Human Pluripotent Stem Cell-Derived Skeletal Muscle Organoid Model of Aging-Induced Sarcopenia

Abstract

Background: Sarcopenia is defined by the age-related loss of muscle mass and function, with an impaired regenerative capacity of satellite cells (SCs). Despite their recognized importance in muscle regeneration, human model-based studies on SCs in sarcopenia are still lacking, limiting our understanding of their role in age-related muscle loss. Here, we aimed to develop a sarcopenia model using human pluripotent stem cells (hPSCs)-derived skeletal muscle organoids (hSkMOs) and prevent the sarcopenia progression by testosterone treatment.

Methods: The 3D hSkMOs were generated from hPSC and exhibited structurally and functionally mature muscle fibres and spinal-derived neurons including motor neurons and interneurons. The proportion of muscle and the diameter of muscle fibres were assessed. To investigate the acute pro-inflammatory response and intrinsic regenerative capacity of hSkMOs, we induced sarcopenia-like conditions by TNF-α treatment for 2 days and analysed. To model aging-induced sarcopenia and investigate the preventive effect of testosterone, chronic TNF-α treatment was applied, followed by testosterone administration. Histological, biochemical, molecular and electrophysiological analyses were conducted in various experiments.

Result: We employed a stepwise differentiation protocol from 2D paraxial mesodermal induction to 3D myogenic specification, concluding with a maturation culture system. We observed that the majority of cells were T/BRA- and TBX6-positive (⁺) paraxial mesodermal progenitors (T/BRA⁺, 82.04%; TBX6⁺, 78.18%), whereas the neuromesodermal progenitors demonstrated a relatively low proportion (T/BRA⁺/SOX2⁺, 15.91%; TBX6⁺/SOX2⁺, 11.45%). Single-nucleus RNA-sequencing and extensive immunohistochemistry confirmed the presence of the myogenic lineage cell types (myogenic progenitors/SCs, myocytes, muscle fibres) and the neural lineage cell types (spinal-derived interneurons, motor neurons, glial cells, Schwann cells). Additionally, the growth of MyHC⁺ muscle fibres reached twice the thickness on Day 100 compared to that on Day 50 (p < 0.0001). We subjected them to TNF-α treatment and analysed. Western blot analysis confirmed that TNF-α/NF-κB pathway associated factors such as NF-κB p65, IκB-α and AKT were highly phosphorylated (p < 0.05, p < 0.001). The administration of testosterone increased the proportion of activated SCs (PAX7⁺/MYOD⁺, 7.97%; PAX7⁺/Ki67⁺, 7.03%) compared to the TNF-α group (PAX7⁺/MYOD⁺, 2.29%; PAX7⁺/Ki67⁺, 2.07%, p < 0.001). The administration of testosterone increased the Cross-Sectional-Area (987.1 μm²) compared to the TNF-α group (644.7 μm², p < 0.01).

Conclusions: We successfully developed a hSkMOs to demonstrate the structural maturity of the skeletal muscle and its functional interaction with spinal-derived interneurons and motor neurons. Furthermore, we demonstrated that our hSkMOs are useful for modelling aging-induced sarcopenia and providing a valuable platform for testing therapeutic interventions.

Keywords: aging; human pluripotent stem cell; human skeletal muscle organoid; sarcopenia; satellite cell; testosterone.

FIGURE 1

Efficient generation of hPSC‐derived 3D hSkMOs. (A) A schematic diagram showing the overall strategy to generate hSkMOs from hPSCs. The representative images of hSkMOs morphology are shown at each time point. Scale bar = 500 μm. The growth diameter demonstrates the average size (mean ± SEM; n = 10). (B) Immunocytochemistry analyses of Day 3 PMPs stained for T/BRA, TBX6, SOX2 and DAPI (mean ± SEM; n = 5). Scale bars = 200 μm. (C) A whole mount staining image of a Day 9 hSkMOs stained for PAX3, TBX6 and DAPI (mean ± SEM; n = 5). Scale bar = 100 μm. (D) Cryosection of Day 20 hSkMOs stained for PAX7, Ki67 and DAPI (mean ± SEM; n = 5). White scale bar = 200. Yellow scale bar = 20 μm. (E) Cryosection of Day 20 hSkMOs stained for PAX7, MYOD and DAPI (mean ± SEM; n = 5). White scale bar = 200. Yellow scale bar = 10 μm. (F) Cryosection of Day 20 hSkMOs stained for PAX7, MYOG and DAPI (mean ± SEM; n = 5). White scale bar = 200. Yellow scale bar = 20 μm.

FIGURE 2

Structural characterization of mature muscle fibres in hSkMOs. (A) A schematic illustration of mature muscle fibres and functional connection with spinal‐derived motor neurons. (B) Cryosection of Days 50 and 100 hSkMOs stained for MyHC, TUJ1 and DAPI. White scale bar = 500. Yellow scale bar = 100 μm. (C) Quantification of the proportion of TUJ1⁺ neural and MyHC⁺ muscle region at day 50 and 100 (mean ± SEM; n = 14). (D) Quantification of the mean fibre diameter stained for MyHC (mean ± SEM; n = 32). (E) Cryosection of Day 100 hSkMOs stained for MyHC, LAMININ and DAPI. Scale bar = 10 μm. (F) Cryosection of Day 100 hSkMOs stained for PAX7, LAMININ and DAPI. Scale bar = 10 μm.

FIGURE 3

Structural characterization of spinal‐derived motor neurons and functional connection with muscle fibres in hSkMOs. (A) Cryosection of Day 100 hSkMOs stained for ChAT, TUJ1 and DAPI. Scale bar = 50 μm. (B) Cryosection of Day 100 hSkMOs stained for GFAP, MAP2 and DAPI. Scale bar = 20 μm. (C) Cryosection of Day 100 hSkMOs stained for αBTX, TUJ1 and MyHC. Scale bar = 10 μm. (D) Cryosection of Day 100 hSkMOs stained for MyHC, S100β, αBTX and DAPI. Scale bar = 10 μm. (E) and (F) Electron microscopic images to show skeletal muscle fibres of Day 100 hSkMOs. Mi: Mitochondria, M: M‐band, Z: z line., SV: Synaptic vesicles, BL: Basal lamina, Ax‐in: Axonal innervation, MC: Muscle cell, Sch: Schwann cell, My: Myelin‐like sheath. Black scale bar = 500 nm. White scale bar = 2 μm.

FIGURE 4

Single‐nucleus transcriptomic profiling reveals dynamic cellular heterogeneity in hSkMOs. (A) UMAP plot of snRNA‐seq data from Day 50 hSkMOs, showing distinct clusters representing major cell types including myogenic progenitors/SCs, myocytes, muscle fibres, neural cells, FAPs and sclerotome. Cell type proportions at Day 50. (B) Dot plot showing expression of representative genes in each cluster in Day 50 hSkMOs. (C) Subclustering of the myogenic progenitors/SCs identifies quiescent SCs, activated/proliferating SCs and differentiating SCs. Dot plot showing the expression of representative genes across the three subclusters. (D) Neural subclustering reveals five distinct cell populations including GABAergic interneurons, glutamatergic interneurons, motor neurons, proliferative neural progenitors and glial cells/Schwann cells. Myogenic prog.: Myogenic progenitors, Quiesc. SCs: Quiescent SCs, Activ./Prolif. SCs: Activated and proliferating SCs, Differ. SCs: Differentiating SCs, GABA. Interneurons: GABAergic interneurons, Glut. Interneuron: Glutamatergic interneurons, Prolif. neural prog.: Proliferative neural progenitor cells, Sch. cells: Schwann cells.

FIGURE 5

Functional characterization of hSkMOs. (A) Spontaneous contraction and (B) calcium imaging with Fluo‐4 AM for visualizing calcium influx of Day 100 hSkMOs. Scale bar = 50 μm. (C) Representative trace of the spontaneous postsynaptic potential (PSP) response obtained from the section of the organoid with muscle fibres (top). Enlarged PSP response observed (bottom left). Their amplitude and frequency (bottom right) from individual muscle fibres targeted (mean ± SEM; n = 8). (D) Representative trace graph depicting spontaneous contractions in mature hSkMOs, and their pharmacological modulation by acetylcholine (ACh) and curare. (E) Quantitative analysis of contraction dynamics in control, ACh‐treated, and curare‐treated hSkMO (mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001; n = 3). (F) ACh and curare affect the postsynaptic potential response from the section of hSkMOs. Traces obtained from three independent cells presented and responses are marked with asterisks.

FIGURE 6

TNF‐α/NF‐κB pathway‐induced muscular damage and regenerative capacity of hSkMOs. (A) A schematic diagram showing the overall strategy to acute treatment of TNF‐α on day 100 hSkMOs. (B) Western blot of p‐NF‐κB‐p65, p‐IκB‐α, p‐AKT, MYOG and β‐actin expression from Vehicle, Days 0, 3, 7 after TNF‐α treatment (mean ± SEM; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; n = 3). (C) Cryosection of Day 100 hSkMOs stained for PAX7, MYOD and DAPI (mean ± SEM; *p < 0.05, ***p < 0.001; n = 3). Scale bar = 50 μm. (D) Cryosection of Day 100 hSkMOs stained for PAX7, Ki67 and DAPI (mean ± SEM; *p < 0.05, ***p < 0.001; n = 3). Scale bar = 50 μm. (E) Cryosection of Day 100 hSkMOs stained for MYOG and DAPI (mean ± SEM; ***p < 0.001; n = 3). Scale bar = 100 μm. (F) Cryosection of Day 100 hSkMOs stained for MyHC, TUJ1 and αBTX (mean ± SEM; ****p < 0.0001; n = 20, ***p < 0.001, **p < 0.01; n = 3). Scale bar = 100 μm.

FIGURE 7

Sarcopenia modelling in hSkMOs and testosterone effects on attenuating muscle wasting. (A) A schematic diagram showing the overall strategy to chronic treatment of TNF‐α and testosterone on Day 100 hSkMOs. (B) Cryosection of Day 100 hSkMOs from each Vehicle, TNF‐α, and TNFα with Testosterone group stained for PAX7, androgen receptor (AR) and DAPI (mean ± SEM; **p < 0.01, ***p < 0.001; n = 3). Scale bar = 50 μm. (C) The representative images of hSkMOs from each Vehicle, TNFα, and TNFα with testosterone group (mean ± SEM; **p < 0.01, ****p < 0.0001; n = 6). Scale bar = 500 μm.

FIGURE 8

Regenerative capacity of hSkMOs upon testosterone treatment. (A) Cryosection of Day 100 hSkMOs from each Vehicle, TNF‐ɑ and TNFα with testosterone group stained for PAX7, MYOD and DAPI (mean ± SEM; **p < 0.01, ***p < 0.001; n = 3). Scale bar = 100 μm. (B) Cryosection of Day 100 hSkMOs from each Vehicle, TNF‐α and TNF‐α with testosterone group stained for PAX7, Ki67 and DAPI (mean ± SEM; **p < 0.01; n = 3). Scale bar = 100 μm. (C) Cryosection of Day 100 hSkMOs from each Vehicle, TNF‐α and TNF‐α with Testosterone group stained for MyHC, MYOG and DAPI (mean ± SEM; ****p < 0.0001; n = 5 and **p < 0.01; n = 50). Scale bar = 100 μm. (D) Cryosection of Day 100 hSkMOs from each Vehicle, TNF‐α and TNFα with testosterone group stained for αBTX (mean ± SEM; **p < 0.01, ***p < 0.001; n = 3). Scale bar = 20 μm. (E) Representative voltage traces of cells from the section of hSkMO subjected to various treatment conditions. (F) Co‐treatment with testosterone prevented the loss of postsynaptic potential observed in TNF‐α exposed cells (mean ± SEM; ***p < 0.001; Control: n = 14, TNF‐α: n = 14, TNF‐α + T: n = 13).