Polyethyleneimine Facilitates the Growth and Electrophysiological 2 Characterization of iPSC-Derived Motor Neurons

Abstract

Induced Pluripotent Stem Cell (iPSC) technology, in combination with electrophysiological characterization via Multielectrode Array (MEA), has facilitated the utilization of iPSC-derived Motor Neurons (iPSC-MNs) as highly valuable models for underpinning pathogenic mechanisms and developing novel therapeutic interventions for Motor Neuron Diseases (MNDs). However, the challenge of adherence of MNs to MEA plates and the heterogeneity presented in iPSC-derived cultures raise concerns about the reliability of the findings obtained from these cellular models. We discovered that one factor modulating the electrophysiological activity of iPSC-MNs is the Extracellular Matrix (ECM) used in coating to support the in vitro growth, differentiation and maturation of iPSC-MNs. This study here showed that two coating conditions, namely, Poly-L-ornithine/Matrigel (POM) and Polyethyleneimine (PEI) strongly promoted attachment of iPSC-MNs on MEA culture dishes compared to the other three conditions and both facilitated the maturation of iPSC-MNs as characterized by the detection of extensive electrophysiological activities from the MEA plates. POM coating accelerated the maturation of the iPSC-MNs for up to 5 weeks, which facilitates the modeling of neurodevelopmental disorders. However, the application of PEI resulted in more even distribution of the MNs on the culture dish and reduced variability of electrophysiological signals from the iPSC-MNs in 7-week cultures, which permitted the detection of enhanced excitability in iPSC-MNs from patients with Amyotrophic Lateral Sclerosis (ALS). This study provides a comprehensive comparison of five coating conditions and offers POM and PEI as favourable coatings for in vitro modelling of neurodevelopmental and neurodegenerative disorders, respectively. Significant statement: Improvement of motor neuron differentiation protocol and production of highly pure functional Motor Neurons (MNs) are essential for modelling motor neuron diseases, drug screening and cell replacement therapy. However, the coating matrix can significantly affect the in vitro performance of motor neurons. In this study, we provided a comprehensive comparison of five coating conditions and offers POM and PEI as favourable coating for in vitro modelling neurodevelopment and neurodegenerative disorders. This also permitted identification of increased excitability phenotypes in the sporadic ALS iPSCs-derived spinal MNs using both POM and PEI coating. Therefore, this novel protocol and derived functional MNs may find a wide application in disease modelling, drug discovery and cell therapy.

Keywords

iPSC-derived motor neurons; Extracellular matrix; Poly-l-ornithine; Matrigel; Polyethyleneimine; Multielectrode array

Abbreviations

ALS: Amyotrophic Lateral Sclerosis; CHAT: Choline Acetyltransferase; ECM: Extracellular Matrix; iPSC: induced Pluripotent Stem Cell; ISL1: HD protein Insulin gene enhancer 1; MEA: Multi-Electrode Array; MNDs: Motor Neuron Diseases; MNPs: MN Progenitors; MNs: Motor Neurons; MNX1: Motor Neuron pancreas homeobox 1; mRNA: messenger Ribonucleic Acid; NESTIN: Neuroepithelial Stem Cell Protein; NSCs: Neural Stem Cells; OCT4: Octamer- Binding Transcription Factor 3/4; OLIG2: Oligodendrocyte Transcription Factor 2; PEI: Polyethyleneimine; POL: Poly-l-ornithine/laminin; POLF: Poly-lornithine/ laminin/fibronectin; POLFM: Poly-l-ornithine/laminin/fibronectin/Matrigel POM: Poly-l-ornithine/Matrigel; RA: Retinoic Acid; RT-PCR: Reverse Transcription Polymerase Chain Reaction; sALS: sporadic Amyotrophic Lateral Sclerosis; MFR: Mean Fring Rate; SHH: Sonic Hedgehog; SOX2: Sex Determining Region Y-box 2; wMFR: Weighted Mean Fring Rate; WNT: Wingless-Type MMTV integration site

Introduction

Motor Neuron Diseases (MNDs) are a group of incurable diseases including Amyotrophic Lateral Sclerosis (ALS) and spinal muscular atrophy, that are characterized by progressive degeneration of Motor Neurons (MNs) located in the central nervous system. MND research has been hindered for a long time due to the inaccessibility to patients’ MNs. Induced Pluripotent Stem Cell (iPSC) technology and human iPSC-MNs from patients provide an opportunity for developing human in vitro models of MNDs [1].

However, the variability and reproducibility of experimental findings present significant challenges in utilizing iPSCs [2-4]. The variations may result from the heterogeneity of iPSC clones and this may be minimized by standardization of iPSC clones, the use of a statistically feasible number of iPSC clones from multiple donors and the establishment of isogenic iPSC lines [5]. Remarkably, even employing identical cell lines and methodologies, there is a notable variation in experimental outcomes across different laboratories, which is challenging to repeat the findings discovered by one lab to another [6]. But following the same culture protocol, including handling procedure, culture medium, passage method, coating matrix, differentiation method and so on, would be able to minimize intra-laboratory or intra-individual variance. In short, this is crucial to ensure the reproducibility of a manifested phenotype as the underlying pathology of MN degeneration within the same laboratory.

Adherence to the culture dish is fundamental to neuronal culture and functional characterization. Extracellular Matrix (ECM) is widely used as a coating matrix for culture dishes to support neuronal culture in vitro and plays a crucial role in multiple neuronal functions, such as cell adhesion, proliferation, migration, differentiation, maturation and communication. Over the past decades, there has been significant progress in the field of artificial ECM materials. One notable advancement is the utilization of ECM-mimicking substrates, such as Matrigel or Geltrex, which have facilitated reliable and scalable in vitro expansion and differentiation of iPSCs across laboratories worldwide. However, this has not been successfully extended to long-term neuronal cell culture. Additionally, Matrigel and Geltrex are derived from tumour and composed of undefined ECM components and growth factors, making them potentially xenogeneic.

Researchers have started using synthetic ploycatinoic peptides in combination with chemically defined ECM components, such as, poly-l-ornithine/laminin, poly-l-ornithine/laminin/fibronectin and laminin/COLI/COLIV/fibronectin for the in vitro culture of MNs. The use of these defined mixtures of substrates is beneficial for MN culture. However, it is notable that they do not prevent the formation of cell aggregates during neuronal differentiation, as protein- or peptide-based substrates are susceptible to degradation by enzymes that are secreted by cultured cells [7,8]. Furthermore, purification of ECM components might present challenges due to their complex nature and variations may also exist in quality control across different batches. The cost associated with these proteins can also be significant.

To address this particular circumstance, researchers have started to develop non-peptide polymer substrates, that are resistant to cellular enzymatic degradation, such as, cytocompatible polypyrrole, polyethyleneimine, polypropyleneimine, poly(allylguanidine) (and polyelectrolyte multilayers [9-13]. The synthetic polymers can be utilized either alone or in combination with peptide-based substrates to reach an optimal balance between resistance to protein breakdown and the absence of inherent biological activity. Among the synthetic polymers, Polyethyleneimine (PEI) has recently been used for in vitro neuronal studies (but not MNs), especially in coating Microelectrode Array (MEA) plates.

MEA is a valuable platform used to study the electrophysiology of electrogenic cells, such as neurons or cardiomyocytes. It consists of dozens to hundreds of planar electrodes embedded in the base of a culture dish. MEA records extracellular potential changes in electrogenic cells, enabling non-invasive, repetitive and long-term monitoring of electrophysiological activities, particularly neuronal network activities. This platform allows a labour-saving and relatively high-throughput manner to investigate electrophysiological activities. To accurately record electrical signals emitted by neurons, neurons are usually seeded at a very high density ranging from 30,000 to 120,000 cells per mm2 on MEA plates, which ensures that the microelectrodes are fully covered by the cells. It is well recognized that neurons grown in vitro have a tendency to aggregate into clusters as they mature. This phenomenon is particularly pronounced when neurons are seeded at high density, aggravating cell attachment. Therefore, when neurons are cultured on MEA plates, notable cell detachment and loss of cultures may occur, thereby hindering the precise recording of genuine neuronal electrical signals. Thus, it is crucial to identify a coating matrix that facilitates an even distribution and firm attachment of neurons to attain reliable electrophysiological recording of true neuronal signals from the MEA plates. PEI presents a high density of cationic charges under the physiological pH conditions. Yet, unlike poly-lysine and poly-ornithine, PEI does not contain peptide bonds, making it remarkably resistant to proteolysis. In comparison to other commonly used coating substrates, such as poly-lysine, laminin, poly-l-ornithine and fibronectin, PEI alone or in conjunction with laminin results in enhanced cell adherence to the culture dish and a more homogeneous distribution of neurons. However, until now, there have been no reports on its use for the MNs or its potential impact on electrophysiological properties in culture.

In this study, we compared the cell attachment and electrophysiology of iPSC-MNs among five coating conditions and both PEI and POM (poly-l-ornithine/Matrigel) were found to support the differentiation and maturation of MNs. Furthermore, both PEI and POM coatings can be used to detect aberrant electrophysiological phenotypes of iPSC-MNs from individuals with sporadic ALS (sALS) using MEA technology. While considering the reproducibility and stability of investigation into the pathophysiology of MNs or neurons from MND patients, we recommend giving priority to the utilization of PEI, which allows continuous evaluation of spontaneous network activities of neurons as they mature and ‘aged’ in an in vitro environment.

Materials and Methods

Maintenance of iPSCs

The iPSC lines used in this study. It were characterized and described previously. All iPSC lines were maintained on Geltrex™ (A1413302, Gibco)-coated 6-well plates in essential 8™ flex medium (A2858501, Gibco).

MN differentiation

MNs were initially differentiated following a previously published protocol with slight modification for comparison of four coating conditions [14]. In this study, iPSCs at passage 20-30 were dissociated with Accutase (A6964, Sigma) and seeded at 30,000 cells/cm2 on (1:100) Geltrex (A1413302, Gibco)-coated 6-well plates in essential 8 flex medium supplemented with 10 μM Y-27632 (72304, STEMCELL) on day 1. The Neuronal Induction Medium (NIM) consisted of 1:1 DMEM/F12 (BE-12- 719F, Lonza) and Neurobasal medium (21103049, Gibco), 1% P/S (#15140122, Gibco), 0.5X N2 (17502048, Gibco), 0.5X B27 (17504044, Gibco), 0.1 mM ascorbic acid (A4403, Sigma) and 1% GlutaMAX (35050061, Gibco). On day 0 and every other day for the subsequent 6 days, the cells were replenished with NIM medium and 3 μM CHIR99021 (#HY-10182, MCE), 2 μM SB (HY- 10431, MCE) and 2 μM DMH1 (HY-12273, MCE) were freshly added. Cells were subsequently split 1:3 onto Geltrex-coated plates in NIM containing 1 μM CHIR, 2 μM SB, 2 μM DMH1, 0.1 μM RA (HY-14649, MCE) and 0.5 μM Purmorphamine (HY-15108, MCE) for the following 6 days, with the medium changed every two days.

On day 12 of differentiation, patterned cells (defined as MNPs) were accutased into single cells and split 1:3 into 6-well plates pre-treated with anti-adherence rinsing solution (07010, stem cell Technologies) for suspension culture in NIM supplemented with 0.5 μM RA and 0.1 μM Purmorphamine from day 12 until day 18. Then, the suspension cultures were accutased into single cells and replated on culture ware pre-coated with POL, POLF, POLFM or POM in the NIM with 0.5 μM RA, 0.1 μM Purmorphamine, 0.1 μM Compound E (HY-14176, MCE), along with three Neurotrophic Factors (NTFs) of 10 ng/μl BDNF (450-02, PeproTech), 10 ng/μl CTNF (50-13, PeproTech) and 10 ng/μl IGF-1 (450-10, PeproTech) for 10 days. Half of the medium was gently changed every other day to prevent disruption to the neurons until day 28 of differentiation. From day 28 onwards, RA, purmorphamine and compound E were all removed from the medium until the cells were ready for subsequent analysis.

For electrophysiological comparative analysis, MNs were differentiated using our recently published protocol. In short, cells, after 12 days of differentiation with Du’s protocol, were dissociated and replated for monolayer culture onto culture ware pre-coated with POM or PEI in the NIM supplemented with 0.5 μM RA, 0.1 μM Purmorphamine and 0.1 μM compound E, along with three NTFs to facilitate differentiation and maturation of iPSC-MNs. On day 18 of differentiation, RA, Purmorphamine and compound E were removed from the medium and half of the medium was carefully changed every other day until day 48.

Coating matrix preparation

The detailed information of coating reagents is listed in Table 1.

Table 1. ECM coating conditions used in this study.

POL: 100 μg/ml Poly-L-ornithine (P4957, Sigma) was diluted to 20 μg/ml in Dulbecco’s Phosphate Buffered Saline (DPBS, D8662, Sigma). The culture ware was coated with 20 μg/ml poly-L-ornithine at 4°C overnight. Laminin at 1 mg/ml (L2020, Sigma) was thawed at 4°C overnight and diluted to 20 μg/ml in DPBS. On the second day of coating, Poly-L-ornithine was removed and the plates were rinsed with sterile water twice and then coated with 20 μg/ml of Laminin at 37°C for 2 hrs. The coating matrix was removed and rinsed once with PBS before plating cells.

POLF: The dishes were first coasted with poly-l-ornithine as described above. The 1 mg/ml laminin (L2020, Sigma) and 1 mg/ml Fibronectin (F1141, Sigma) stocks were pre-thawed at 4°C and diluted with DPBS to contain 20 μg/ml laminin and 10 μg/ml fibronectin. Then poly- ornithine pre-coated plates were coated with laminin/fibronectin solution at 37°C for 2 hours.

The coating matrix was removed and rinsed once with PBS before plating cells.

POLFM: After poly-l-ornithine, laminin and fibronectin coating as described above, Matrigel was thawed at 4°C and diluted 20x in KnockOut™ DMEM (10829018, Gibco) and then added to the same wells. The plates were kept at 37°C for 2 hrs and Matrigel was removed before use.

POM: After overnight coating of poly-l-ornithine, the same wells were coated with Matrigel (1:50) at 37°C for 2 hrs and Matrigel was removed before use.

0.1% PEI solution: 50% PEI (P3143, Sigma) was diluted to 10% with sterile water as recommended by FujiFilm cellular dynamics, Inc. (https://www.fujifilmcdi.com/icell-motor-neurons-01279-gmnc01279). The 0.1% PEI working solution was freshly made by diluting 10% PEI at a 1: 100 ratio with 1x borate buffer, which was made by diluting 20x borate buffer (28341, Thermo Scientific) with sterile water. The 0.1% PEI was filtered with a 0.22μM sterile filter (A16534K, Lennox) before use.

MEA recording and data analyses

MEA recording was performed as described in a recent publication.

Statistical analysis

Statistical analyses were performed with GraphPad Prism version 9.3.1 using Mann-Whitney test with a ^*p<0.05, ^**p<0.01, ^***p<0.001. N represents the total number of cell lines; n is the number of independent experiments or the MEA wells used. Data are presented as the mean ± SEM.

Results

CHAT+ spinal MNs are derived from iPSCs in 28 days by following a published protocol

iPSCs were first differentiated into spinal MNs using a previously established protocol. The differentiation process was artificially divided into four stages: iPSC neutralization, MN patterning (ventralization and caudalization), MN induction and MN maturation (Figure 1). The morphology of differentiated cells was monitored throughout the differentiation until day 28. To induce MNs, cells were adherently cultured for 12 days and then in suspension for another 6 days. On day 18, the cells were dissociated into single cells, replated onto poly-l-ornithine/laminin (POL)-coated plates and treated with 0.1 μM compound E to promote the differentiation and maturation of spinal MNs. After replating, iPSC-derived cells began to project neurites, displaying neuronal identity. However, cells progressively formed long neurite bindles and cell bodies aggregated into large clusters by day 28 of differentiation. The identity of differentiated cells was verified at each differentiation stage by quantitative RT-PCR and immunocytochemistry for the expression of specific markers. The iPSCs, confirmed by high expression of the pluripotent marker OCT4/SOX2 (Figure 1C), were induced to homogenous Neural Stem Cells (NSCs) after 6 days of treatment with the WNT activator CHIR99021, dual-SMAD inhibitor SB431542 and DMH1. Approximately 98.7% ± 0.2% of the cells were positive for the NSC marker of PAX6, 96.5 ± 0.3% of the cells expressed NESTIN (Figure 1C), with suppressed OCT4 expression. Subsequently, the NSCs were directed into MN Progenitors (MNPs) after another 6-day treatment of Retinoic Acid (RA) and purmorphamine (an agonist of SHH) in addition to SMAD inhibition and WNT activation (Figure 1A) and 95.8 ± 0.5% of the cells expressed the MNP marker of OLIG2 (Figure 1C and 1D). The OLIG2+ MNPs were cultured in suspension with RA and purmorphamine for 6 days (Figure 1A), resulting in the generation of 73.8 ± 2.3% ISL+ and 49.3 ± 1.7% MNX1+ MNs by day 18 of differentiation (Figure 1C and 1D). Ultimately, Compound E was applied to the cultures to promote MN maturation. However, only 45.5 ± 4.2% of cells were found to express the mature motor neuron marker of CHAT after a maturation period of 10 days. The expression of specific markers at each stage was also validated at the mRNA level using RT-PCR analysis. The proportion of mature MNs was substantially less than previously reported, however, the low differentiation efficacy was reproduced in three iPSC lines.

Figure 1. CHAT+ spinal MNs can be induced from iPSCs in 28 days using a previously published protocol. Note: (A) Schematic diagram of the 28-day MN differentiation protocol established by Du; (B) Representative morphology of cells during the course of differentiation from day 0 to day 28; (C) Representative staining of cells at each differentiation stage with stage-specific markers. OCT4/SOX2 staining for iPSCs on day 0, PAX6 and NESTIN staining for NSCs on day 6, OLIG2 staining for MNPs on day 12, ISL1 and MNX1 staining for early MNs on day 18 and CHAT staining for mature MNs on day 28 of differentiation; (D) The proportion of PAX6+ and NESTIN+ cells on day 6, OLIG2+ cells on day 12, ISL1+ and MNX1+ cells on day 18 and CHAT+ cells on day 28 of differentiation. N=3 cell lines, n=3 replicates; (E) RT-PCR data showing the mRNA expression of PAX6 on day 6 (left up), OLIG2 on day 12 (right up), ISL1 and MNX1 on day 18 (left bottom) and CHAT on day 28 (right bottom) of differentiation compared to iPSCs on day 0. GAPDH was used as a reference gene.

Figure

Optimal adherence and morphology of iPSC-MNs under poly-l-ornithine/matrigel coating for over 4 weeks

MNs are non-adherent cells and require an ECM for their attachment and growth in vitro. Our initial culture experiments showed that the commonly used coating condition of poly-l-ornithine/laminin (POL) could not support the long-term adherence of iPSC-MNs and this was also reported previously. The cells showed the tendency to aggregate into large clusters (started from day 24 of differentiton) and eventually detached (after medium change on day 30 of differentiation) from the culture dishes prior to downstream analyses (Figure 2A Lane 1), which is consistant with previous studies. Three multiple peptide substrates, namely POLF (POL+Fibronectin), POLFM (POLF+Matrigel) and POM (Poly-L-ornithine/Matrigel) were used in previous MN cultures (Table 1) [15].

We therefore conducted a comparative analysis of four coating conditions (POL, POLF, POLFM and POM) to assess which condition could best promote firm attachment and even distribution of iPSC-MNs.

On day 18 of differentiation, suspension cultures were dissociated into single cells and replated onto a 96-well plate, which was pre-coated with four different coating conditions and cell attachment was closely observed throughout the differentiation (Figure 2). From day 24 onwards, cells on POL- and POLF-coated wells started to cluster into small cell aggregates, whereas cells cultured on POLFM and POM-coated wells continued to display monolayer growth. During the subsequent culture, the aggregates of cells on POL- and POLF-coatings progressively increased in size, resulting in loose adhesion to the culture dish. Consequently, the aggregates could easily be detached from the dish during medium replenishment (Around day 30 of differentiation). Although cell aggregates were also observed on wells coated with POLFM or POM, they were substantially smaller in size than those grown on POL and POLF-coated wells. The size of cell aggregates on each coating condition was quantified on day 28 of differentiation. In addition, iPSC-MNs exhibited a significant increase in the formation of neurite networks on POLFM or POM coating conditions. These findings suggest that neurons developed a higher maturity under these two coating conditions.

In summary, the cell attachment assay indicates that the POM coating is the optimal choice among the four tested conditions that improved the attachment and maturation of iPSC-MNs over POL or POLF coating. Additionally, the POM coating also offers an easier/cheaper alternative to the POLFM method which contains additional fibronectin and Matrigel (Table 1). Consequently, the POM coating was employed for the subsequent electrophysiological research.

Figure 2. POM coating condition showed optimal adherence, neurite development and smallest aggregates of iPSC-MNs. Note: A) Representative morphology of the cells grown on dishes pre-coated with POL, POLF, POLFM or POM (from left to right) and images were taken on day 18, 24, 28 and 30 (from top to bottom); Scale bar 100 μm. B) The quantification of aggregate size among four coatings. The size of aggregates was measured on day 28 of differentiation; N=3 biological replicates.

PEI coating promotes firmer attachment and more homogeneous distribution of iPSC-MNs on MEA plates compared to POM coating in 7-week culture

MEA plates were coated with POM and whether it supported long-term culture by MN adherence and functional maturation of iPSC-MNs was investigated by electrophysiological characterization. As noted earlier, small clusters started to appear from day 30 on the POM-coated MEA plates and large MN aggregates formed by day 48 (Figure 3). In searching for more optimal coating substrates for MN aging on MEA plates, PEI, a non-peptide polymer, was found employed in an early study for neuronal culture due to its ability to enhance neuronal adhesion and minimize cell aggregation on culture dishes on MEA plates. However, there has been no report documenting its utilization for in vitro MN culture. Therefore, PEI was tested in this study to compare with POM for the MN growth on MEA plates.

As sMN differentiation efficiency was found to be substantially lower than that in the previous report, we recently developed a novel monolayer sMN differentiation protocol to improve MN production. The addition of a NOTCH pathway inhibitor, Compound E, was advanced to day 12 (instead of from day 18) for a duration of 6 days and this was found to timely promote the conversion of MNPs to MNs, resulting in the efficient generation of 91.2% ± 7.0% of MNs expressing CHAT.

We therefore switched to this monolayer differentiation protocol to generate MNs for the subsequent electrophysiological investigation. MNPs were dissociated on day 12 and then reseeded at a cell density of 50,000 cells/well onto 48-well MEA plates coated with either POM or PEI. Subsequently, neuronal attachment on the MEA plates was closely monitored throughout the culture process. As anticipated, MNs on POM-coated MEA plates displayed small clusters by day 30 of differentiation, 18 days after replating cells and this was consistent with the cells cultured on POM-coated conventional culture dishes. However, in the subsequent 18 days, they grew into large aggregates by day 48. On the other hand, the MNs cultured in the PEI-coated wells did not display obvious cell clusters on day 30 and no large aggregates appeared by day 48 of differentiation. Taken together, the PEI coating was shown to be more suitable to enhance firm adhesion in a long-term and homogeneous distribution of iPSC-MNs among different coating conditions tested and no major cell clustering issue occurred in over an extended period of 36 days of culture on MEA plates, even with a high seeding density of 50,000 cells on a 1.1 × 1.1 mm recording area of 48-well MEA plate.

Figure 3. PEI coating condition promoted even distribution of iPSC-MNs on MEA plates.

Representative morphology of iPSC-MN cultures on day 18, day 30 and day 48 from 48-well MEA plates coated with POM (A) or PEI (B). The top panels were taken under 4x magnification and the bottom panels were taken from the respective well on the top but under 5x magnification. Scale bar, 100 μM. N=3 biological replicates.

Electrophysiological signals of iPSC-MNs are influenced by coating conditions

Electrophysiological characteristics are essential for evaluating the functionality of mature neurons, but little is known about whether they could be influenced by coating conditions. To compare the electrical performance of hiPSC-MNs on the POM and PEI-coating conditions, the firing properties of the MNs were recorded every other day from day 18 to day 48 of differentiation on 48-well MEA apparatus. Distinct firing patterns of spontaneous activities were observed under the two different coating conditions, as the iPSC-MNs progressively matured on the MEA plates. In comparison to the PEI-coating, the iPSC-MNs on the POM-coated MEA wells exhibited accelerated development of spontaneous firing activities at the early stage of differentiation (Figure 4). The number of active electrodes and Mean Firing Rate (MFR) from iPSC-MNs on the POMcoating increased sharply from day 22 and peaked at day 30 and up to day 34 of differentiation, they were significantly greater than those cells grown on the PEI-coated wells. In addition, the cells on the POM-coated wells showed stronger network activities by day 30 and higher synchronization by day 34 in comparison to those grown on the PEI-coated wells, indicating a greater functional connectivity among neurons. After peaks of the neuronal activities between day 30 and 34 of differentiation, however, there was a subsequent decline from day 32 in the electrical activities from the POM-coated wells and this coincided with progressive formation of large clusters of MNs during the in vitro “aging” and aligned with previous findings on conventional culture ware. On the PEI-coated wells, a uniform distribution of cells was observed throughout the entire culture duration on the PEI-coated wells up to the end of recording by day 48 (Figures 3B). The iPSC-MNs exhibited a slower development of functional firing at the early stage of differentiation compared to cells on the POM-coating, but a rapid increase of the firing activities appeared from day 34. The electrical activities on the PEI-coated wells continued to rise and were significantly higher than those on the POM-coated wells from day 40 to day 48 Figures 4A-4D). This firing pattern was repeatedly shown among the different iPSC lines. These findings indicate that different coating conditions have the potential to influence the maturation speed of iPSC-MNs in an in vitro setting. Furthermore, coating conditions can also affect the stability of electrical signals detected by the MEA.

Figure 4. The iPSC-MNs showed different electrophysiological characteristics under the POM and PEI coating conditions.

(A-D) plots of MEA recording data for the number of active electrodes (A), Mean firing rate (Hz) (B), Network burst frequency (C) and synchrony index (D) from the longitudinal recording of day 18 to day 48 of iPSC-MNs on the POM or PEI-coating conditions. (E- H) Roaster plot of MEA recordings showing representative changes in spike firing pattern on day 30 E and F) or day 48 (G and H) on POM-(E and G) or PEI- (F and H) coated 48-well MEA plates. The total recorded wells were n=24 in the POM-coating condition and n=51 in the PEI coating condition from three independent cell lines (N=3). The data are presented as the mean ± SEM.

ALS iPSC-MNs display spontaneous hyper excitability

Hanges in the excitability of MNs were previously reported in both in vivo and in vitro models of ALS, such as in ALS patients and in MNs derived from patients with familial ALS who carried known mutations, as well as in transgenic animal models of ALS. This prompted us to investigate whether these coating conditions equally impacted the electrophysiological properties of iPSC-MNs from sALS patients and controls. Both groups of MNs were cultured in parallel on the coating substrates of POM or PEI (Figure Increased excitability was consistently observed from sALS iPSC-MNs under the PEI-coating condition, when compared to the MNs derived from control iPSCs, which was evidenced by the significantly higher values of the number of active electrodes, MFR and network burst frequency. On the POM-coated wells, sALS iPSC-MNs also showed a similar trend of higher excitability compared to the control iPSC-MNs, but not as significantly as the signals on the PEI coating (Figure 5). This discrepancy might be attributed to the formation of neuronal aggregates and subsequent loose contact of the cells on the POM-coated wells (Figure 3), which led to greater variations compared to cells cultured on the PEI-coated wells. To minimize the impact of inactivated electrodes resulting from neuron detachment, we compared the weighted MFR (wMFR), which exclusively considered the firing rate of active electrodes at each time point and discarded those that were inactive. The value of wMFR was found to be higher than that of MFR when inactive electrodes were omitted. However, this did not affect the hyperexcitability phenotype of sALS iPSC-MNs, regardless of the coating conditions used (Figures 5G and 5H). The raw data of the control and patient groups are included. The enhanced excitability of sALS iPSC-MNs was consistently reproduced in subsequent experiments using a larger sample scale under the PEI coating, which was recently reported. In conclusion, altered excitability was effectively captured by sALS iPSC-MNs under PEI-coating condition, in addition to consistent firing qualities and enhanced synchronization of neurons. Therefore, PEI is a more suitable choice for investigating the electrophysiological activities of cells on MEA plates, particularly for long-term studies of neurodegenerative disorders.

Figure 5. PEI and POM coating enabled detection of hyperexcitability of iPSC-MNs from sALS patients.

Plots of MEA recording data for the number of active electrodes (A and B), mean firing rate (Hz) (C and D), network burst frequency (E and F) and weighted mean firing rate (G and H) from the longitudinal recording of day 18 to day 48 of iPSC-MNs on the PEI or POM-coating conditions. The total recorded wells were n=51 for the PEI coating condition and n=24 for the POM-coating condition from three independent cell lines (N=2 sALS iPSC lines and 1 control iPSC line; n=2 technical replicates). The data are presented as the mean ± SEM.

Discussion

MEA has been extensively employed in research on cardiac disease and neurodevelopmental/neurodegenerative disorders due to its non-invasiveness, labour-saving and relatively high readouts. These attributes make MEA suitable for investigating disease progression and conducting in vitro drug testing. In combination with iPSC-MN models, MEA enabled the recapitulation of abnormal electrophysiological activities from MND patients, which led to the discovery of novel candidate drugs such as retigabine [16-19]. However, the clustering of iPSC-MNs takes place on MEA plates during the long-term culture in vitro, which not only creates a significant challenge for investigating age-related phenotypes, but also potentially hinders the reproducibility of pre-symptomatic data and limits screening of novel drugs for the treatment of MNDs.

In this study, a comparative analysis was conducted to examine five coating conditions for their impact on the attachment and electrophysiological performance of iPSC-MNs. The initial step involved comparison of POL, POLF, POLFM and POM on traditional culture ware for up to 30 days of culture and the iPSC-MNs exhibited better distribution on the surfaces of POLFM and POM. Considering that POM coating, which contains poly-l-ornithine/Matrigel but lacks laminin and fibronectin, yields a similar outcome to POLFM with laminin and fibronectin, POM coating is recommended for MN culture on conventional or MEA dishes for up to 5 weeks. In addition, POM appeared to accelerate neuronal maturation in the early phase of MN differentiation, strongly and linearly increased neuronal activities were detected in iPSC-MNs from day 24 to day 30 (Figure 3). Therefore, POM coating can be adopted to investigate stem cell models of neurodevelopmental disorders.

However, large aggregates did form in the prolonged culture on the POM-coating condition, which led to reduced contacts with microelectrodes and decreased MEA signals detected during the subsequent 2 weeks of MN culture. This is likely due to the degradation of the peptide matrix of poly-l-ornithine and Matrigel by the enzyme secreted from the cells.

In terms of neurodegenerative dieases, it is more desirable to use “aging” cell models. It has been reported that reprogramming iPSCs revert some key hallmarks of cellular age, such as epigenetic age. However, a more stable culture condition that can extend the in vitro culture time of iPSC-derived neurons would be partially benefit and facilitate the “aging” of the cultures and the recapitulation of aging-related phenotypes. Therefore, an optimal ECM or co-culture with astrocytes or other coating methods were developed to achieve more homogeneous cell distribution in the longer term, which would benefit the utilization of MEA for electrophysiological phenotyping for neurodegenerative diseases [20,21]. In this respect, we compared the POM and PEI coating conditions and observed that the PEI exhibited greater suitability for MND modeling. There was no significant formation of large aggregates over the course of 7 weeks of MN cultures in the PEI wells and stable neuronal electrical activities consistently increased until the end of 48 days of differentiation. Notably, the cells demonstrated greater synchronization on the PEI-coated wells than on POM-coated wells. Electrophysiological characteristics of iPSC-MNs showed that the PEI-coating condition enabled the capture of increased spontaneous excitability of MNs derived from sALS iPSC lines, which was consistent with previous findings in familial ALS iPSC-MN models.

Therefore, we recommend PEI-coating for neurodegenerative research.

Conclusion

However, it is worth mentioning that there was a delay in detecting electrical signals from the PEI-coated wells during the initial phase and the strength of the MEA signals was also reduced in comparison to the POM-coated wells, which may be due to the PEI is lack of bioactivity and may induce toxic effects. Therefore, the apoptosis and senescence of the neurons on PEI coating would be measured to address this question. Another limitation is the percentage of functional mature neurons out of all the cells has not been determined on PEI coating in this study as the cells are still formed aggregates. Above all, the coating condition for iPSC-MN culture and the properties of iPSC-MN on different coating conditions need be further optimized and comprehensively analysed.

Recently, a new non-peptide polymer, known as dendritic polyglycerol, an amine-based substrate, in combination with Matrigel, was reported to significantly improve the long-term culture of iPSC-MNs. This enhanced culture system allowed sustained investigation of various aspects, including cell viability, molecular characteristics, spontaneous network electrophysiological activity and single-cell RNA sequencing of iPSC-derived mature MNs for up to two months, where the current study was limited to 7 weeks of culture. Matrigel, being a composite of ECM components, is deemed suboptimal for in vitro investigations. Laminin, an essential component of Matrigel, has been extensively employed in neuronal research in combination with other synthetic peptide, such as poly-l-lysine, poly-D-lysine and poly-l-ornithine. A potential resolution to accelerate maturation of iPSC-MNs might be achieved using PEI coating in combination with laminin or other synthetic peptide, which could be examined in a forthcoming investigation.

In summary, this study demonstrated the beneficial effects of the PEI coating for the long- term investigation of iPSC-MNs, leading to a more uniform distribution, the feasibility of investigating mature iPSC-MNs and the stable acquisition of electrophysiological activities. Consequently, improved long-term culture of iPSC-MNs will contribute to the investigation of neurodegenerative diseases and support the early-stage of drug discovery efforts for MND diseases.

Acknowledgement

We acknowledge the volunteers for participating in this study.

Funding

This research was supported by the SFI Investigator award (13/IA/1787) and Centre Grant Number 16/RC/3948 (which was co-funded under the European Regional Development Fund and by FutureNeuro industry partners), Galway University Foundation, Confucius Institute of Chinese and Regenerative Medicine at University of Galway. This research was supported by the HRB-Clinical Research Facility Galway, a unit of University of Galway and Saolta University Health Care Group and with scientific and technical assistance of the NCBES Genomics Facility and the Centre for Microscopy & Imaging, which are funded by University of Galway and the Irish Government's Programme for Research in Third Level Institutions, Cycles 4/5, National Development Plan 2007-2013.

Availability of Data and Materials

The dataset supporting the conclusions of this article are included within the article.

Author Contributions

Conceptualization: S.S., O.H., T.O. M.Y. and D.C.H.

Experimental investigation: S.S., M.Y., M.L.

Data analysis: M.Y., M.L. D.Y., G.Y., Y.L. and G.L.

Original draft: M.Y. and M.L.

Revision: S.S., O.H., T.O., D.C.H., G.Y., Y.L. G.L.

Ethics Declaration

All procedures and protocols have been reviewed and approved by the Galway Clinical Research Ethics Committee. Title: "Making Stem Cells from Somatic Tissues (iPS Study)," Approval number: C.A.750, Date of approval: 12 March 2020. Patients recruited in this study as skin biopsy sample donors provided written consent for use of tissue, cell reprogramming and differentiation and result disclosure.

Competing Interests

The authors declare that they have no competing interests.