Abstract
Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease in which the upper and lower motor neurons die. Most studies aimed at elucidating the cause of this disease have focused on the motor neuron cell body. However, recent work has suggested that the disease may begin in motor nerve terminals. The experiments described in Chapters 2-4 of this dissertation studied functional defects in motor nerve terminals of mice expressing mutant human superoxide dismutase 1 (SOD1-G93A, SOD1-G85R), models of familial ALS. In Chapter 2, the proximal hind limbs of SOD1-G93A mice were subjected to varying durations of a tourniquet-induced ischemia/reperfusion injury to determine whether these motor terminals were more vulnerable to this stress than wild-type terminals. Confocal imaging of yellow fluorescent protein (YFP expressed in neurons) and alpha-bungarotoxin (labels acetylcholine receptors on muscle) was used to determine endplate occupancy. In the distal hind limb of SOD1-G93A/YFP terminals innervating fast type muscles (extensor digitorum longus (EDL) and plantaris) were more vulnerable to ischemia/reperfusion injury than those occupying the slow type muscle (soleus). Increased vulnerability to endplate denervation was evident in presymptomatic mice as early as 31 days old. Experiments in Chapters 3 tested whether mitochondrial handling of Ca2+ loads is altered at presymptomatic stages. These experiements used rhodamine-123 to measure depolarization of the mitochondrial membrane potential (Ψm) evoked by trains of action potentials delivered to the motor nerve in levator auris longus motor terminals. These Ψm depolarizations depended on Ca2+ entry into motor terminals and were relatively small (~1-2 mV) in wild-type terminals. Consistent with the hypothesis that reduced ability to accelerate the electron transport chain (ETC) activity results in larger stimulation-induced Ψm depolarizations, presymptomatic SOD1-G93A (maintains dismutase activity) and SOD1-G85R (lacks dismutase activity) terminals displayed ~5 times greater depolarizations than wild-type terminals. Expression of normal human SOD1 or knockout of SOD1 did not significantly alter Ψm depolarizations. In the presence of a low concentration of rotenone (inhibits complex 1 of the ETC) wild-type terminals also displayed larger Ψm depolarizations. Experiments in Chapter 4 studied stimulation-induced Ψm depolarizations in terminals of older, symptomatic SOD1-G93A and SOD1-G85R mice. These depolarizations decayed more slowly than those in wild-type terminals and incremented with successive trains. Asynchronous depolarizations that were not time linked to the stimulus train were also noted. These behaviors were attenuated when opening of the mitochondrial permeability transition pore (mPTP) was inhibited with cyclosporin A or by replacing bath Ca2+ with Sr2+. Incrementing Ψm depolarizations could be elicited in wild-type terminals when subjected to an oxidative stress (diamide-induced depletion of glutathione). These findings indicate that motor terminals in mutant SOD1 mice display functional deficits even at presymptomatic ages, and that deficits associated with mitochondrial handling of stimulation-induced Ca2+ loads increase with age and may contribute to motor terminal degeneration in mutant SOD1 mice.