Amid the buzz over passive immunotherapy as a potential treatment for Alzheimer disease, the prevention chorus has swelled a tad. Researchers led by David Morgan of the University of South Florida, Tampa, have used an AD mouse model to distinguish between two explanations for how amyloid accumulates in the brain. Their findings, reported in the 15 April Journal of Neuroscience, offer strong support for a prevailing hypothesis—that plaques grow from amyloid-β “seeds” that sneak onto the scene when Aβ production exceeds clearance by small amounts over time. An unfortunate therapeutic implication of this model is that reducing Aβ production or increasing its clearance may alleviate future Aβ deposition but would not remove existing seeds. However, the new study offered a silver lining. “A little bit of prevention of Aβ deposition very early has very, very significant benefits later on. It’s going to suppress the entire disorder back in time dramatically,” Morgan told ARF. He estimated that scaling back Aβ deposition by just a year or two in its early stages could delay the appearance of dementia by five to 10 years.

According to a leading theory, which the new paper calls the “accumulation hypothesis,” the brain’s amyloid load reflects the cumulative effect of Aβ production outweighing clearance to a slight extent. With time, these small amounts of extra Aβ huddle together and seed the comparatively quicker outgrowth of fibrillar Aβ that researchers can see under a microscope. However, amyloid buildup could also be explained by an alternative “equilibrium hypothesis,” which suggests that steady-state rates of Aβ production and clearance change with age and that this imbalance determines the extent of amyloid deposition.

Support for the latter notion comes from studies showing that older people have reduced levels of Aβ-degrading proteases in AD-affected brain areas (Iwata et al., 2002; Caccamo et al., 2005). In addition, several reports have documented the rapid reappearance of Aβ plaques after amyloid load had been dramatically reduced by brain injections of LPS (Herber et al., 2004) or anti-Aβ antibody (Oddo et al., 2004 and ARF related news story). In both cases, Aβ levels “returned back in a matter of weeks to the levels they were at previously—in spite of the fact that they had been building over a period of months,” Morgan said. These observations led his team to consider that the extent of amyloid deposition might reflect disproportion between production and clearance—whether induced by age or treatment—and that Aβ levels shift back when rates of these processes are restored to equilibrium.

To distinguish the two hypotheses, first author Rachel Karlnoski and colleagues suppressed amyloid deposition in Tg2576 mice by systemic treatment with a humanized, monoclonal antibody against Aβ’s C-terminal end. (Originally developed at Rinat Neuroscience, which has since been acquired by Pfizer, this antibody [PF-04360365] is currently in Phase 2 trials for AD.) The researchers began treating the mice when they were eight months old, about a month before amyloid deposition typically begins in this strain, and continued weekly injections for six months. At that point, they stopped treatment and used immunohistochemistry and ELISA to monitor what happened to amyloid load over the next three months. If the accumulation hypothesis is correct, they reasoned, amyloid load in the antibody-treated mice would begin increasing at the same rate as in the control mice, but never catch up in absolute levels, starting from the point at which treatment was stopped. The equilibrium hypothesis, on the other hand, predicts that amyloid deposition would occur at a faster rate in the antibody-treated mice once Aβ suppression was lifted, such that their amyloid load would eventually catch up to that of the control mice.

In this study, the accumulation hypothesis came out as the hands-down winner. By histology, amyloid load in the frontal cortex dropped sharply after six months of immunotherapy, and Aβ levels remained much lower in the treated mice three months after immunizations were stopped. ELISA measurements of Aβ40 and Aβ42 backed these findings. The trends also held when the researchers stained frontal cortex for several microglial activation markers corresponding to early, intermediate, and mature stages of Aβ deposition.

All told, the findings are consistent with the idea that “there is some crystal of Aβ somewhere that is driving deposition of the Aβ peptides,” Morgan said. “Even if we whittle away at the ends…it can rapidly regrow. The rate limitation is the number of ‘seeds,’ and our treatments are not clearing the seeds out of the brain.”

Previous work in AD transgenic mice has shown that anti-Aβ immunization is most effective during the early stages of Aβ deposition and does not work well once amyloidosis is well underway (see, e.g., Das et al., 2001; Levites et al., 2006; and ARF related news story). An important piece of data from the new paper “is that even a window treatment at the right stage can delay future accumulation of pathology,” wrote Yona Levites, first author of one of the previous studies, in an e-mail to ARF. Levites, a former postdoctoral fellow in Todd Golde’s lab at Mayo Clinic in Jacksonville, Florida.

Ongoing work by Golde and colleagues aims to tease out at which stage one can achieve maximal efficacy with amyloid-reducing drugs such as γ-secretase inhibitors (see ARF related news story). In the meantime, Morgan’s group plans to address another aspect of the accumulation hypothesis—the idea that after initial seeding, amyloid deposition proceeds apace for a while but eventually reaches a plateau. To examine this, the researchers will essentially repeat the current study, but instead of seeing how amyloid load changes three months after suppression of treatment, they will wait six to eight months—“to determine whether there is some upper limit beyond which the animals cannot deposit amyloid,” Morgan said.—Esther Landhuis


  1. This is an intriguing and nice study. It tests the hypothesis whether Aβ plaques are in direct equilibrium with soluble Aβ. The results point to an accumulation model, and the given explanation for discrepancies with other well-reproducible observations is evident. It seems appropriate to speculate that plaque formation in vivo is a two-step process that involves a slow buildup of an Aβ seed (which contains oligomeric Aβ forms) and is followed by a rather fast and reversible (e.g., through immunotherapy) second step of growth to a histologically detectable and well-defined aggregate. Assuming that this is correct, it raises further questions that are extremely interesting:

    1. Why is such a small seed stable enough not to be dissolved upon treatment with antibodies?
    2. Why do at least some antibodies permanently block the conversion from a seed to a mature plaque (e.g., Meyer-Luehmann et al., 2006?
    3. Is it possible to isolate such “core seeds” and will they still function as such when transferred to another host animal?
    4. If so, what other components are contained in such a “core seed”?


    . Exogenous induction of cerebral beta-amyloidogenesis is governed by agent and host. Science. 2006 Sep 22;313(5794):1781-4. PubMed.

    View all comments by Frank Baumann
  2. This very important paper confirms the hypothesis that amyloid deposition is driven by accumulation of insoluble amyloid, probably following a seeding stage. It has been extensively shown previously by us and others that if you start an immunization paradigm early on in a mouse model, before deposition begins, you can prevent the deposition. Treatment is much less effective in clearing existing pathology. An important point observed in this paper is that even a window treatment at the right stage can delay future accumulation of pathology. It would be interesting to see, and is currently work in progress in our group, which stage is critical, i.e., which window treatment results in the most efficacious reduction of pathology: early, “seeding,” or later, “accumulation and plaque formation” stages.

    View all comments by Yona Levites
  3. To our knowledge, this study from David Morgan’s group provides the first published data showing that passive immunization during a restricted time window can provide a benefit that outlasts treatment for a considerable period of time (several months in mice). These data complement findings that Todd Golde, Pritam Das, and colleagues reported in Alzforum. Using the experimental γ-secretase inhibitor LY411575 to reduce Aβ production in Tg2576 during three different time windows (4M-7M, before plaques begin to accumulate; 7M-10M, when plaques are accumulating; 12M-15M, when plaques continue to accumulate at an accelerated rate), Golde’s group found that three months of drug treatment conferred lasting protection only when the drug was administered to the youngest age group, before plaques have accumulated.

    Taken together with the findings of Golde’s group, Morgan’s results could have important therapeutic implications. Rapid plaque accumulation appears to follow a slow “seeding” phase, and a growing body of evidence suggests that plaque-reducing treatments are most effective when administered before or during the seeding phase, but that it is much more difficult to slow or reverse plaque accumulation once seeds are formed (Das et al., 2001; Jankowsky et al., 2005). The results of the study by Karlnoski et al. are consistent with this conclusion, but since these authors only looked at the effects of antibody treatment initiated relatively early (8M of age), we do not know the extent or duration of the efficacy of their treatment regimen if initiated later. In addition, plaque burden in the current study was quantified at only three time points, 8M, 14M, and 17M of age. With only three time points per group in this study, it is difficult to get a sense of the shape of the curves describing the kinetics of plaque accumulation (That is, is there a six-month shift to the right of the curves of plaque burden versus time, or do the curves change shape? The limited data suggest the latter.) Given a longer time after cessation of antibody treatment, would the anti-Aβ group catch up to the control group?

    Perhaps the most important question yet to be answered is whether slowing plaque accumulation will also slow cognitive decline. The dissociation between amyloid burden and cognitive dysfunction in human AD and APP transgenic mice is well known; Aβ immunotherapy has been reported to reduce plaque burden without affecting cognitive function and vice versa (recently reviewed by Brody and Holtzman, 2008; Holmes et al., 2008). In APP transgenic mice, which we believe to be models of preclinical AD, soluble Aβ oligomers are associated with deficits in memory function (Lesné et al., 2006; Cheng et al., 2007; Meilandt et al., 2009). What is the effect of passive Aβ immunotherapy on soluble Aβ oligomers thought to mediate cognitive dysfunction? If Aβ immunotherapy also reduces levels of these soluble oligomers, when is treatment most effective? Is there a period during AD progression when plaques themselves mediate cognitive dysfunction, perhaps through local effects on neurite architecture? In AD patients, cognitive dysfunction might be mediated by different mechanisms at different stages of disease progression; while Aβ oligomers might initiate the disease, cognitive dysfunction at later stages might be mediated primarily by pathological species of tau, which are unaffected by Aβ immunotherapy.

    The results from Morgan’s and Golde’s groups confirm that any particular therapy is likely to be effective during a restricted phase of disease progression, and they highlight the importance of understanding which molecules are toxic at different stages of AD.


    . Active and passive immunotherapy for neurodegenerative disorders. Annu Rev Neurosci. 2008;31:175-93. PubMed.

    . Accelerating amyloid-beta fibrillization reduces oligomer levels and functional deficits in Alzheimer disease mouse models. J Biol Chem. 2007 Aug 17;282(33):23818-28. PubMed.

    . Reduced effectiveness of Abeta1-42 immunization in APP transgenic mice with significant amyloid deposition. Neurobiol Aging. 2001 Sep-Oct;22(5):721-7. PubMed.

    . Long-term effects of Abeta42 immunisation in Alzheimer's disease: follow-up of a randomised, placebo-controlled phase I trial. Lancet. 2008 Jul 19;372(9634):216-23. PubMed.

    . Persistent amyloidosis following suppression of Abeta production in a transgenic model of Alzheimer disease. PLoS Med. 2005 Dec;2(12):e355. Epub 2005 Nov 15 PubMed.

    . A specific amyloid-beta protein assembly in the brain impairs memory. Nature. 2006 Mar 16;440(7082):352-7. PubMed.

    . Neprilysin overexpression inhibits plaque formation but fails to reduce pathogenic Abeta oligomers and associated cognitive deficits in human amyloid precursor protein transgenic mice. J Neurosci. 2009 Feb 18;29(7):1977-86. PubMed.

    View all comments by Kathleen Zahs

Make a Comment

To make a comment you must login or register.


News Citations

  1. Tackling Alzheimer’s from the Outside In
  2. Passive Aggressive—Must Antibody Therapy Start Early to Be Effective?
  3. Eibsee: Keynote on Anti-amyloid Drugs, Prevention

Paper Citations

  1. . Region-specific reduction of A beta-degrading endopeptidase, neprilysin, in mouse hippocampus upon aging. J Neurosci Res. 2002 Nov 1;70(3):493-500. PubMed.
  2. . Age- and region-dependent alterations in Abeta-degrading enzymes: implications for Abeta-induced disorders. Neurobiol Aging. 2005 May;26(5):645-54. PubMed.
  3. . Time-dependent reduction in Abeta levels after intracranial LPS administration in APP transgenic mice. Exp Neurol. 2004 Nov;190(1):245-53. PubMed.
  4. . Abeta immunotherapy leads to clearance of early, but not late, hyperphosphorylated tau aggregates via the proteasome. Neuron. 2004 Aug 5;43(3):321-32. PubMed.
  5. . Reduced effectiveness of Abeta1-42 immunization in APP transgenic mice with significant amyloid deposition. Neurobiol Aging. 2001 Sep-Oct;22(5):721-7. PubMed.
  6. . Anti-Abeta42- and anti-Abeta40-specific mAbs attenuate amyloid deposition in an Alzheimer disease mouse model. J Clin Invest. 2006 Jan;116(1):193-201. PubMed.

Other Citations

  1. Tg2576

External Citations

  1. PF-04360365

Further Reading

Primary Papers

  1. . Suppression of amyloid deposition leads to long-term reductions in Alzheimer's pathologies in Tg2576 mice. J Neurosci. 2009 Apr 15;29(15):4964-71. PubMed.