. Virosome-based active immunization targets soluble amyloid species rather than plaques in a transgenic mouse model of Alzheimer's disease. J Mol Neurosci. 2005;27(2):157-66. PubMed.


Please login to recommend the paper.


  1. This paper reports a reduction in soluble, ELISA-assayable Aβ, in the absence of significant reductions in other pools of Aβ (insoluble, immunohistological) using a novel vaccination method based on influenza virus. The major advance in this manuscript is the demonstration that the soluble ELISA-assayable Aβ pool appears to be the most sensitive to the active immunization protocol. This pool is reduced to the greatest extent (60-75 percent) by five monthly immunizations. One assumes that the actual titer of anti-Aβ antibodies obtained was low, as the titers in immunized mice exceeded the titers in mice with control immunizations at P Interestingly, the absolute reduction in soluble Aβ by ELISA was 0.6 ng per gm tissue, while the nonsignificant reduction in the insoluble pool was almost 10 ng per gm. Thus, the absolute reduction of insoluble Aβ was much greater than soluble Aβ, even though its relative reduction to control values was less. This suggests that reductions in the soluble pool may lead to slow removal of Aβ from the insoluble pool.

    Perhaps most importantly, this paper highlights a problem in amyloid research. This regards how to relate the analyses of Aβ measured with different methods. The ELISA chemist has soluble and insoluble fractions (either formate or guanidinium miscible; it is unclear whether these truly extract everything). The histologist has compacted (thioflavin S or Congo red positive) and diffuse Aβ deposits (often the bulk of immunohistological Aβ measured with pan-Aβ antibodies). The physical chemist has monomeric, oligomeric, protofibrillar, and fibrillar Aβ. Learning how these different pools of Aβ accumulation interrelate is critical to understanding the Aβ economy in the brain.

    Another critical concern in establishing methods to detect various forms of Aβ is that these pools are likely in some form of equilibrium. The Heisenberg uncertainty principle may apply here, as the act of measuring the Aβ pools may modify their distributions (this may be particularly acute for Western analyses).

    Still, this paper clearly indicates that the soluble ELISA pool is the most readily affected. As this would also likely be the most accessible pool to freely diffusing ligands such as antibodies, this is a logical conclusion. As such, the paper forms an important contribution to the Aβ immunotherapy field.

  2. This paper reports a novel active immunization method against Aβ peptide by displaying the peptide Aβ1-16 on virosome of influenza, which confirmed previous findings that Aβ fragments devoid of T cell epitopes are able to induce antibodies that recognize whole Aβ peptide.

    The presented data show significant changes in soluble Aβ in treated mice, related to controls; however, the total amount of insoluble Aβ was also reduced significantly.
    In general, antibodies against the N-terminal region of Aβ, which is the immunodominant region of the whole peptide, recognize both soluble and insoluble Aβ.
    The paper does not show the antibody titer obtained by this immunization procedure.

    The authors mention no overactivation of microglia, which is partially induced by Fc antibodies by binding to microglia Fc receptors. The low titer of antibodies, if this is the case, may explain why the plaque load and microglia were unaffected.
    The main concern is why drastic reduction in soluble oligomers of Aβ, a form of this protein suggested to affect the cognitive functions of Tg mice, did not improve in this immunization approach.

  3. This paper is an important contribution to the still unsettled issue of immunization as a approach to AD prevention, but the comments of Morgan are equally noteworthy. He raises questions as to how Abeta preparations are measured that are not often considered in the interpretation of most published studies. He points out the difficulty of comparing materials subjected to guanidinium or formate treatments with thioflavin/congo red reactivity or other physical measurements. His point that components of native macromolecules are always assumed to be in equilibrium is a another consideration that must influence the interpretation of many studies that measure monomer/dimer/ oligomer states of Abeta or any other component.

  4. This study presents an interesting immunotherapy approach using influenza virosomes to present Aβ1-16 antigens for immunization. It appears to be targeted at soluble, oligomeric Aβ. Follow-up experiments addressing several technical questions may help advance the goal of understanding the workings of this antibody in more detail.

    For example, the paper's conclusion that the immunization achieved "highly specific antibody responses without eliciting any T cell reactivity" could be substantiated further by examining T cells in the brain and showing how they react to restimulation with Aβ peptide in culture. It would also be good to know if the antibodies used for immunization recognize full-length Aβ.

    An antibody that moves into the clinic should have supporting data clarifying whether the antibodies bind plaques, which is important for plaque-clearing. The claim that this vaccine clears Aβ oligomers is tantalizing, but the follow-up paper should show that the antibodies actually recognize Aβ oligomers.

    The study appears to have achieved a large reduction in soluble Aβ but not insoluble Aβ. That is a surprising finding. According to the authors, it may be due to the mouse model's accelerated AD pathology resulting from double transgenes for APP and PS1. The authors immunized from 2-6 months and analyzed the mice at 7.5 months, soon after the initiation of plaque deposition. These results are at variance from ours. We showed reduction of plaque burden and of insoluble Aβ in PSAPP mice in an even more accelerated model (Lemere et al., 2003). The mice we used develop plaques at 8-10 weeks of age. We immunized with full-length Aβ from 8-13 weeks and saw a substantial reduction in both insoluble Aβ40 and Aβ42. An alternative explanation is that the antibodies used in this study do not recognize a broad range of Aβ species and thus do not lead to a reduction in plaques. However, further studies characterizing exactly which species of Aβ are recognized by the antibodies would substantiate this hypothesis.


    . Evidence for peripheral clearance of cerebral Abeta protein following chronic, active Abeta immunization in PSAPP mice. Neurobiol Dis. 2003 Oct;14(1):10-8. PubMed.

Make a Comment

To make a comment you must login or register.