CNS Drugs 2003; 17 (9): 641-652

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 641 1. Alzheimer’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 642 1.1 Overview of the Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 642 1.2 Role of Oxidative Stress in Aetiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 642 2. The Glutamatergic System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 642 2.1 Role in Learning and Memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 642 2.2 Changes in Alzheimer’s Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 643 3. Potential Therapies Directed at the Glutamatergic System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 645 3.1 The Two-Stage Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 645 3.1.1 Therapies Aimed at Glutamatergic Hyperactivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 645 3.1.2 Therapies Aimed at Glutamatergic Hypoactivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 647 4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 648 Alzheimer’s disease affects nearly 5 million Americans currently and, as a Abstract result of the baby boomer cohort, is predicted to affect 14 million Americans and 22 million persons totally worldwide in just a few decades. Alzheimer’s disease is present in nearly half of individuals aged 85 years. The main symptom of Alzheimer’s disease is a gradual loss of cognitive function. Glutamatergic neurotransmission, an important process in learning and memory, is severely disrupted in patients with Alzheimer’s disease. Loss of glutamatergic function in Alzheimer’s disease may be related to the increase in oxidative stress associated with the amyloid β-peptide that is found in the brains of individuals who have the disease. Therefore, therapeutic strategies directed at the glutamatergic system may hold promise. Therapies addressing oxidative stress induced by hyperactivity of glutamate receptors include supplementation with estrogen and antioxidants such as tocopherol (vitamin E) and acetylcysteine (N-acetylcysteine). Therapy for hypoactivity of glutamate receptors is aimed at inducing the NMDA receptor with glycine and cycloserine (D-cycloserine). Recently, memantine, an NMDA receptor antagonist that addresses the hyperactivity of these receptors, has been approved in some countries for use in Alzheimer’s disease. 642 Butterfield & Pocernich 1. Alzheimer’s Disease thought to be central to the pathogenesis of the disease.[37] Our laboratory combined these two concepts into a comprehensive model for neurodegener1.1 Overview of the Disease ation in Alzheimer’s disease based on oxidative stress associated with Aβ(1-42).[17,36,38,39] Aβ(1-42) Alzheimer’s disease is the major dementing disinduces protein oxidation, lipid peroxidation, reacorder of the elderly.[1] According to Katzman,[2] in tive oxygen species formation, stimulation of nitric the absence of any intervention, patients with oxide synthase, alteration of mitochondria and many Alzheimer’s disease will number 22 million worldother markers of oxidative stress, all of which are wide in just over 2 decades. Current estimates suginhibited by antioxidants.[17,36-40] gest that the healthcare costs in the US in the nearly In addition to the above effects, Aβ(1-42) also 5 million patients with Alzheimer’s disease apinhibits aspects of the glutamatergic system, includproach $US100 billion annually.[1,2] In the absence ing causing oxidative modification of glutamine of an effective therapy, the costs associated with 14 synthetase (the enzyme that catalyses the conversion million Americans with Alzheimer’s disease in a of glutamate, the excitatory neurotransmitter and few decades will be enormous. activator of NMDA receptors, to glutamine) and the Alzheimer’s disease is associated with three maformation of small soluble, highly neurotoxic aggrejor pathological hallmarks: loss of synapses and the gates of Aβ(1-42).[18,40-42] Moreover, glutamate presence of neurofibrillary tangles and senile plapotentiates the toxicity of Aβ peptides.[43] Conceivaques. Amyloid β-peptide (Aβ), in the form of insolbly, this effect may reflect the increased free radical uble fibril deposits, is the major component of senile production observed following NMDA receptor plaques.[3,4] Senile plaques are surrounded by degenstimulation.[44] Thus, although glutamate serves an erated neurons,[1] and Aβ is toxic to neurons in important role in neurotransmission, excess glutaculture.[5-19] Genetic studies of familial Alzheimer’s mate-induced receptor stimulation can be toxic. disease offer the strongest evidence for a central role The involvement of glutamatergic pathways in of Aβ in the pathogenesis of the disease.[20] Several Alzheimer’s disease, the potential role played by familial Alzheimer’s disease mutations have been Aβ(1-42) in the alterations in the glutamatergic sysfound in the amyloid precursor protein (APP) and tem seen in the disease and possible therapies aimed presenilin genes; these mutations invariably lead to at the glutamatergic system are addressed in this increased Aβ deposition.[20,21] APP is expressed on review. chromosome 21, as is Down’s trisomy, and persons with Down’s syndrome have increased Aβ depo2. The Glutamatergic System sits[22] and develop Alzheimer’s disease eventually. APP-overexpressing mice exhibit some characteristics of Alzheimer’s disease pathology.[23-32] Increas2.1 Role in Learning and Memory ing evidence suggests that the toxic species of the Activation of NMDA receptors in different ways 42-mer Aβ(1-42) is small aggregates of the pepcan lead to either long-term potentiation (LTP) or tide.[33-35] long-term depression of synaptic strength. These forms of synaptic plasticity may represent ways of 1.2 Role of Oxidative Stress in Aetiology encoding memories in the brain. The size and nature The Alzheimer’s disease brain is characterised by of the changes in synaptic strength are highly reguextensive oxidative stress manifested by increased lated processes in learning and memory. The protein oxidation, lipid peroxidation, free radical strength of the synapse can be altered in several formation and DNA and RNA oxidation.[36] The Aβ ways. The end result can be affected by the – particularly Aβ(1-42), known to accumulate in the probability of transmitter release from an activated brains of individuals with Alzheimer’s disease – is presynaptic terminal, a change in the number of  Adis Data Information BV 2003. All rights reserved. CNS Drugs 2003; 17 (9) Glutamatergic System and Alzheimer’s Disease 643 receptors, a change in the size of the current proand contribute relatively little to the clearance of duced by each receptor at a postsynaptic site, a glutamate from the synaptic cleft.[52,54] change in the excitability of the dendritic membrane Glutamate can be neurotoxic through a stimulatoand/or changes in the cytoskeleton and membrane ry effect on NMDA, AMPA, kainate or group 1 trafficking that ultimately produce a new connecmetabotropic receptors (mGluR1), but selective tion.[45,46] Alzheimer’s disease is considered a synneuronal death in Alzheimer’s disease appears to be aptic failure.[47] The degree of cognitive decline in dependent primarily on NMDA receptor activapatients with Alzheimer’s disease has been correlattion.[55] Indeed, a recent study suggests that NMDA ed with synaptic loss.[48] A study of temporal and receptor activation stimulates APP processing to frontal cortical biopsies performed within an averproduce Aβ.[56] Consequently, overstimulation of age of 2–4 years of the onset of clinical Alzheimer’s the receptor by excess glutamate could lead to exdisease revealed a 25–35% decrease in the numericess Aβ(1-42) production with consequent oxidative cal density of synapses and a 15–35% decrease in stress-induced neurotoxicity.[36,38,39] In turn, Aβ sethe number of synapses per cortical neuron.[49] lectively depresses excitatory synaptic transmission onto neurons and is NMDA receptor activity dependent.[57] Nontoxic concentrations of Aβ can produce 2.2 Changes in Alzheimer’s Disease a rapid inhibition of LTP, although there is no longterm effect on normal synaptic transmission.[58] Aβ The possibility that activation of glutamate recan also inhibit NMDA receptor-mediated synaptic ceptors contributes to cell death in neurodegenerapotentials; however, results suggest that Aβ does not tive disorders such as Alzheimer’s disease has been inhibit LTP via effects on NMDA receptors but postulated. As mentioned in section 2.1, short-term rather interferes with a downstream pathway.[58] release of glutamate is involved in important proSynaptic depression from excessive Aβ could concesses such as learning and memory,[50] possibly tribute to cognitive decline during early Alzheimer’s involving synaptic protein remodelling.[46] Howdisease. ever, abnormally prolonged release of glutamate Glutamatergic neurotransmission in neocortical causes excitotoxicity and cell death and may play a regions and the hippocampus is severely disrupted role in the pathogenesis of chronic neurodegenerain Alzheimer’s disease.[59-61] In addition, a reduction tive disorders such as Alzheimer’s disease.[51] in the number of NMDA receptors is reported in No enzymes exist in the synaptic cleft to degrade individuals with the disease.[59] Once transported to glutamate; thus, this neurotransmitter must be glia, glutamate is converted to glutamine by glutacleared from the synapse by high-affinity presynapmine synthetase. The glutamine is released, taken up tic and glial transporters. Five different forms of into neurons and converted to glutamate by mitoglutamate transporters have been identified: GLAST chondrial glutaminase. Noninvasive detection of (EAAT1), GLT-1 (EAAT2), EAAC-1 (EAAT3), glutamate + glutamine (GLX) in vivo showed a EAAT4 and EAAT5. GLAST and GLT-1 are resignificant reduction in GLX in the cingulate cortex stricted to astroglial cells, with GLAST expression of patients with Alzheimer’s disease that strongly observed during early stages of development and correlates with both their cognitive and functional GLT-1 expression observed throughout maturity.[52] status.[62] Two other in vivo reports of GLX in In the adult brain, GLAST and GLT-1 are differenAlzheimer’s disease revealed no significant differtially located in astroglia.[53] In particular, levels of ence in the mid-frontal or temporoparietal GLX[63] GLT-1 expression are highest in the forebrain with but an increase in GLX in the occipital lobe[64] of moderate expression throughout the remaining patients with Alzheimer’s disease. neuroaxis, whereas expression of GLAST is highest in the cerebellum.[53] The neuronal glutamate transOne enzyme that is particularly sensitive to oxiporters EAAC1 and EAAT4 are located outside of dative stress is glutamine synthetase.[18,41,42] Novel  Adis Data Information BV 2003. All rights reserved. CNS Drugs 2003; 17 (9) 644 Butterfield & Pocernich electron paramagnetic resonance (EPR) approaches suggested that glutamine synthetase isolated and purified from the brains of individuals who had Alzheimer’s disease was more oxidised than that isolated and purified from the brains of healthy agematched individuals.[42] A recent study has shown, using proteomics, that glutamine synthetase indeed is oxidised in the brains of individuals with Alzheimer’s disease relative to control individuals.[65] Consistent with these findings, a significant decrease in glia-resident glutamine synthetase activity in the hippocampus and neocortex of those with Alzheimer’s disease has been reported.[66] A Glutamate Excitotoxicity Removal by GLT-1 and GS Disruption of Ca2+ homeostasis; free radical formation