Alzheimer's Drug Candidates Target Form of Amyloid Beta


by Jim Schnabel

November 4, 2010

One of the most popular ideas in Alzheimer’s drug development has been to make vaccines or antibodies against the amyloid-beta (A-beta) protein, whose deposits in the brain are a classic sign of the disease. The hope is that clearing A-beta from the brain early enough would stop the disease in its tracks.

But there is a potential problem with this strategy. A-beta occurs naturally in the healthy brain, and a broad immunological attack on it could bring unacceptable side effects. Such bad reactions have already occurred in clinical trials of A-beta vaccines; the first large-scale trial was shut down in 2002 after 19 people developed inflammation of the brain and the membranes around it.

To prevent that, some laboratories have sought a way to attack not all forms of A-beta but only the smaller, more toxic clumps known as A-beta oligomers. In a mark of progress, a team of researchers from Abbott Laboratories has announced its development of monoclonal antibodies that bind to unique shapes, or conformations, found on these oligomers, and thereby reverse cognitive deficits in “Alzheimer’s mice”— without binding to other forms of A-beta.

“It’s the right idea; these conformation-dependent antibodies have the potential to distinguish between pathological and benign forms of A-beta,” says Charles Glabe, a researcher at the University of California at Irvine whose lab is active in this area.

“The big issue is specificity, and it seems that this group has done a very good job at developing an antibody that is specific to oligomers,” says Cynthia Lemere, an A-beta vaccine researcher at Brigham and Women’s Hospital and Harvard Medical School.

The Abbott team’s research was led by senior investigators Heinz Hillen and Ulrich Ebert at Abbott’s R&D center in Ludwigshafen, Germany, and published in the Aug. 4 issue of the Journal of Neuroscience.

To begin with, the researchers devised an artificial oligomer—a “globulomer”—made of 12 identical fragments of A-beta. When injected into mice, it prompted their immune system B-cells to start producing antibodies against it. The researchers then sifted through the active B-cell population for the specific type, or clone, of B-cell that made the most oligomer-specific antibody.

The most promising of these antibodies, A-887755, bound tightly to oligomers from Alzheimer’s mouse and human brain tissue, without binding to individual A-beta proteins or to the long fibrils of clumped-together A-beta normally seen in Alzheimer’s brain deposits. When given to transgenic mice that overproduce A-beta and develop an Alzheimer’s-like condition, A-887755 reversed their usual memory-test impairments. Brain-tissue analysis also showed that the treated mice regained a normal number of synapse-laden neuronal input stalks, known as synaptic spines, in the memory-related hippocampus region of the brain. Thus, it appeared that the usual brain-impairing effects of excess A-beta in these mice could be reversed by targeting merely one narrow class of oligomers.

Hillen suspects that in the brain tissue of people with Alzheimer’s there are “probably a whole variety” of toxic A-beta oligomers. But apparently, all these oligomer species share a particular structural shape that makes them toxic—and it is to this special toxic shape, or “epitope,” to which A-887755 antibodies bind. “What’s important is that one antibody is sufficient to detect all of these species if you preserve the real pathogenic epitope,” its toxic shape, says Hillen.

Hillen and his colleagues already have applied for the relevant patents, and in principle, once they have optimized a monoclonal antibody for use in humans, they can begin a small “phase 1” trial to determine a safe dose range, and proceed to larger-scale trials. An Abbott spokewoman described the company’s drug development effort in this area as active, but in an “early” stage.

One immunotherapy to bind them all?

Other academic groups active in this area include the laboratory of William Klein at Northwestern University in Chicago; that of Dominic Walsh at University College, Dublin; Dennis Selkoe’s lab at Harvard, and Glabe’s lab at the University of California at Irvine.

In 2003, Glabe’s team reported producing rabbit antibodies that bound moderately sized A-beta oligomers but not small oligomers, A-beta monomers or A-beta fibrils. The antibodies were “polyclonal,” meaning that they originated from a mix of different B-cells. Since then Glabe and his colleagues have isolated single clones of such antibodies, and have obtained Alzheimer’s mouse results like those reported by the Abbott group. They described these findings at the Society for Neuroscience conference in Chicago in 2009, but have delayed publication in a journal while their licensee, Kinexis, Inc., a biotechnology company in Carlsbad, CA, develops the monoclonal antibodies for possible human testing.

Like other researchers, Glabe sees A-beta as one of at least several proteins that can trigger disease by clumping into oligomers and larger structures—which then seed the formation of new clumpings, so that the pathology spreads an infection-like manner. [See also story, "Researchers Eye Role of ‘Infectious’ Proteins in Neurodegenerative Disease."] Other examples include the alpha-synuclein protein in Parkinson’s disease, tau protein in the later stages of Alzheimer’s and in other neurodegenerative diseases, islet amyloid polypeptide in type 2 diabetes, and prion protein in Creutzfeldt-Jakob disease.

Despite having different amino-acid sequences, these proteins when they form oligomers may all develop the same toxic shape, or epitope. If so, an antibody recognizing that epitope could bind them all—and potentially treat or prevent all these age-related diseases. The Abbott researchers reported that their A-887755 antibodies don’t have such “cross-reactivity” with other oligomeric disease proteins, but Glabe’s antibodies may.

“The antibodies we’re moving forward with also recognize generic epitopes on alpha-synuclein, on tau, on islet amyloid polypeptide, and on prion peptide” in addition to those on A-beta, he says.

“It may be a one-stop shopping type of situation for various age-related diseases that involve these aggregation-prone proteins,” says Lemere. “And there are definitely other companies working on that.”

Start Early

For Alzheimer’s, at least, any drug aimed at A-beta faces a serious hurdle:  In humans, the clump-prone protein appears to do most of its damage in the years before dementia symptoms appear—so that by the time a diagnosis is made, other neuron-harming processes such as inflammation are underway, and many neurons already have been irretrievably lost. Studies suggest that clearing A-beta with immunotherapies, inhibiting its production, or even specifically inhibiting the formation of oligomers, does little or nothing for people who already have dementia.

Thus the testing of oligomer-specific as well as broader immunotherapies against Alzheimer’s may occur only in a new kind of clinical trial—essentially a prevention trial in which subjects do not yet have cognitive symptoms but show signs of A-beta buildup on brain imaging and in spinal fluid tests. “Prevention is where I really think this is going to work,” says Lemere.

“Fifty years from now,” says Hillen, “we might even be preventing the eventual accumulation of A-beta pathology by vaccinating newborns.”