While the amyloid hypothesis-that aggregates of Aβ peptide are the primary pathogen in AD-drives much of Alzheimer's disease research, it remains controversial as some observations have been difficult to reconcile with it. For example, the degree of dementia correlates only weakly with the burden of amyloid plaques in the brain at the time of death. Conversely, transgenic mice carrying the disease-causing human amyloid precursor protein (AβPP) exhibit synaptic, electrophysiological, and behavioral deficits before amyloid plaques form.

One line of thought holds that freely circulating Aβ, in the form of oligomers that have not yet formed large fibrillar aggregates, are the truly toxic species. Two papers in yesterdays’ Nature go a way toward showing how small amyloid aggregates could interfere with normal cellular function or even kill cells, not just in AD, but in other neurodegenerative diseases as well.

Dominic Walsh, working with Dennis Selkoe and colleagues at Harvard Medical School and at Trinity College in Dublin, Ireland, generated human Aβ (from the disease-causing V717F mutation of APP[751]) in Chinese hamster ovary cells. First author Dominic Walsh et al. were able to show that soon after the synthesis of Aβ monomers, these single peptide species combined to form dimers, trimers, and perhaps even larger oligomers, which were subsequently secreted into the culture medium. The oligomers, but not the monomers, in this culture medium, were able to impair hippocampal long-term potentiation (LTP) when injected into rat brain. When the Aβ-secreting cells were treated with an inhibitor of the enzyme γ-secretase (which cleaves AβPP to form Aβ) the culture medium no longer contained Aβ oligomers (though it did contain monomers) and it no longer disrupted LTP. Since LTP is thought of as a mechanism of memory storage, these data support a candidate mechanism-and possible solution-for the memory deficits of AD, the authors write.

In the larger scope of degenerative disease, protein aggregates are implicated in spongiform encephalopathies, Parkinson's disease, and other neurological and non-neurological conditions. In the same issue of Nature, researchers from the University of Florence in Italy and the University of Cambridge in England show that several seemingly benign proteins not associated with any degenerative diseases can be toxic to cells in the early stages of aggregation. The researchers showed that during the early of stages of aggregation, when the protein has aggregated into small oligomers, it is highly toxic to cells in culture. Later, when highly structured fibrils have time to form, the cytotoxic effect disappears.

"Together, the [two studies] suggest that damage to cells can be caused by misfolded intermediates generated during the production of amyloid fibrils, whether or not the fibrils-or the normal proteins from which they are derived-are also toxic. The toxicity of these early aggregates depends upon some as-yet-undefined structural features, and not upon their amino acid sequence," write John Ellis and Teresa Pinheiro of the University of Warwick, in Covington, England, in an accompanying News and Views article.—Hakon Heimer


  1. Our study identifies a specific form of naturally produced human amyloid beta protein, namely stable low-n oligomers, as directly interrupting a key correlate of memory and learning in a living animal. Previous research by many scientists had linked Ab in general to interruption of neural function, but precisely which form of the protein and how that occurred under natural conditions remained obscure. We now identify a specific form of naturally secreted Ab and show directly in living, anesthetized rats that it blocks long-term potentiation in the absence of monomers, protofibrils and fibrils. Thus soluble, diffusible Ab oligomers can interrupt memory circuits in the brain.

    Finally, we use a chemical compound that inhibits the production of Aβ to lower the oligomers enough to completely prevent the synaptic interruption, while still leaving appreciable monomer levels (60 percent of normal). This supports the potential utility of modest doses of β- or γ-secretase inhibitors in the disease.

  2. This work adds to increasing evidence implicating proteins aggregates in the pathophysiology of neurodegenerative diseases.

    View all comments by Benjamin Wolozin
  3. Previous work by the labs of Klein and Krafft showed that synthetic Aβ aggregates, termed ADDLs, are highly toxic and inhibit hippocampal function. This work now elegantly shows that Aβ oligomers generated in cells are also toxic. One puzzling aspect of this work is why such oligomers would not cause disease in familial AD patients early on in life, since they are presumably being generated continually. This suggests that although the oligomers are toxic, other factors might also contribute to the ultimate neurodegenerative toxicity - perhaps factors related to the accumulation of Aβ oligomers.

    View all comments by Benjamin Wolozin
  4. Please see the following letter published at British Medical Journal that comments on this article:

    Amyloid hypothesis, synaptic function, and Alzheimer’s disease, or: Beware: the dogma is revitalized
    Alexei R. Koudinov, Natalia V. Koudinova
    BMJ online, Published 15 May 2002 [ Full Text ].

    View all comments by Alexei Koudinov
  5. This milestone paper has addressed several issues with elegant experiments, dealing with intracellular generation of Aβ peptides and its oligomerization in cellular compartments. It is now clear that the tendency of the Aβ peptides to oligomerize and form fibrils is high. Several ideas seem striking; for example, is a particular milieu required to form these fibrils inside the cellular system? Probably the answer is “yes” but we need much more data. It has also been established that synthetic peptides have less fibrillar toxicity on synaptic plasticity when compared to Aβ generated inside cells. Now the important questions that need to be addressed from several aspects are, What is the biological significance of fibril formation and is there any process inside the cell to prevent formation of these fibrils? Scientists around the globe have been trying to understand this phenomenon and have gotten important information; this will help to understand disease pathology and its progression.

    Through the results in this article, we came to know that these Aβ peptides are present in cellular compartments, particularly in recycling endosomes. These organelles help maintain the density of receptors at various synapses; therefore, the integrity of the recycling endosomes is very important. Recently, Wang et al. reported that upon activation of NMDA receptors during LTP, the calcium sensor motor molecule Myosin Vb is recruited to recycling endosomes and is responsible for the carriage of glutamate receptors to dendritic spines, hence provides a basis for LTP and, in turn, learning and memory. In the future it will be important to see if Aβ in the recycling endosomes interferes with receptor recruitment. This can open new vistas in terms of understanding disease pathophysiology.


    . Myosin Vb mobilizes recycling endosomes and AMPA receptors for postsynaptic plasticity. Cell. 2008 Oct 31;135(3):535-48. PubMed.

    View all comments by Touqeer Ahmed

Make a Comment

To make a comment you must login or register.


No Available References

Further Reading

No Available Further Reading

Primary Papers

  1. . Naturally secreted oligomers of amyloid beta protein potently inhibit hippocampal long-term potentiation in vivo. Nature. 2002 Apr 4;416(6880):535-9. PubMed.
  2. . Inherent toxicity of aggregates implies a common mechanism for protein misfolding diseases. Nature. 2002 Apr 4;416(6880):507-11. PubMed.
  3. . Medicine: danger--misfolding proteins. Nature. 2002 Apr 4;416(6880):483-4. PubMed.