸ Our observations are in agreement with CS10 viability data reported for other iPSC-derived dopaminergic neurons, with viability rates reported at approximately 70%.⁵⁸ However, viability rates of iPSC-derived dopaminergic neurospheres, cryopreserved in CS10, were much lower at 45%,⁵⁹ likely due to the variability in neurosphere size leading to lower penetration of the CPA. Although there was a delay in establishing synchronous and spontaneous activity on MEAs compared to what we routinely observed for non-frozen controls (18 vs 27 days),¹⁰ this delay was much shorter compared to CS10 frozen controls. These findings suggest that 2FA may have aided in preserving the functionality of the iPSC-Ns, as evidenced by increased post-thaw maturation and electrophysiological activity. It further highlights the requirement for long-term functional studies post-thaw as a reliable measure of cell recovery and quality. The 2FA cryopreserved iPSC-Ns retained expression of key neuronal specific and terminally differentiated markers and displayed functional neuropharmacological responses following treatment with a panel of neuroactive agonists and antagonists. These results confirm that 2FA supplementation did not negatively affect the function of these receptors and channels. Most intriguingly, 2FA cryopreserved neurons retained high spiking and bursting activity over long-term cultures (up to 236 days), significantly longer and with a higher mean percent of active electrodes than for CS10 controls. These long-term functional parameters are important considerations toward ensuring high cell quality of post-thaw cells, particularly if they are to be administered to patients. Given the complexity and duration of iPSC-N differentiation strategies, optimizing cryopreservation of terminally mature and functional iPSC-Ns, with no post-thaw manipulation, demonstrates pre-clinical feasibility for clinical applications in a variety of neurodegenerative disorders as well as broader implications of cryopreserved cell-based therapeutics covering a vast array of regenerative medicine applications. However, using terminally differentiated iPSC-Ns for cell replacement therapy may not be desirable for all types of neurons or disease indications. In some cases, neural stem and/or progenitor cells may be more suitable,^58,94-96 as they can be differentiated in vivo into specific neuronal subtypes in response to the microenvironmental cues post-implantation. The choice of cellular intermediates depends on the specific context, disease model, and desired therapeutic outcomes. Future studies focused on assessing in vivo function of 2FA cryopreserved iPSC-Ns in transplantation studies will bring further insights into the post-thaw recovery and functionality of 2FA cryopreserved iPSC-Ns.

The major constituent of both mFreSR and CS10 cryomedia used in these studies is 10% DMSO. Although DMSO is regarded as relatively nontoxic, the clinical use of frozen/thawed cells treated with DMSO can cause many adverse effects and toxic reactions.^20,97-99 Furthermore, DMSO is not only cytotoxic but it also induces differentiation and epigenetic status of iPSC/ESCs when added to the cell culture medium.^97,100-103 Therefore, it remains desirable to develop cryopreservation protocols either with lower concentrations of DMSO or with nontoxic alternatives to DMSO for iPSC and iPSC-derived cell products. While very limited, studies employing non-DMSO cryoprotective agents are showing promising results for iPSCs and iPSC-derived sensory neurons at different stages of differentiation.⁵⁴ IRI technology may represent a differentiated modality in iPSC cryobiology research aiding to develop improved cryo-formulations with lower or DMSO-free concentrations while retaining high post-thaw viability, recovery, and functionality of iPSC and iPSC-derived cell therapy products. While the mechanisms of cellular damage during cryopreservation are diverse, complex, and cell-specific, future studies will focus on validating the 2FA outcomes in additional iPSC lines, alternative neuronal subtypes, and other iPSC-derived cell types of representatives of all 3 germ layers, to assess the versatility of 2FA across multiple cell types and stages of differentiation. Furthermore, future studies will also focus on assessing the scalability of 2FA supplementation based on cell number and volume (cryovials to cryobags) and under different cooling rates to enable future scale-up for clinically relevant applications. Continued research into freeze-thawing of iPSC and iPSC-derived cell products may provide crucial advantages to increasing both the safety and efficacy of cellular therapeutics.

In summary, this study demonstrates that small molecules are extremely potent inhibitors of ice recrystallization in iPSC and iPSC-Ns. Further studies with these compounds in vitro and in vivo will elucidate their effectiveness as cryoprotectants representing an enabling and scalable technology to support an effective cryopreservation strategy to enable the delivery of successful iPSC-derived therapies to the clinic.

Funding

This work was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC Discovery Grant), Canadian Institutes of Health Research (CIHR), GlycoNet and the National Research Council of Canada.

Conflict of Interest

Robert N. Ben is Chief Scientific Officer and cofounder of panTHERa Cryosolutions Inc. All of the other authors declare no potential conflict of interest.

Author Contributions

J.H.: iPSC experimentation, data analysis, manuscript ­writing. S.A.: IRI preparation, iPSC-N differentiation, cryopreservation, MEA plating, and analysis. E.B.: iPSC-N methodology, experimentation, data analysis, writing. A.A: assembly and analysis of MEA data. K.C.: iPSC experimentation, data analysis. M.H.: histochemistry for teratoma assays. J.K.S.: analysis of teratoma histology. J.S.T.: data analysis and interpretation of iPSC-N MEA data. R.B.: provision of IRI, data analysis, supervision. A.J.: conceptualization, data analysis, interpretation, supervision, and manuscript writing.

Data Availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author upon request.

Author Agreement

All authors reviewed the manuscript, agreed with its content, and agreed to its submission.