ALS as a distal axonopathy: molecular mechanisms affecting neuromuscular junction stability in the presymptomatic stages of the disease

Abstract

Amyotrophic Lateral Sclerosis (ALS) is being redefined as a distal axonopathy, in that many molecular changes influencing motor neuron degeneration occur at the neuromuscular junction (NMJ) at very early stages of the disease prior to symptom onset. A huge variety of genetic and environmental causes have been associated with ALS, and interestingly, although the cause of the disease can differ, both sporadic and familial forms of ALS show a remarkable similarity in terms of disease progression and clinical manifestation. The NMJ is a highly specialized synapse, allowing for controlled signaling between muscle and nerve necessary for skeletal muscle function. In this review we will evaluate the clinical, animal experimental and cellular/molecular evidence that supports the idea of ALS as a distal axonopathy. We will discuss the early molecular mechanisms that occur at the NMJ, which alter the functional abilities of the NMJ. Specifically, we focus on the role of axon guidance molecules on the stability of the cytoskeleton and how these molecules may directly influence the cells of the NMJ in a way that may initiate or facilitate the dismantling of the neuromuscular synapse in the presymptomatic stages of ALS.

Keywords: Amyotrophic Lateral Sclerosis; axon guidance molecules; distal axonopathy; motor neuron; neuromuscular junction; skeletal muscle; terminal Schwann cell.

Figure 1

(A) A healthy neuromuscular junction (NMJ) is a tripartite synapse composed of a motor neuron (MN) terminal synapsing onto a muscle fiber. Terminal Schwann cells (TSCs) that envelop the junction have a role in supporting and maintaining the motor neuron terminal. (B) Potential endogenous compensatory effects that could counteract NMJ instability in ALS. CD44 is upregulated by TSCs in ALS mice and is known to encourage IGF-1 production to provide trophic support to damaged muscle fibers. CD44 has also been found to colocalize with ErbB, and can amplify the NRG1-ErbB pathway with two potential outcomes (1) indirectly via paracrine stimulation of the NRG1-ErbB pathway in MN, which is involved in synapse formation/maturation and/or (2) directly via NRG1-ErbB pathway in TSCs, which is involved in survival and maintenance of TSC (vital for correct synapse function). APP (in muscle) can be cleaved (by α-secretase followed by γ-secretase) to form secreted-APP-α (sAPP-α), which can bind CRMP2 (and maintain it in a non-phosphorylated, neuroprotective state) resulting in stabilization of the neuronal cytoskeleton. sAPP-α is also known to bind SEMA3A and prevent its repulsive effects on the nerve terminal (presumably by preventing SEMA3A binding to NRP1, which leaves CRMP2 in an unphosphorylated, neuroprotective state, thus helping to stabilize the neuronal cytoskeleton). (C) Detrimental effects of non-guidance and axon guidance molecules on NMJ stability in ALS. Non-guidance molecules (in purple) such as galectin-1 (gal-1; secreted from TSCs) and amyloid precursor protein (APP; expressed by damaged muscle fibers) can both directly cause synapse degradation (see main text for further detail). Axon guidance molecules (in green) like SEMA3A (expressed by TSCs), NOGO-A (expressed by muscle fibers) and EPHA4 (expressed by motor neurons) are upregulated in ALS mice. SEMA3A and NOGO-A can influence the stability of the neuronal cytoskeleton via activation of the collapsing response mediator protein (CRMP)-pathway. SEMA3A does this via interaction with neuropilin-1 (NRP1) on the plasma membrane of MN, resulting in the phosphorylation of CRMP2. NOGO-A (in muscle) interacts with its receptor, NgR (on MN), to activate the CRMP4-RhoA kinase pathway. CRMP2 and CRMP4 can interact directly with cytoskeletal proteins (microtubulin and actin respectively) to negatively alter their dynamics. EPHA4 (expressed by MN) feeds into the RhoA-kinase pathway, which also regulates cytoskeletal dynamics. (D) Potential therapies that target specific molecular components of the NMJ to improve NMJ stability in ALS. The administration of trophic factors (such as VEGF, IGF-1, and GDNF) has been shown to have direct neuroprotective effects and/or effects on TSC survival and maintenance (Krakora et al., 2012). Administration of recombinant gal-1 (gal-1^oxd; the oxidized form) is neuroprotective; presumably through a mechanism that enhances VEGF-signaling (gal-1^oxd binds NRP1, which enhances binding of co-receptor VEGFR2). Administration of β-secretase blockers has the potential to enhance sAPP-α production (by enhancing the non-amyloidogenic cleavage pathway). sAPP-α binds to unphosphorlyated CRMP2 which may help maintain CRMP2 in an non-phosphorylated state and thus aid in the stabilization of the neuronal cytoskeleton. Neutralization strategies using antibodies directed toward NRP1 (Venkova et al., 2014) or NOGO-A (Bros-Facer et al., 2014) have the ability to inhibit these signaling pathways, probably by preventing phosphorylation of the appropriate CRMPs, and thus enhancing neuronal cytoskeleton stabilization. Fasudil, a RhoA-kinase inhibitor, can be administered to prevent EPHA4 function by inhibiting the activation of downstream mediators that usually lead to the destabilization of the neuronal cytoskeleton. The ideal treatment for ALS would therefore combine the various interventions listed above to increase the potential success of the therapeutic approach by enhancing the beneficial pathways and inhibiting the destructive pathways respectively.