More Evidence that Clusters of Proteins May Drive Disease


by Jim Schnabel

December 6, 2010

What if Alzheimer’s, in yielding up its long-held secrets, also handed over the keys to Parkinson’s, Huntington’s, Creutzfeldt-Jakob, and many other diseases? What if all these diseases are driven by a single toxic structure that appears when proteins cluster together in a certain way—a structure that could be targeted with a single drug or vaccine?

That surprising possibility was a big theme among Alzheimer’s scientists gathered at the Society for Neuroscience annual meeting in San Diego in November. “It is fascinating to me that one therapy could serve against all these diseases,” said Ottavio Arancio, a researcher at Columbia University who presented data at the meeting.

Tau oligomers

The amyloid-beta (A-beta) protein in Alzheimer’s, and alpha-synuclein protein in Parkinson’s disease, are already known to form oligomers that can harm neurons in lab tests. (See “Amyloid-Beta ‘Oligomers’ May Be Link to Alzheimer’s Dementia.”)

On Nov. 16, Arancio, a pathologist at Columbia’s Taub Institute for Research on Alzheimer’s Disease and the Aging Brain, added another neuron-harming oligomer to the list. He reported for the first time that oligomers of tau protein, but not single copies of the protein, impair synaptic functions and memory-related behavior in ordinary mice.

Insoluble aggregates of tau, known as neurofibrillary tangles, are one of the two major pathological signs—along with A-beta plaques—found in the brains of Alzheimer’s patients at autopsy. Tau tangles are also found in a number of other neurodegenerative diseases, including frontotemporal dementia and dementias related to chronic head trauma.

Tau tangles don’t seem particularly toxic, yet soluble tau proteins have been reported to harm neurons in lab tests. With the shift in focus towards oligomeric protein clusters in Alzheimer’s and other diseases, it made sense to test whether tau oligomers were responsible for soluble tau’s apparent toxic effect. Arancio and his colleagues at Columbia already had developed a system for modeling memory impairments in mice, and together with researchers at a New York City biotechnology company, Oligomerix, they decided to find out whether tau oligomers could cause such impairments.

Oligomerix, which hopes to develop a vaccine against tau oligomers, supplied the clustered tau molecules, which had been harvested from the brains of Alzheimer’s patients. Arancio’s team used small, through-the-skull tubes to deliver these human oligomers directly to the memory-making hippocampus regions of ordinary mice—a technique that avoids the time, expense, and complications of genetically re-engineering the mice.

Mice that received these doses of tau oligomers soon showed significant impairments on a standard fear-memory test, in contrast to control mice that received human tau in monomer, or single-copy form. Normally, after receiving an electric shock in a certain room within a maze, a mouse will freeze with anticipatory fear if placed again in that room. “But if the mouse receives these tau oligomers, it will not freeze, because it doesn’t remember the place where it received the electric shock,” Arancio says. Analyses of hippocampal tissue from the oligomer-dosed mice revealed weakening of synapses as well as reduced levels of proteins that normally activate synapse-maintaining genes—effects which were not seen in control mice.

Arancio notes that the synapse-impairing effect in these experiments is very similar to those reported in previous experiments with A-beta oligomers, suggesting that the tau and A-beta oligomers, along with other disease-linked oligomers, are working through the same molecular mechanisms. Those mechanisms aren’t yet clear, he said, “but there are specialized portions of the cell membrane where these oligomers could bind, affecting the function of receptors that are relevant for synaptic plasticity.”

The silver bullet?

Although Oligomerix is using human tau oligomers to develop a vaccine, other researchers see this as the wrong approach. “Generally speaking, it’s not a good idea to vaccinate humans with a human protein,” molecular biologist Charles Glabe from the University of California at Irvine told meeting attendees. “It’s hard to get an immune reponse against it, and then you have the potential for autoimmune complications.” Clinical trials of vaccines targeted against A-beta monomer and fibrils have been stymied by serious immunological side effects.

Glabe, Arancio, and others believe that disease-linked oligomers such as tau and A-beta are toxic because of an abnormal type of “beta sheet” shape they all share. As Glabe reported at the meeting, his lab recently used an artificially generated, non-human sequence of 20 oligomer-forming amino acids to raise antibodies in rabbits that specifically recognized this abnormal shape. These antibodies were able to block A-beta aggregate formation and improve memory performance in transgenic Alzheimer’s-simulating mice, just as much as antibodies raised specifically against A-beta oligomers. The same antibodies also recognized oligomers of Parkinson’s-linked alpha-synuclein protein and Huntington’s disease-linked polyglutamine protein, Glabe said.

Besides having the potential to prevent multiple diseases, this potential vaccine had the advantage that it recognized only the abnormal, disease-related beta-sheet shape on oligomers, rather than shapes found on healthy proteins, added Glabe. “Even if you vaccinate with this peptide, and you get off-target immune responses, there’s nothing expressed in the human genome that cytotoxic T cells or antibodies [evoked by the vaccine] would react to. So we hope we can develop an effective vaccine that would avoid immune complications.”

How oligomers escape cellular housekeeping mechanisms and start spreading through the brain is still a mystery. But Arancio noted that A-beta in lab tests has been found to promote, or “seed” the clustering of tau, which might help explain why A-beta deposits appear and spread first during Alzheimer’s, and are followed by the appearance and spread of tau tangles. “Perhaps it’s easier for A-beta oligomers to form first, and then tau oligomers,” he said.