Originally, vaccination with amyloid-β (Aβ) peptides aimed to fight Alzheimer disease by enlisting the immune system to clear aggregated amyloid plaques from the brain. But there is mounting evidence that soluble forms of Aβ, and particularly toxic multimers, cause considerable neuronal damage long before the amyloid deposits form. Soluble multimers of Aβ are neurotoxic, and interfere with synaptic function and memory storage by inhibiting hippocampal long-term potentiation (LTP) (see Lambert et al., 1998 and ARF related news story). Accordingly, immunotherapy should block the early toxic action of soluble Aβ, to have any chance of slowing or reversing the cognitive and memory deficits that occur even before plaque deposition.

A new study from the labs of Michael Rowan of Trinity College in Dublin and Dennis Selkoe at Harvard Medical School shows that Aβ-targeted immunotherapy can block the action of soluble Aβ in vivo. Their paper, which appeared online in Nature Medicine on April 17, shows that treatment with antibodies or a vaccine prevents inhibition of LTP in living animals after intraventricular injection of soluble Aβ. Their demonstration that antibodies can neutralize the short-term effects of soluble Aβ in vivo may explain how immunotherapy can rapidly improve cognitive function in AD, even before significant changes in plaque deposition are seen (see ARF related news story).

The Selkoe and Rowan groups teamed up 3 years ago to investigate the effects of naturally produced soluble Aβ oligomers on LTP in rat brain. They showed that intraventricular injection of conditioned medium from CHO cells transfected with the V717F mutant form of amyloid precursor protein (APP) inhibited LTP measured in living rats (see ARF related news story). In the current report, the researchers, led this time by first author Igor Klyubin, used the same experimental system to investigate whether antibodies to Aβ could block the effect. First, they showed that two commercially available monoclonal antibodies to Aβ (6E10 or 4G8) prevented the inhibition of LTP when injected along with Aβ-containing conditioned medium. Reasoning that the antibodies probably neutralized Aβ before the mixture was administered to the rats, they tried giving the animals Aβ first, followed 10 minutes later by the antibody. This gave the same results, showing that the antibodies could neutralize Aβ in vivo.

The conditioned medium Klyubin et al. used contained several multimeric forms of Aβ that were recognized by the antibodies, including mostly monomeric Aβ, but also dimers and trimers. To determine which fraction was responsible for the inhibition of LTP, they used exclusion chromatography to separate the different size oligomers. Only the fraction enriched in dimers and trimers inhibited LTP, and its activity was blocked by the 6E10 antibody.

To see if naturally produced antibodies could provide resistance to Aβ, the researchers immunized rats with a synthetic, preaggregated mix of Aβ40/42. The immunized animals were classified as responders if they had detectable plasma antibodies reactive with soluble Aβ, or nonresponders if they had little or no detectable Aβ immunoreactivity. Both immunized and nonimmunized animals showed normal LTP. But immunized responders did not show as much inhibition of LTP after intraventricular Aβ injection as did nonresponders or nonimmunized rats. The resistance to Aβ seemed to correlate with plasma antibody titers, because the strongest responders showed the highest levels of LTP after Aβ injection. However, the protective effect of immunization was only partial and appeared transient. LTP was restored only about halfway in the strong responders, and inhibition by Aβ increased over the 3-hour test period. The authors speculate that the partial and transient effects may be due to low serum titer of antibody, or the antibody failing to enter the hippocampus in high enough levels to sustain the response.

To date, the emphasis in immunotherapy has been on clearing of excess Aβ from the brain, as measured by the disappearance of amyloid plaques. The current results suggest that immunotherapy could also be effective at blocking immediate early effects of soluble Aβ toxicity, particularly those that lead to the memory impairment seen early in AD.—Pat McCaffrey


  1. This study represents an important advance in establishing solid in vivo proof-of-concept for eventual human therapeutics directed toward soluble, neurotoxic Aβ oligomers, also known as ADDLs. This study confirms results from other studies that passive immunization with anti-Aβ antibodies can reverse memory deficits, and it lends further support for the idea that functional deficits stem from synaptic failure and signaling malfunction, rather than nerve cell death. This study nicely demonstrates that symptomatic efficacy can be achieved by active vaccination with Aβ oligomers, but these results also suggest that alternate strategies that use more sophisticated immunogens will be required to generate a robust and sustained ADDL-neutralizing immune response.

    This study builds very nicely on concepts and in vitro experiments that we published in 2001, establishing the idea that ADDL-directed immunotherapy for AD could effectively reverse memory deficits, and demonstrating that ADDL-selective polyclonal antibodies were highly effective at blocking ADDL toxicity.

    From the standpoint of experimental methodology, we note that Western blots are cut off at 16 kDA, showing only the oligomers in the dimer, trimer, and tetramer range. We have noticed that 6E10, the antibody used in this study, does not recognize larger oligomers as well as it recognizes trimers and tetramers, which may explain why these investigators have not included these larger structures in their analysis. In our recent publication (Lacor et al., 2004), we demonstrated that only larger oligomers, including a prevalent 54-kDa assembly also observed in AD brain extracts, are capable of binding to postsynaptic neuronal receptors. A lower molecular weight fraction containing monomer through tetramer did not show any binding to synaptic receptors, nor did this fraction exhibit any neurotoxicity. We are also aware of results from Karen Ashe’s lab at the University of Minnesota, made public recently at an Alzheimer symposium in Dublin, Ireland, that certain synaptic receptors associate only with 12-mers, and not with dimers, trimers, or tetramers.

    In summary, we are very enthusiastic that this study provides continuing support for the hypothesis that ADDLs trigger memory malfunction, a hypothesis to which these authors continue to contribute with highly significant studies. We continue to believe that ADDLs represent a highly attractive target for therapeutic discovery efforts.


    . Synaptic targeting by Alzheimer's-related amyloid beta oligomers. J Neurosci. 2004 Nov 10;24(45):10191-200. PubMed.

  2. Recently, Cleary et al. reported that conditioned medium from a 7PA2 cell line that stably expresses mutant APP disrupts learning behavior when microinjected intracerebroventricularly into rats. More importantly, the authors demonstrated that oligomers (trimer and dimers), but not monomers or possibly other high-molecular-weight components of Aβ42 from 7PA2 cells (for example, 55kD) disrupt learned behavior. This is very exciting data that is further supported by this latest paper from the same group. Klyubin et al. demonstrate that monoclonal antibodies to Aβ prevent the inhibition of long-term potentiation (LTP) by conditioned medium from 7PA2 cells (containing Aβ trimers and dimers). The same effect was achieved after active immunizations of the rats with Aβ40 and Aβ42 peptides (protofibrils and some fibrils) formulated into CFA/IFA. However, it looks like only half of rats produced anti-Aβ antibodies (296 +/-94 ug/ml antibodies) and only these animals were capable of partial prevention of the inhibition of LTP by conditioned medium from 7PA2 cells. I think that these results suggest that more potent immunogen(s) should be used to generate a robust antibody production in animals, and, of course, in human populations (although it is not clear what titers of such antibodies will be therapeutic).

    For example, we have found that an epitope vaccine, composed of the immunodominant B cell epitope from β amyloid and the promiscuous T cell epitope PADRE, induced high titers of antibodies specific to Aβ1-11 (but not to PADRE peptide) in wild-type and three different APP/Tg mouse models of AD (Agadjanyan et al., 2005, and papers in prep). On the contrary, in all of the mice, this epitope vaccine induced a T cell response against PADRE only, and no anti-Aβ T cell responses. Recently, we isolated polyclonal antibodies specific to anti-Aβ1-11 from APP/Tg mice vaccinated with epitope vaccine. It is likely that these antibodies will bind oligomers, since they have the same specificity as 6E10 used in the Klyubin et al. study. However, we should be cautious in binding experiments, because our anti-Aβ1-11 antibodies, as well as 6E10 monoclonal antibody, also bind very well to other species of Aβ42. Probably it will be important to have a gold standard for evaluation of polyclonal and monoclonal antibodies specific to particular forms of Aβ (fibrils, high- and low-molecular-weight oligomers, or monomers). It looks like 7PA2 cells naturally secreted only p3, monomers, dimers, and trimers, although it will be more convincing if the authors showed in Figures 1 and 3 that 6E10 did not recognize larger molecules. It will be interesting to test the binding of M71, M93, or M94 rabbit polyclonal antibodies, specific to ADDLs at low concentrations (Gong et al., 2003), to species from 7PA2 conditioned medium.


    . Natural oligomers of the amyloid-beta protein specifically disrupt cognitive function. Nat Neurosci. 2005 Jan;8(1):79-84. PubMed.

    . Amyloid beta protein immunotherapy neutralizes Abeta oligomers that disrupt synaptic plasticity in vivo. Nat Med. 2005 May;11(5):556-61. PubMed.

    . Prototype Alzheimer's disease vaccine using the immunodominant B cell epitope from beta-amyloid and promiscuous T cell epitope pan HLA DR-binding peptide. J Immunol. 2005 Feb 1;174(3):1580-6. PubMed.

    . Alzheimer's disease-affected brain: presence of oligomeric A beta ligands (ADDLs) suggests a molecular basis for reversible memory loss. Proc Natl Acad Sci U S A. 2003 Sep 2;100(18):10417-22. PubMed.

  3. Michio Hashimoto has shown significant improvement in cognition in rats given β-amyloid infusions into the cerebral ventricles, when the brain was protected with orally administered omega-3 fatty acids (DHA) (Hashimoto et al.,2005.) In addition to better cognition, she saw a reduction in markers of lipid peroxidation and neuronal apoptosis.

    Giving long-chain fish oil fatty acids for Alzheimer's is far more practical than immunotherapy, besides being cheaper. The uncertain effects of Vitamin E in established Alzheimer disease (M Sano, 1999) might be greatly improved, if long-chain synaptic essential fatty acids were given with smaller, more sensible and less vasculotoxic doses of the vitamin.

    The recent failure of high-dose vitamin E to prevent progression of mild cognitive impairment suggests that better protection of synapses against amyloid polymers may be afforded by restocking the membranes with long-chain omega-3 and omega-6 fatty acids (fish oil plus evening primrose oil), under moderate antioxidant protection.

    Not only is high-dose vitamin E alone unable to do this job, but such high doses (e.g., 2,000 mg daily) may promote vascular smooth muscle proliferation, increasing microvascular damage in the brain itself. An instructive example of what could be achieved in established AD is Frank Corrigan's effective regimen of evening primrose oil and modest vitamin E dose (50 mg daily) (Corrigan et al., 1991). This study obviously needs to be repeated, with fish oil added.


    . Chronic administration of docosahexaenoic acid ameliorates the impairment of spatial cognition learning ability in amyloid beta-infused rats. J Nutr. 2005 Mar;135(3):549-55. PubMed.

    . Essential fatty acids in Alzheimer's disease. Ann N Y Acad Sci. 1991;640:250-2. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Earliest Amyloid Aggregates Fingered As Culprits, Disrupt Synapse Function in Rats
  2. One-Shot Deal? Mice Regain Memory Day After Vaccination, Plaques Stay Put

Paper Citations

  1. . Diffusible, nonfibrillar ligands derived from Abeta1-42 are potent central nervous system neurotoxins. Proc Natl Acad Sci U S A. 1998 May 26;95(11):6448-53. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Amyloid beta protein immunotherapy neutralizes Abeta oligomers that disrupt synaptic plasticity in vivo. Nat Med. 2005 May;11(5):556-61. PubMed.