An illustration of our previous method to generate fetal iMuSCs is shown in Supplementary Figure S1A (Zhao et al., 2020). To isolate these iMuSCs, hiPSCs carrying a PAX7 (satellite cell marker)-Venus reporter was used (Nalbandian et al., 2021). After approximately 80 days of culture, Venus-positive cells appeared in the differentiation culture (Supplementary Figure S1B) and were purified by flow cytometry (Supplementary Figure S1C). To replace the PAX7-Venus reporter hiPSCs, we developed a purification method using antibodies for surface markers. Although several surface markers are reported, we utilized CD82, a marker of progenitor cells in human skeletal muscles (Alexander et al., 2016; Uezumi, A. et al., 2016), as a positive selection marker, and CD57 (also known as HNK-1), a marker of neural crest cells, as a negative selection marker for iMuSCs (Borchin et al., 2013). After staining day-80 differentiated cells with CD82 and CD57 antibodies (see Methods), the differentiated cells were separated into four fractions: CD57⁺CD82⁺, CD57⁻–CD82⁺, CD57⁺CD82⁻, and CD57⁻–CD82⁻ (Supplementary Figure S2A). A gene expression analysis revealed that the CD57⁻–CD82⁺ fraction showed the highest expression level for several muscle stem/progenitor markers including PAX7, MYF5, and MYOD among the four fractions (Supplementary Figure S2B). An immunocytochemical analysis demonstrated that more than 80% of CD57⁻–CD82⁺ cells expressed PAX7 (Supplementary Figure S2C). Together, these results suggest that cell sorting the CD57⁻–CD82⁺ fraction efficiently purifies iMuSCs from non-reporter hiPSCs. Thus, we could obtain iMuSCs by PAX7-Venus fluorescence monitoring and by CD57⁻–CD82⁺ antibody selection.

We then optimized the in vitro myotube differentiation of the purified iMuSCs. Figure 1A shows a simple scheme of the in vitro myotube differentiation protocol. iMuSCs were proliferated to 100% confluence by AK02 media for 5 days (Supplementary Figure S3A), at which time we examined the component of differentiation media that affected the myotube differentiation efficiency. N2 supplement was found to promote the induction of myosin heavy chain (MHC)-positive myotubes, which is a hallmark of myogenic differentiation efficiency compared with 2% horse serum (HS) supplement (Figures 1B,C).

After switching to differentiation media with N2 supplement, iMuSCs displayed a spindle-like morphology at day 7 of the culture (Supplementary Figure S3B). Generally, myotubes are surrounded by connective tissues, such as extracellular matrix, in vivo. To imitate this environment, we developed a novel method to embed a high concentration of Matrigel to the culture dish in vitro (Supplementary Figure S4A) by referring to a previous report (Falcone, S. et al., 2014) that demonstrated murine primary satellite cell maturation in vitro. After embedding the Matrigel, iMuSC-derived myotubes began to fuse to each other and displayed spontaneous self-contractions at day 14 of the culture (Supplementary Figure S3C and Supplementary Movie S1). This feature was more obvious at day 42 of the culture (Supplementary Figure S3D).

Mature skeletal muscle is equipped with structures for contraction, such as the sarcomere and triad structures, which is composed of t-tubules and the SR. At day 42 of the culture, the intramuscular structure was evaluated by electron microscopy (Figure 1D). Well-aligned sarcomere and triad structures were visible, suggesting that the myotubes were considerably mature.

Finally, we evaluated the effect of Matrigel embedding by comparing the condition with Matrigel embedding (m-condition) to without Matrigel embedding (c-condition) (Supplementary Figure S4B). The size of the myotubes was measured between two conditions at day 42 of the culture to assess the maturation level of myotubes. m-condition increased the myotube diameter by 16.5% (p < 0.013) if using PAX7-Venus iMuSCs and by 19.8% if using CD57⁻–CD82⁺ iMuSCs (p < 0.009) (Supplementary Figures S4C,D).

To assess the maturity of differentiated myotubes, the expression of MYH genes was investigated between c-condition and m-condition at days 14 and 42 using PAX7-Venus and CD57⁻–CD82⁺ iMuSC-derived myotubes. The expression of MYH3 (embryonic subtype) was decreased from day 14 to day 42 in both iMuSC-derived myotube types (Figure 2A). This pattern of expression, which indicates a shift from developmental to adult myosin, is consistent with the findings during native muscle development (Schiaffino et al., 2015) and differentiation (Brown et al., 2012). Moreover, the expression of MYH3 itself was higher in m-condition than in c-condition.

On day 14, the expression of four different MYH subtypes associated with postnatal to adult development (MYH2,4,7,8) was elevated. Especially on day 42, the expression of MYH4 was elevated in m-condition compared with that in c-condition in PAX7-Venus iMuSC-derived myotubes (Figure 2A).

Moreover, the expression of genes related to SR and t-tubules was investigated. On day 42, the expression of SR-related markers (CASQ1 and ATP2A1) was notably elevated in m-condition compared with that in c-condition in PAX7-Venus and CD57⁻–CD82⁺ iMuSC-derived myotubes (Figure 2B). Also on day 42, the expression of t-tubule-related genes (DHPRβ and DHPRγ) was elevated in m-condition in both myotube types (Figure 2B). Altogether, these results suggest that m-condition promotes mature myotubes in vitro more than c-condition.

Finally, the expression of MYH, t-tubules, and SR related genes was investigated in m-condition with or without Agrin to clarify the effect of Agrin for myotube differentiation. The expression of most genes was comparable between with and without Agrin except a slight difference of MYH3 and DHPRβ at day 42. These data indicate that Agrin itself does not have significant effect on the expression of myogenic markers (Supplementary Figure S5). However, the triad structure was observed only in the presence of Agrin (Figure 1D) and never detected without Agrin condition (data not shown), suggesting that Agrin might have an important role for forming the triad structure.

Next, to assess physical maturity, the electrical properties of the myotubes derived from iMuSC were investigated. The gene expressions of SCN4A and SCN5A were investigated in both PAX7-Venus and CD57⁻–CD82⁺ iMuSC-derived myotubes (Figure 3A). Interestingly, SCN4A expression both in PAX7-Venus and in CD57⁻–CD82⁺ iMuSC-derived myotubes was increased dramatically in a time-dependent manner with the culture. The expression of SCN5A was increased in CD57⁻–CD82⁺ iMuSC-derived myotubes as the cells matured but was relatively stable in PAX7-Venus iMuSC-derived myotubes. These results suggest that Nav1.4 is dominant in both PAX7-Venus and CD57⁻–CD82⁺ iMuSC-derived myotubes.

Additionally, voltage-dependent sodium currents were measured in PAX7-Venus iMuSC-derived myotubes using the whole-cell patch-clamp technique. Representative raw traces of sodium currents recorded from cultured cells around day 7 of the culture are shown (Figure 3B left). Transient currents were elicited in a voltage-dependent manner, suggesting that currents were derived from Nav channels. To characterize the sodium currents, the kinetics of the channel gating was examined. The voltage dependence of the activation (Figure 3B, right, open circles) and the steady-state fast inactivation (Figure 3B, filled circles) are shown. The estimated parameters for the intermediate potentials ( V 1 / 2 ) and slope factors ( k ) are listed in Supplementary Table S2. Next, we investigated the TTX sensitivity of the Na⁺ current by the out-side-out patch clamp technique, which enables us to expose the membrane to TTX without physical prevention by the Matrigel. In the membrane excised from the myotube around day 11, approximately 140 pA of Na⁺ current was recorded in the conventional bath solution without TTX (Figure 3C, black). After exposure of 100 nM TTX, the current was blocked nearly completely (Figure 3C, gray), suggesting that the excitability of PAX7-Venus iMuSC-derived myotubes relies on the expression of Nav1.4 rather than Nav1.5.

hiPSC-based phenotype screening can be used to develop novel drugs (Yamashita, A et al., 2014). To use our myotubes as a resource for drug discovery, iMuSCs were differentiated to myotubes in a 96-well plate to model a multi-well drug screening system. Only the internal 60 wells in the plate were used because external margins of the plate are not always suitable due to instability of the cell culture (Lundholt, BK et al., 2003). At 28 days of culture, iMuSCs in each of the 60 wells were differentiated to myotubes with high homogeneity and positive for MHC (Figure 4A). The average differentiation efficiency was over 79%, and the coefficient of variation number (CV number) was under 3.3% (Figure 4B), suggesting that this model satisfied criteria for a multi-well drug screening system.

Experiments using hiPSCs were approved by the Ethics Committee at the Graduate School and Faculty of Medicine Kyoto University (approval numbers #E1762, #G567, and #Rinsho71). This study was performed conforming to the guidelines of the Declaration of Helsinki and was conducted after obtaining written informed consent.

The hiPSC line 201B7 was generated from purchased fibroblasts of a healthy donor by retroviral transduction (Takahashi et al., 2007). Ff-WJ14s516 (an HLA-homozygous hiPSC line with the most frequent haplotype in Japan; abbreviated as S516 in this manuscript) was established from cord blood cells by an episomal vector system (Okita et al., 2013) at the Facility for iPS cell Therapy, CiRA, Kyoto University. HiPSCs were cultured in StemFit media (AK02N; Ajinomoto) and passaged once a week. At the passage, the cells were completely detached from a 3.5-cm well in a dish by incubating in 500 μl of Accutase (Nacalai Tesque) for 10 min followed by neutralization with 2 ml of Stemfit media. After pipetting gently, 3 × 10⁵ hiPSCs were seeded on the dish in Stemfit media containing a Rock inhibitor, 10 μM Y-27632 (Nacalai Tesque). The medium was replaced with 1.5 ml of StemFit media without the Rock inhibitor 2, 4, 5, and 6 days after passage. The maximum passage number was 30.

The myogenic differentiation of hiPSCs was performed following a previous protocol (Zhao M. et al., 2020). Cell populations including fetal iMuSCs were obtained at days 80–100. The purification of iMuSCs was performed by FACS sorting (BD FACS aria 650110J1). For the 201B7-Pax7 Venus line (Nalbandian et al., 2021), the iMuSCs were purified by Venus fluorescence. For non-reporter lines, iMuSCs were purified as the CD57⁻–CD82⁺ population. Allophycocyanin-conjugated CD82 antibody (Biolegend 342108) was used at a final concentration of 1:100. Phycoerythrin-conjugated CD57 antibody (BD Pharmingen 560844) was used at a final concentration of 1:200. The cells were stained with the antibodies for 30 min, followed by washing with HBSS buffer once.

Purified iMuSCs were plated onto a 1:100 Matrigel growth factor reduced (Corning 354230) in a StemFit media-coated 96-well microplate (IWAKI 3860–096). The microplate was pre-coated with Matrigel diluted by StemFit media at 1:100 for 1 h at 37°C. Then, 5,000 purified iMuSCs were seeded onto the plate and cultured in StemFit media containing 10 μM Y-27632 for expansion.

iMuSCs started to proliferate and reach confluency after approximately 5 days of culture. In this process, the medium was changed every day until day 5. After the day of seeding, the medium was replaced to Stemfit media without Y-27632. Differentiation was induced by switching to differentiation media composed of DMEM (Invitrogen 11885084) supplemented with 1% N2 supplement (Gibco 17502048) or 2% HS (Sigma H1138). Recombinant rat Agrin (R&D Systems 550-AG) was used for better differentiation.

At day 7 of the culture, 3D Matrigel covering was performed. A Matrigel mixture composed of cold DMEM media (40 μl) and Matrigel (20 μl) was prepared. First, the differentiation media was completely removed from the well, and myotubes were covered with the Matrigel mixture as previously described (Falcone S. et al., 2014). Afterward, the culture dish was incubated at 37°C for 30 min. As a result, myotubes were covered with a high concentration of Matrigel mixture. Next, the differentiation medium containing Agrin at a final concentration of 80 μg/ml was gently added on the 3D Matrigel. Finally, the culture dish was moved to a 37°C incubator, and the medium was changed every 2 days. After the Matrigel coating, Agrin-containing DMEM/N2 media was used as the differentiation media.

Total RNA was isolated using a ReliaPrep RNA cell Miniprep Kit (Promega Z6012) according to the manufacturer’s instructions. Reverse transcription was performed by ReveTra Ace qPCR RT Mater Mix with gDNA remover (TOYOBO FSQ-301). SYBR Green reagent (Applied Biosystems) was used to detect the target sequences. qPCR was performed using StepOnePlus. (Applied Biosystems) The list of primers used in this study is shown in Supplementary Table S1.

Myotubes were fixed with PBS containing 2% (w/v) paraformaldehyde for 10 min at 4°C. Samples were washed with PBS twice, followed by blocking with Blocking One (Nacalai Tesque) for 60 min at 4 °C and incubated overnight at 4 °C with primary antibodies diluted in 10% (v/v) Blocking One in PBS-T (PBS with 0.2% (v/v) Triton X-100 solution (Nacalai Tesque). The samples were then washed 3 times with PBS-T and incubated for 1 h at room temperature with secondary antibodies diluted in 10% (v/v) Blocking One in PBS-T. Finally, the samples were incubated with DAPI for 5 min to visualize cellular nuclei, followed by washing with PBS twice.

An analysis of myosin heavy chain (MHC)-positive cells was performed using Arrayscan High-Content Systems (Thermofisher). After immunostaining of the myotubes with MHC antibody, the ratio of nuclei with the surrounding MHC fluorescence to total nuclei was calculated. The threshold was changed depending on the background signal.

The samples were fixed with 2% paraformaldehyde (PFA) and 2% glutaraldehyde (GA) in 0.1 M phosphate buffer (PB), pH 7.4, at the incubation temperature (37°C) and then put into a refrigerator for 30 min to lower the temperature to 4°C. Thereafter, they were fixed with 2% GA in 0.1 M PB at 4°C overnight. After the fixation, the samples were washed 3 times with 0.1 M PB at 4°C for 1 h. The samples were dehydrated in graded ethanol solutions (50%, 70%, 90%, 100%) as follows: 50% and 70% for 5 min each at 4°C, 90% for 5 min at room temperature, and three changes of 100% for 5 min each at room temperature. Then, the samples were transferred to a resin (Quetol-812; Nissin EM Co., Tokyo, Japan) and polymerized at 60°C for 48 h. The polymerized resins were ultra-thin-sectioned at 70 nm with a diamond knife using an ultramicrotome (Ultracut UCT; Leica, Vienna, Austria) and mounted on copper grids. They were stained with 2% uranyl acetate at room temperature for 15 min and then washed with distilled water, followed by secondary staining with Lead stain solution (Sigma-Aldrich Co., Tokyo, Japan) at room temperature for 3 min. The grids were observed by a transmission electron microscope (JEM-1400Plus; JEOL Ltd., Tokyo, Japan) at an acceleration voltage of 100 kV. Digital images (3,296 × 2,472 pixels) were taken with a CCD camera (EM-14830RUBY2; JEOL Ltd., Tokyo, Japan). Samples were analyzed by Tokai Electron Microscopy, Inc. (Nagoya, Japan).

To measure sodium currents, a modified culture protocol was used to easily access the patch electrode to an isolated cell. Briefly, 12-mm round glass coverslips (Warner Instruments, Holliston, MA) were used with Matrigel the day before seeding. The cells were harvested on the Matrigel-coated coverslips without overlaying the Matrigel mixture. The myotubes were cultured in the differentiation medium until the patch-clamp experiment.

Sodium currents were measured using a patch-clamp electrode and the whole-cell recording technique. Recordings were made using an Axopatch 200B amplifier (Molecular Devices, San Jose, CA), and data acquisition was done using a Digidata 1440A (Molecular Devices, San Jose, CA). pCLAMP10.7 software was used (Molecular Devices, San Jose, CA) for the data collection and analysis. Patch electrodes were fabricated from borosilicate glass tubes (Sutter, Novato, CA) using a P-97 Flaming/Brown Micropipette Puller (Sutter, Novato, CA). The electrode’s tip was heat-polished with a final resistance of 1.5–2.3 MΩ in the bath solution. Series resistance was compensated at 80–85% by the analog circuitry of the amplifier. All recordings were conducted at room temperature (25°C). The pipette internal solution contained (in mM) 105 CsF, 35 NaCl, 10 ethylene glycol tetraacetic acid (EGTA), and 10 Cs-HEPES (pH 7.4). The bath solution contained (in mM) 140 NaCl, 4 KCl, 2 CaCl₂, 1 MgCl₂, 5 glucose, and 10 Na-HEPES (pH 7.4).

Before starting a measurement, the membrane potential was held at −100 mV for about 5 min to ensure a recovery from slow inactivation. The activation of Na currents was measured by applying a 10 ms depolarizing step pulse from −80 to +80 mV at 5 mV increments. Any cells with peak currents <1 or >10 nA on a step depolarization from −120 mV to −10 mV were excluded. The voltage dependence of the steady-state fast inactivation was measured as peak inward currents elicited by a −10 mV test pulse after a 300 ms conditioning potential of −120 mV to −10 mV. For out-side-out configuration, we prepared electrodes with a final resistance of 1.5 MΩ. A control current without TTX was obtained by a −10 mV test pulse from −120 mV, Then, the current in the presence of 100 nM TTX was obtained by the identical test pulse.

For the validation of screening system, myotubes differentiated at 28 days with Matrigel covering the in vitro culture were used. The myotubes were stained with MHC antibody (Santacruz sc-20641) followed by analysis using a high-content screening instrument (Thermo Scientific Arrayscan System).

Curve fitting was manually performed using Origin (OriginLab Northampton, MA). Conductance was calculated using the following equation: G ( V ) = I p e a k / ( V − E r e v ) The reversal potential, E r e v , was measured experimentally for each cell. The voltage dependence of activation was calculated from the following Boltzmann equation: G ( V ) = G m a x / [ 1 + e x p ( − ( V − V 1 / 2 ) / k ) ] Steady-state fast inactivation was fitted to the Boltzmann function calculated from the following equation, where V 1 / 2 is the half-maximum voltage, and k is the slope factor: I / I m a x = 1 / [ 1 + e x p ( ( V − V 1 / 2 ) / k ) ] I m a x was calculated as the average of the three I values before and after the potential, where I shows the maximum point.

Medium components, myofiber diameter, and qPCR (MYH, SCN, SR-related, and t-tubule-related genes) were analyzed using an unpaired two-tailed Student’s t-test. The expressions of Pax7, MyoD, and Myf5 were analyzed using a t-test with Bonferroni correction. The expressions of SCN4A and SCN5A were analyzed using Tukey’s test. Data are presented as means ± SEM. p < 0.05 was considered statistically significant.