As molecular research into Alzheimer disease shifts increasingly toward analysis of soluble oligomers of the amyloid-β (Aβ) peptide, scientists are looking for therapeutic strategies to interfere with these oligomers in vivo. The gold standard is still a pill that a person will be able to swallow and be sure its effect reaches across the blood-brain barrier to the brain’s beleaguered synapses, where Aβ oligomers are thought to do their dirty work. One such compound, scyllo-inositol or AZD-103, drew attention last summer when Canadian researchers described its effect on amyloid deposition and water maze performance in transgenic mice (see McLaurin et al., 2006). Now, this story took another step forward with a report by Mathew Townsend and colleagues from Brigham and Women’s Hospital in Boston and elsewhere, who tested scyllo-inositol in their own experimental systems to study the effects of Aβ oligomers.

In the December 22 Annals of Neurology, the scientists report that scyllo-inositol rescued the impairment of synaptic function in slices of wild-type mouse hippocampus that the researchers had previously described. It also restored the flagging performance of normal adult rats in a lever pressing behavioral task that the scientists had previously established to characterize the effect of Aβ oligomers in vivo. Both experiments use Aβ dimers, trimers, and tetramers released by cultured cells, not synthetic Aβ. The study acknowledges support from the Canadian biotech company Transition Therapeutics, which is testing AZD-103 for human use as an AD drug. According to the company’s website, an initial phase 1 safety study ended without side effects, a second one testing higher doses is underway, and the much larger pharmaceutical company Elan recently partnered with Transition Therapeutics to develop AZD-103.—Gabrielle Strobel


  1. This is the latest in a series of papers evaluating the efficacy of specific inositol stereoisomers to ameliorate phenotypes of Aβ pathology. It extends investigation of effects of Aβ oligomer-containing 7PA2 cell culture media on synaptic function into behavioral tasks in a rat infusion model. The observed effects appear to be due to blocking oligomers binding to neuronal membranes, which would directly affect toxicity while also improving the chances for clearance of small oligomers.

    Importantly, the investigators measure and report dose-dependent levels of AZD-103 in CSF of rats who have had either 30, 100, or 300 mg/kg/day of AZD-103 added to their drinking water. The concentrations attained are 20-50 microM, well above the single-digit microM required to neutralize the effects of the infused 7PA2 cell media. This direct demonstration of AZD-103 bioavailability is often a difficult hurdle to overcome for putative therapeutics. Of course, the pharmacodynamics and pharmacokinetics remain to be worked out, since a continuous dosing in patients is unlikely.

    The results raise a number of questions worth considering:

    1. The experiments on AZD-103 competition with antibodies to different epitopes are hard to interpret because the apparent affinity of AZD-103 for oligomers (~100 nM) is relatively low compared to that of most antibodies (~1 nM). In addition, as a consequence of the proximity effect of having multiple antibody epitopes on the oligomers and the high density of antibody on the beads, the antibodies might easily out-compete the compound for binding to oligomers.

    2. Does AZD-103-coupled epoxy-Sepharose remove oligomer activity from 7PA2 culture media (controlled with chiro-inositol-Sepharose)? The Western blots (Figs 3d and Suppl. 1E) are very light and not very convincing. There may be subpopulations of small-n oligomers that bind tightly but are minimally involved in activity against LTP.

    3. Is the continued presence of AZD-103 with 7PA2 cell culture media required for neutralization? Since its affinity is relatively low, implying exchangeability, can AZD-103 be removed by gel filtration or dialysis? If it can be removed, and activity against LTP is still blocked, this would suggest a “conversion” activity, which would amplify the utility of the compound. The observed decreases in A11 anti-oligomer reactivity could be due not to disassembly of oligomers, but to a conformational rearrangement. Previously published studies (McLaurin et al., 2000) indicate this for synthetic Aβ(1-42).

    4. There is always an issue about the extrapolation of synthetic peptide experiments to biologically produced material, hence the 7PA2 cell culture media experiments. The reverse question is also important in terms of being able to have sufficient quantities of reproducible preparations that can be characterized more completely than cell culture media. Do similar experiments performed with purified synthetic oligomers of Aβ1-42, both in the hippocampal slices and the rat infusion model, show activity, and is AZD-103 as effective at similar concentrations of Aβ and in the same size fractions (SEC) of oligomers?

    Looking to the future, it is even more important to ask the question of whether the cell culture-derived material is representative of the soluble oligomers found in humans, including controls (some nondemented people have significant Aβ), MCI, and early and late AD. Oligomers produced by a transfected overexpressing cell line may differ in important ways from the assemblies found in human AD brain or even Tg mouse brain. How do the sizes and concentrations of cell culture Aβ oligomers compare with Tg mouse and human brain and CSF material?

    The investigators now have the systems (hippocampal slices, rat ICV infusion) to potentially, after dialysis, disentangle the effects of other substances in tissue extracts. While purification may be difficult, it should be possible to differentiate the effects of the oligomers in soluble Tg mouse or human brain and CSF extracts from those of contaminants by judicious dose control and IP of Aβ and AZD-103 neutralization.


    . Inositol stereoisomers stabilize an oligomeric aggregate of Alzheimer amyloid beta peptide and inhibit abeta -induced toxicity. J Biol Chem. 2000 Jun 16;275(24):18495-502. PubMed.

    View all comments by Harry LeVine III

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Cyclohexanehexol inhibitors of Abeta aggregation prevent and reverse Alzheimer phenotype in a mouse model. Nat Med. 2006 Jul;12(7):801-8. PubMed.

External Citations

  1. Transition Therapeutics

Further Reading

No Available Further Reading

Primary Papers

  1. . Orally available compound prevents deficits in memory caused by the Alzheimer amyloid-beta oligomers. Ann Neurol. 2006 Dec;60(6):668-76. PubMed.