Neuromuscular disorders: Controlling contractions with light

Neuromuscular disorders include ALS, myasthenia gravis, and the muscular dystrophies such as Duchenne's. Collectively these disorders exceed an incidence of 1 in 3,000. Although there is a strong genetic understanding of many of these disorders, the poor translatability of animal models to humans has hindered the development of treatments for these diseases. Consequently, there is a need for a model that more faithfully recapitulates the physiology of the human neuromuscular junction. In this webinar, Dr. Elliot Swartz (UCLA) discusses how he is building a light controlled hiPSC model of a neuromuscular junction to help better understand neuromuscular disorders.

Transcript of webinar:

Thank you for joining us on today's Coffee Break Webinar. Today's topic is Controlling contractions with light, building a functional human neuromuscular junction in a dish.

The neuromuscular junction is a synapse formed between a motor neuron and a muscle fiber when an action potential reaches the presynaptic terminal of a motor neuron. It can trigger the release of the neurotransmitter acetylcholine into the synaptic cleft the binding of acetylcholine to the receptors on the muscle fiber depolarizes the cell causing a cascade that eventually results in muscle contraction. Diseases that impair the functioning of muscles either directly or indirectly via motor neurons are referred to as neuromuscular disorders. Neuromuscular disorders include myasthenia gravis, ALS, and the muscular dystrophies such as Duchenne's, collectively these disorders exceed an incidence of 1 in 3,000. Although there is a strong genetic understanding of many of these disorders, the poor translatability of animal models to humans has hindered the development of treatments for these diseases. Consequently there is a need for a model that more faithfully recapitulates the physiology of the human neuromuscular junction. Today's presenter Elliott Swartz, is a PhD candidate in Giovanni Coppola's lab at UCLA. His expertise is in stem cell biology and mechanisms of neurodegeneration. Elliott's research uses induced pluripotent stem cells to create in vitro model systems which recapitulate the physiology of neurodegenerative and neuromuscular disorders. He will discuss how he is building a patient-derived model of a neuromuscular junction in a dish with the aim of providing a greater understanding of these neuromuscular disorders.

Thank you for the introduction, I'll put the context of this webinar under the umbrella of neuromuscular disorders, these disorders represent a group of over 200 rare monogenic disorders that collectively exceed an incidence of 1 in 3,000. Neuromuscular disorders are grouped into 16 broad categories, in some cases, such as the muscular dystrophies muscle is the primarily affected tissue, whereas in diseases such as amyotrophic lateral sclerosis also known as ALS or Lou Gehrig's disease, motor neurons are primarily affected.

We've gained a strong genetic understanding for many of these disorders with over 400 genes being implicated thus far. Nevertheless patients with defects in these genes suffer long term chronic illnesses which are often severe or fatal with few if any effective treatments outside of recent advances in spinal muscular atrophy. The neuromuscular junction shown here is the specialized synapse between motor neurons and skeletal muscle that is responsible for motor function. In many neuromuscular disorders, we know that neuromuscular junction collapse precedes symptom onset and neuronal loss, suggesting that aiming to ameliorate this upstream event may be beneficial therapeutically, however, little is known about how denervation actually occurs in disease.

To create new models of neuromuscular disorders our lab has utilized patient-specific induced pluripotent stem cells or iPSCs. To narrow the scope of this talk I'll focus on ALS as a representative model, although you'll see that the experiments performed can be applicable to a wide range of neuromuscular disorders. To model ALS our lab has created iPSCs from patients with hexanucleotide repeat expansion mutations in the gene C9ORF72, now known to be the most common cause of ALS and frontotemporal dementia, which is often comorbid in these patients. Additionally, we have created genetically corrected isogenic control lines to increase detection of mutation specific phenotypes, many studies modeling ALS with iPSCs have focused solely on creating spinal motor neurons losing the physiological importance of the neuromuscular junction connection, additionally the lack of reliable biomarkers or access to tissues in living humans due to ethical boundaries or a loss of tissue at end-stage disease has prevented knowledge from being gained from human disease models. Therefore much of our knowledge of neuromuscular function and disease comes from animal models such as the mouse, which have classically had poor translational outcomes.

We think therefore that by deriving both skeletal muscle and motor neurons this will create an all human patient derived model of the neuromuscular junction, which will enable new avenues for research in a basic biology of neuromuscular function and therapeutic discovery and disease. To establish a neuromuscular model we first derived motor neurons using a published protocol where motor neurons are efficiently obtained from suspension culture within approximately 30 days. We utilized our previously published protocol to generate skeletal muscle which expresses mature cytoskeletal markers, and displays sarcomeric organization and spontaneous contractility in vitro, as shown in the video on the right. By deriving each cell type independently and recombining at a later stage, we could obtain functional neuromuscular connections in less than seven days in vitro. Through patch-clamp electrophysiology recording, spontaneous activity could be abolished with the neuromuscular blocking agent kirari, as shown at the bottom.

An example of the co culture is shown here, where multiple myotubes contract synchronously in response to spontaneous motor neuron firing, which resembles the formation of a motor unit, in vitro. In order to gain experimental control over our system we express channelrhodopsin under control of a motor neuron specific promoter, to enable optogenetic control of IPSC derived motor neurons. To test the system we recorded optically evoked calcium transients using the red shifted calcium sensor X ROD1 am in innervated skeletal myotubes as shown in this diagram. Here we demonstrated evoked calcium response in to innovative myotubes where the calcium intensity changes in response to multiple pulses of blue light firing. Despite the success this methodology is low throughput and requires manual recordings making phenotypic analyses difficult.

In order to increase throughput we utilize multi-electrode arrays or MEA's where each well contains a four by four grid of electrodes that enable filtered spiking activity to be recorded from electrophysiologically active cells. Using the MEA we can quantify parameters related to firing rate, bursting, and network activity. Additionally, using the Lumos device we can perform simultaneous optical stimulation during live electrophysiological recordings. We created co cultures on a 48-well MEA plate shown in this video, which is sped up two times, each flash represents spontaneous activity in the corresponding well. Shown here, is a representative well, where the electrode grid as well as motor neuron clusters expressing channelrhodopsin can be seen. When we begin to provide blue light stimulation you can observe the synchronized activity within the same pictured well. Quantifying this across the plate shows that evoked stimulation increases the number of spikes and bursts per electrode, as well as the overall number of active electrodes. Network activity is also increased upon evoked stimulation suggesting that the optogenetic system performs as expected.

Given that both skeletal muscle and motor neurons are electrophysiologically active we attempted to determine the origin of the recorded signals, to do this we utilize two neuromuscular antagonists vecuronium and decamethonium bromide, which work via different mechanisms to produce blockade of the neuromuscular junction. For the assay we quantified the evoked spike counts, or spikes that occur in a 500 millisecond window following light stimulation, shown in the pink highlighted regions of the representative raster plot. In these preliminary data both vecuronium and decamethonium bromide completely abolished evoked activity, suggesting that it recorded activity as mediated by neuromuscular transmission. The variance in the vehicle treatment may be explained by electrodes with very low evoked spike counts, which were not excluded from this analysis.

Additionally and data not shown here, we have determined that IPSC derived skeletal myotubes are electrotonically coupled via gap junctions, which can be blocked by the gap junction inhibitor o1-heptanol. Treating co-cultures with heptanol also abolish signal, with some rescue of signals at lower dosages or following washout. These data suggest that the cell type being recorded is actually skeletal myotubes rather than motor neurons, where bursting activity represents electronically coupled myotubes activated via light-evoked neuromuscular transmission. We looked at cellular morphology corresponding to the active electrodes which revealed dense regions of myotubes on top of electrodes, with motor neuron clusters nearby. As a model to explain these data we first coat electrodes in a protein matrix and plate Myotubes at high density on top, presumably making contact with the electrodes. After a few days, we add motor neurons on top of the skeletal muscle with an additional protein matrix. This protocol seems to prevent motor neurons from making direct contact with the electrodes due to physical separation, leaving the myotubes as the cell type in contact. Although we cannot rule out individual or small clusters of motor neurons contributing to recorded signals.

In conclusion, we established a scalable, tunable, all human iPSC-derived neuromuscular co culture system. We show that neuromuscular activity can be precisely controlled via optogenetics and this tool can be combined with MEAs and characterized drugs to determine signal origin in a co culture system. In future experiments, we plan to collect additional pharmacological data over time to support our model of co cultures on MEAs. We are also interested in using isogenic IPSC lines to determine if physiological disease-relevant phenotypes exist, as well as test or screen potential therapeutic candidates which modify the function of the neuromuscular junction.

I'd like to acknowledge Giovanni Coppola and the members of the Coppola lab that contributed to this project as well as the Novak's lab at UCLA for lending the optogenetic constructs. Thank you for listening.

And, that is the conclusion for today's coffee break webinar, if you have any questions you would like to ask regarding the research presented, or if you are interested in presenting your own research with microelectrode array technology please forward them to For questions submitted for Elliott Swartz, he will be in touch with you shortly. Thank you for joining in on today's coffee break webinar and we look forward to seeing you again.