Engineering Stem Cell-Derived Neuromuscular Circuits to Model Amyotrophic Lateral Sclerosis (ALS)

Abstract

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease characterised by degeneration of motor neurons in the brain and spinal cord, leading to progressive paralysis and death. Currently no cure exists, and the only treatments available extend life by a matter of months. Cytoplasmic aggregation of TAR DNA-binding protein 43 (TDP-43) is a key pathological hallmark, seen in the vast majority of patients with ALS. The mechanisms by which TDP-43 contributes to motor neuron degeneration are not fully understood, although aberrant alternative splicing and perturbed RNA metabolism are thought to play pivotal roles. Emerging evidence suggests that some of the earliest pathological events in the progression of ALS occur peripherally. These include neuronal hyperexcitability, spontaneous muscle fasciculations (twitching), impaired neuromuscular transmission, neuromuscular synapse dismantling and axonal degeneration. To develop drugs that can target these early peripheral events, better in vitro disease models are required that accurately reflect these complex phenotypes.

This work describes the development of human induced pluripotent stem cell (hiPSC) derived neuromuscular circuits (Chapter 4), based on preliminary work using mouse embryonic stem cells (mESCs) and primary chick myoblasts (Chapter 3). CRISPR-Cas9 genome editing was used to correct a pathogenic ALS-related TDP-43^G298S mutation in patient derived hiPSCs. hiPSC lines were also engineered to express a MACS sortable motor neuron marker and the optogenetic actuator CHR2-YFP to enrich and control the activity of hiPSC-derived motor neurons. MACS enriched hiPSC-motor neurons were co-cultured with iPAX7 forward programmed hiPSC-myoblasts in compartmentalised microdevices, whereby motor axons were able to grow through micro-channels and form functional neuromuscular junctions (NMJs) with target muscle. Optogenetic stimulation of the motor neurons elicited robust myofiber contractions and daily optogenetic entrainment was found to enhance neuromuscular synapse formation, mirroring the activity-dependence of NMJ formation in vivo. Furthermore, it was found that the neuromuscular co-cultures containing TDP-43^G298S motor neurons displayed increased spontaneous myofiber contractions yet weaker optogenetically evoked maximal contractile output, which was linked to reduced axon outgrowth and a lower number of neuromuscular synapses (Chapter 4). These phenotypes resemble the muscle fasciculations and muscle weakness observed in patients with ALS.

In conjunction with the compartmentalised co-culture format, a high-throughput 96-well hiPSC-neuromuscular co-culture platform was established (Chapter 5), whereby neuromuscular disease phenotypes could be imaged and quantified automatically using high content image analysis more rapidly and at a larger scale than has previously been possible. Using this approach, it was possible to model both ALS related phenotypes and Duchenne muscular dystrophy (DMD) related phenotypes – two contrasting neuromuscular diseases. Furthermore, it was possible to test candidate drugs and it was found the RIPK1 inhibitor necrostatin could partially rescue ALS-related neuromuscular phenotypes, while the TGFβ inhibitor SB431542 could rescue DMD-related phenotypes. Finally, by generating custom- built 96-well plates with suspended electrospun elastic nanofibers it was possible to stabilise contractile myofibers for longer-term neuromuscular culture stability and maturation in the 96-well plate format.

Finally, to uncover mechanistic insights into the hyperactivity of the ALS-TDP-43^G298S neuromuscular co-cultures, whole-cell patch clamp recordings were performed. These showed dysregulated neuronal excitability of TDP-43^G298S motor neurons, relative to wildtype and CRISPR-corrected controls. Interestingly, these changes were associated with structural abnormalities of the axon initial segment (AIS) – the primary site of action potential initiation, as well as perturbed functional plasticity of the AIS. Specifically, early (6-weeks) TDP-43G298S motor neurons showed an increase in the length of the AIS and hyperexcitability, while late (10-weeks) TDP-43^G298S motor neurons showed shortening of the AIS and hypoexcitability. Furthermore, at all stages, activity-dependent plasticity of the AIS was impaired, further contributing to abnormal homeostatic regulation of neuronal excitability (Chapter 6). Taken together these results suggest for the first time that abnormal AIS structure and plasticity may contribute to dysregulated neuronal excitability in ALS-related TDP-43^G298S motor neurons.

Date of Award	1 Jan 2023
Original language	English
Awarding Institution	King's College London
Supervisor	Ivo Lieberam (Supervisor) & Madeline Parsons (Supervisor)