How do we get from hyperexcitability to excitotoxicity in amyotrophic lateral sclerosis?

Abstract

Amyotrophic lateral sclerosis is a devastating neurodegenerative disease that, at present, has no effective cure. Evidence of increased circulating glutamate and hyperexcitability of the motor cortex in patients with amyotrophic lateral sclerosis have provided an empirical support base for the 'dying forward' excitotoxicity hypothesis. The hypothesis postulates that increased activation of upper motor neurons spreads pathology to lower motor neurons in the spinal cord in the form of excessive glutamate release, which triggers excitotoxic processes. Many clinical trials have focused on therapies that target excitotoxicity via dampening neuronal activation, but not all are effective. As such, there is a growing tension between the rising tide of evidence for the 'dying forward' excitotoxicity hypothesis and the failure of therapies that target neuronal activation. One possible solution to these contradictory outcomes is that our interpretation of the current evidence requires revision in the context of appreciating the complexity of the nervous system and the limitations of the neurobiological assays we use to study it. In this review we provide an evaluation of evidence relevant to the 'dying forward' excitotoxicity hypothesis and by doing so, identify key gaps in our knowledge that need to be addressed. We hope to provide a road map from hyperexcitability to excitotoxicity so that we can better develop therapies for patients suffering from amyotrophic lateral sclerosis. We conclude that studies of upper motor neuron activity and their synaptic output will play a decisive role in the future of amyotrophic lateral sclerosis therapy.

Keywords: amyotrophic lateral sclerosis; dying forward; excitotoxicity; homeostasis; hyperexcitability.

Figure 1

The complexity of the corticomotor system and the limitations of the tools we use to study it. Amyotrophic lateral sclerosis has been studied at three levels across the nervous system: networks, neurons and synapses. Network: Our understanding of network physiology has been predominantly derived from research implementing transcranial magnetic stimulation (TMS) to patients with amyotrophic lateral sclerosis. TMS can induce activation of brain regions to measure the excitability of networks (glowing brain region under the coil). Clinical TMS studies have revealed that the cortex of patients is often more excitable, such that peripheral motor responses can be more easily induced for any intensity of stimulus provided to the motor region. TMS is limited by cellular resolution and only comments on excitability. As such, the endogenous activity of motor networks, and the cells within them, in patients remains unknown. Neuron: Our understanding of neuronal physiology in amyotrophic lateral sclerosis has been predominantly derived from research implementing whole-cell patch-clamp of neurons in the corticomotor system of preclinical rodent models. Patch-clamp can measure currents that underlie neuronal function, but this is limited to the region of neurons that can be confidently controlled by the experimenter (blue region of the neuron depicts area of a neuron that is likely clamped by the pipette and accurately recorded from). Patch-clamp studies have revealed that layer 5 projection neurons of the motor cortex are often intrinsically hyperexcitable, such that their likelihood of firing action potentials is increased for any intensity of injected current. Patch-clamp is limited by its poor synaptic resolution, can only measure single neurons in isolation and only comments on excitability. As such, the endogenous activity of upper motor neurons remains unknown. Synapse: Our understanding of synaptic physiology in amyotrophic lateral sclerosis has been predominantly derived from research implementing intracellular recordings of muscle end plate potentials. End plate potential recordings measure properties of synaptic neurotransmission but are biased towards lower motor neurons and neuromuscular physiology. End plate potential recordings across multiple different model organism species have revealed that evoked neurotransmitter release is often decreased in amyotrophic lateral sclerosis. There currently are no assays to measure neurotransmission properties of central neurons of the corticomotor system. As such, basal neurotransmitter release from axons of upper motor neurons remains unknown. Network-, neuron- and synapse-level changes must always be considered in the context of homeostatic adaptions, which occur to stabilize outputs based on global and local set points.