This paper confirms recent studies that demonstrate a decrease in AMPA receptor activity as a consequence of exposure to Aβ peptides, but it is more than just confirmatory. The earlier studies employed exogenous Aβ at relatively high concentrations, experiments that are always open to question. This new work suggests that endogenous Aβ is the likely agent responsible for the decrease in synaptic transmission. Their use of a mutant APP incapable of generating Aβ is a new approach that has great potential for further studies.

View all comments by Vincent Marchesi

Ting et al. provide an interesting and well-done analysis of how endogenous Abeta may depress synaptic transmission, namely by depressing AMPA receptor-mediated EPSCs. Also, the authors find subtle presynaptic deficits in synaptic vesicle cycling with unknown consequences for synaptic communication. The key here is the possibility that cellularly derived Abeta may be causing these effects, thereby bypassing problems related to Abeta concentration or Abeta conformation typically associated with exogenously applied Abeta. It will eventually be useful to know the specific types of Abeta that are responsible for this phenomenon.

Several groups have demonstrated that synaptic activity can regulate release of Abeta from neurons (Kamenetz et al., 2003, Cirrito et al., 2005 ). Is activity-dependent release of Abeta necessary for this phenomenon, or is Abeta release via other mechanisms sufficient to mediate the effect on AMPA receptors? These questions ultimately address whether Abeta may act...  Read more

Ting et al. provide an interesting and well-done analysis of how endogenous Abeta may depress synaptic transmission, namely by depressing AMPA receptor-mediated EPSCs. Also, the authors find subtle presynaptic deficits in synaptic vesicle cycling with unknown consequences for synaptic communication. The key here is the possibility that cellularly derived Abeta may be causing these effects, thereby bypassing problems related to Abeta concentration or Abeta conformation typically associated with exogenously applied Abeta. It will eventually be useful to know the specific types of Abeta that are responsible for this phenomenon.

Several groups have demonstrated that synaptic activity can regulate release of Abeta from neurons (Kamenetz et al., 2003, Cirrito et al., 2005 ). Is activity-dependent release of Abeta necessary for this phenomenon, or is Abeta release via other mechanisms sufficient to mediate the effect on AMPA receptors? These questions ultimately address whether Abeta may act as a negative feedback signal for synaptic transmission.

APP with a mutation at the BACE cleavage site was a very clever tool to use in these studies. As the authors note, while this vector suggests that Abeta could be a key mediator of the effects seen here, other APP cleavage products are also affected and therefore cannot be excluded.

View all comments by John Cirrito

Our PNAS study identifies deficits in synaptic transmission when APP is overexpressed in neurons. We use Semliki Forest virus to rapidly upregulate APP in autaptic (isolated microisland) cultures of hippocampal neurons, and record synaptic responses 12 to 24 hours after infection. Our finding that AMPA receptor-mediated responses are reduced in neurons overexpressing APP is consistent with a number of recent studies reporting APP- or Aβ-mediated internalization of AMPA receptors (e.g., Almeida et al., 2005; Roselli et al., 2005; Hsieh et al., 2006).

One notable difference between our study and that of Hsieh et al. is that we do not observe a decrease in NMDA receptor-mediated synaptic responses. I believe we fortuitously caught our synapses at a point predicted but not seen by Hsieh et al.—that is, after AMPA receptor removal but prior to spine retraction—by recording a few hours earlier after infection than Hsieh et al. We also identified a presynaptic deficit in synaptic vesicle recycling that has implications for neurotransmission in response to extended trains of action...  Read more

Our PNAS study identifies deficits in synaptic transmission when APP is overexpressed in neurons. We use Semliki Forest virus to rapidly upregulate APP in autaptic (isolated microisland) cultures of hippocampal neurons, and record synaptic responses 12 to 24 hours after infection. Our finding that AMPA receptor-mediated responses are reduced in neurons overexpressing APP is consistent with a number of recent studies reporting APP- or Aβ-mediated internalization of AMPA receptors (e.g., Almeida et al., 2005; Roselli et al., 2005; Hsieh et al., 2006).

One notable difference between our study and that of Hsieh et al. is that we do not observe a decrease in NMDA receptor-mediated synaptic responses. I believe we fortuitously caught our synapses at a point predicted but not seen by Hsieh et al.—that is, after AMPA receptor removal but prior to spine retraction—by recording a few hours earlier after infection than Hsieh et al. We also identified a presynaptic deficit in synaptic vesicle recycling that has implications for neurotransmission in response to extended trains of action potentials.

References:
Almeida CG, Tampellini D, Takahashi RH, Greengard P, Lin MT, Snyder EM, Gouras GK. Beta-amyloid accumulation in APP mutant neurons reduces PSD-95 and GluR1 in synapses. Neurobiol Dis. 2005 Nov;20(2):187-98. Abstract

Hsieh H, Boehm J, Sato C, Iwatsubo T, Tomita T, Sisodia S, Malinow R. AMPAR removal underlies Abeta-induced synaptic depression and dendritic spine loss. Neuron. 2006 Dec 7;52(5):831-43. Abstract

Roselli F, Tirard M, Lu J, Hutzler P, Lamberti P, Livrea P, Morabito M, Almeida OF. Soluble beta-amyloid1-40 induces NMDA-dependent degradation of postsynaptic density-95 at glutamatergic synapses. J Neurosci. 2005 Nov 30;25(48):11061-70. Abstract

View all comments by Jane Sullivan

Nicholson and Ferreira’s new findings provide an important advance in our understanding of why high cholesterol levels may increase the vulnerability of neurons to dysfunction and death in Alzheimer disease. Previous studies in my laboratory and other laboratories had shown that Aβ can damage neurons by a mechanism involving membrane-associated oxidative stress and consequent perturbation of membrane proteins involved in the maintenance of cellular calcium homeostasis. As a result, calcium levels inside the neurons may rise excessively, resulting in damage to synapses and cell death. In addition, Ferreira and colleagues had previously provided evidence that the microtubule-associated protein tau is cleaved by calcium-activated proteases to generate a 17 kDa tau fragment that may itself adversely affect neurons.

In the present study, Nicholson and Ferreira manipulated and measured levels of cholesterol, tau proteolysis, and intracellular calcium levels in cultured hippocampal neurons exposed to Aβ. They found that when cholesterol levels are elevated, the neurons are more...  Read more

Nicholson and Ferreira’s new findings provide an important advance in our understanding of why high cholesterol levels may increase the vulnerability of neurons to dysfunction and death in Alzheimer disease. Previous studies in my laboratory and other laboratories had shown that Aβ can damage neurons by a mechanism involving membrane-associated oxidative stress and consequent perturbation of membrane proteins involved in the maintenance of cellular calcium homeostasis. As a result, calcium levels inside the neurons may rise excessively, resulting in damage to synapses and cell death. In addition, Ferreira and colleagues had previously provided evidence that the microtubule-associated protein tau is cleaved by calcium-activated proteases to generate a 17 kDa tau fragment that may itself adversely affect neurons.

In the present study, Nicholson and Ferreira manipulated and measured levels of cholesterol, tau proteolysis, and intracellular calcium levels in cultured hippocampal neurons exposed to Aβ. They found that when cholesterol levels are elevated, the neurons are more vulnerable to Aβ toxicity, and the increased vulnerability is associated with increased intracellular calcium levels and increased production of the 17 kDa tau fragment. Our previous findings indicated that the increased oxidative stress that occurs during aging and AD may itself perturb membrane cholesterol and sphingomyelin metabolism (Cutler et al., 2004). Collectively, the available data suggest that a self-propagating neurodegenerative cascade may occur in AD in which oxidative stress perturbs lipid metabolism, resulting in impaired calcium regulation and pathological changes in tau which, in turn, exacerbate oxidative stress and further disrupt cellular calcium homeostasis. Lowering cholesterol levels through dietary modifications, exercise, and statins may, therefore, stabilize cellular calcium homeostasis and so protect neurons against dysfunction and degeneration in aging and AD.

View all comments by Mark Mattson