26 Jun 2008

Aβ coming from cultured cells, test tubes, and mouse brains has been shown to be toxic to neurons. But what about the stuff found in human AD brain? The major premise of the amyloid hypothesis is that human Aβ is damaging neurons. If the proof of the pudding is in the eating, then savor this—in this week’s Nature Medicine online, researchers led by Dennis Selkoe at Brigham and Women’s Hospital, Boston, report that Aβ peptides—particularly dimers—isolated from Alzheimer disease brains are indeed toxic. Alzforum already described a sampling of this work when it was presented at the Society for Neuroscience annual meeting last November in San Diego, California (see ARF related news story).

First author Ganesh Shankar and colleagues at Brigham and Women’s, together with collaborators at Beaumont Hospital, Trinity College, and University College Dublin, all in Ireland, and Harvard Medical School, isolated Aβ peptides from brain tissue taken at autopsy in Boston and Dublin. As Gabrielle Strobel reported previously, the researchers found freely soluble monomers and dimers of Aβ in tissue samples taken from clinically demented patients. Samples from cognitively normal people appeared free of soluble Aβ but did contain peptides that were solubilized by treatment with the denaturing agent guanidine HCl. There are several novel aspects to the work, suggested Selkoe in an e-mail to ARF. “We isolated Aβ species directly from the brains of AD patients and found that soluble Aβ dimers are the smallest (and thus earliest) synaptotoxic assembly of Aβ, whereas monomers, larger soluble assemblies, and isolated plaque cores have very low specific activity in our assays. Antibodies directed to the free N-terminus of human Aβ are also particularly efficient at capturing the dimers and neutralizing their effects,” he wrote. The latter finding could be particularly important for companies and researchers pursuing antibody therapies for treatment of AD.

The peptides isolated from human brain inhibited long-term potentiation in mouse hippocampal slices and lowered the threshold for induction of long-term depression. The latter effect seems independent of N-methyl-D-aspartate receptors (NMDARs) but dependent on activation of metabotropic glutamate receptors (mGluRs) since mGluR antagonists impaired LTD facilitation by the Aβ isolates. NMDA receptors seem to play a role in spine loss. Aβ peptides isolated from patient tissue halved the spine density in organotypic rat hippocampal slices, and this was blocked in the presence of an NMDA receptor antagonist. “This finding is consistent with earlier data showing that Aβ can influence synaptic plasticity through various receptors, including NMDAR, mGluR, and nicotinic acetylcholine receptors,” write the authors.

Shankar and colleagues found that the major toxic Aβ species in the tissue extracts were dimers. They cannot rule out the possibility that a small molecule might be attached to the dimer and be responsible for the toxicity. However, synthetic dimers (made by oxidizing monomers with cysteine substitutions for serine at position 26) were similarly toxic, showing that dimers alone could scupper synaptic function. The researchers found no evidence of highly toxic oligomers, such as Aβ*56, that have been isolated from mouse brain tissue.

As the authors conclude, “Aβ extracted from human brain can now serve as the most pathophysiologically relevant material for further pathway analysis and for preclinical validation of agents designed to neutralize Aβ aggregates.”—Tom Fagan