Abbreviations used:

Alzheimer’s disease


transmissible spongiform encephalopathy

The increasing burden of Alzheimer’s disease (AD), caused by aging of the world’s population, has led many scientists and policy makers to suggest that AD will become one of the major causes of economic and health distress in the next few decades. Alzheimer’s disease predominantly affects episodic memory causing impaired cognitive function and eventually a loss of one’s identity. The pathological hallmarks of AD are extracellular plaques, formed by aggregation of amyloid-β peptides (Aβ) and other proteins, as well as intracellular neurofibrillary tangles, caused by Tau hyperphosphorylation. In addition to these pathological hallmarks, several lines of evidence indicate that soluble oligomeric forms of Aβ-peptides are likely to be one of the early toxic components in the disease. Albeit still under debate, many researchers accept the notion that increased levels of distinct forms of Aβ oligomers trigger a cascade of pathological changes that culminates with cognitive deficits and ultimately to neuronal death (Ono and Yamada 2011).

There are many aspects of the role of Aβ peptides in Alzheimer’s disease that remain poorly understood. One subject on which we still have scarce knowledge is the relationship between distinct species of Aβ oligomers and generation of amyloid plaques. Remarkably, synaptic activity seems to directly regulate the amount Aβ peptides released into the interstitial fluid; long-term changes in neuronal activity can alter the levels of extracellular Aβ and influence plaque deposition (Bero et al. 2011). These results agree with the proposal that a default mode network in the brain, defined by its functional connectivity during resting state, may be a key player in the cognitive and pathological dysfunction found in AD (Sperling et al. 2009). Indeed, the notion that Aβ deposits may be related to the level of endogenous neuronal connectivity in the resting state is enticing. Importantly, plaques are dynamic entities able to both capture aggregated forms of Aβ as well as to free some previously captured soluble oligomers (Selkoe 2011). Unfortunately, little is known about the equilibrium dynamics of Aβ deposits and its soluble forms.

A number of experiments have recently uncovered novel aspects of the dynamic relationship between Aβ and plaque formation and have done so by examining how Aβ deposits can be seeded. A manuscript published in this issue of the Journal of Neurochemistry by Rebecca Rosen et al. (2011) has further advanced this concept.

Thanks to the efforts of the Jucker and Walker laboratories, we now know that injecting diluted extracts from AD brains, or from old transgenic mice over-expressing APP with familial AD mutations, robustly accelerates plaque deposition in injected transgenic mice expressing mutated APP (Fig. 1, Meyer-Luehmann et al. 2006; Walker et al. 2006). Even peripheral inoculation of these Aβ seeds has an effect on amyloid deposition (Eisele et al. 2010), suggesting that some sort of diffusible Aβ could have an important role in plaque formation. However, most of this work has been done in transgenic mice that would eventually develop plaques when aged. Hence, whether plaque deposition is accelerated by these brain extracts, or there is de novo induction of plaque formation is still unknown. Rosen et al. (2011) have shed light on this process by performing brain extract injection experiments in a transgenic rat line that over-expresses mutated human APP with the ‘Swedish’ double mutation (K670N-M671L) in conjunction with the ‘Indiana’ mutation (V642F). These rats express increased levels of APP and distinct forms of Aβ peptides (Agca et al. 2008), but plaques are not detected in this rat line even at 30 months of age, the median-life span of the parental rat strain. These experiments show that injection of dilute brain extracts from diagnosed AD individuals containing Aβ can induce plaques in this experimental model, which otherwise would seem incapable of doing so (Fig. 1). Control brain extracts did not cause plaque-inducing activity in injected transgenic rats nor did injection of extracts from AD brain into non-transgenic rats. These data support the notion that AD brain extracts can induce Aβ deposits in a new species and also provide evidence that brain extracts containing Aβ peptides can have seeding activity in an animal model that is relatively immune to plaque deposition. This work complements another manuscript published recently by Soto and collaborators (Morales et al. 2011) showing that in a transgenic mouse line over-expressing wild-type APP, which also does not develop plaques during their life span, brain extracts from an AD individual stereotaxically injected in the hippocampus can also induce the formation of plaques. Importantly, this seeding activity appears to be diffusible after hippocampal injection, and the Aβ deposits spread from the injection site. Plaques could be observed even in the cortex as pathology progressed. Therefore, in a mouse line that does not produce plaques during their life span, Aβ extracts seem to induce de novo plaque formation.


Figure 1.  In contrast to established animal models such as APP23 transgenic mice, the homozygous APP21 transgenic rat model developed by Rosen et al. (2011) show no plaque formation during lifetime in the absence of seeding. Plaque formation can be induced by injection of tissue extract from Alzheimer’s disease patients.

Download figure to PowerPoint

These experiments raise important questions about the relationship between diffusible forms of Aβ peptides and this seeding activity. One of the possibilities is that this seeding effect is related to the protein misfolding mechanisms proposed in transmissible spongiform encephalopathy (TSEs) or prion diseases, in which a misfolded form of PrPC, PrPSc, is thought to provide a template for recruiting and misfolding PrPC in a self-perpetuating process. Although there is some experimental support for such a mechanism, it is important to stress that there is currently no evidence that AD would have an infectious profile similar to TSEs.

There are many parallels between Alzheimer’s disease and human forms of TSEs, however whether prion-like mechanisms are in place in AD still remains to be rigorously tested. For example, in transgenic mice and rats over-expressing APP the levels of distinct species of Aβ oligomers are high already (APP transgenic mice present high levels of synaptotoxic Aβ1-42; Mucke et al. 2000), and they might be borderline with respect to the levels required for triggering higher-level aggregation. Increasing the Aβ load of a particular biochemical species could then trigger Aβ deposits. Indeed, in Down’s syndrome a very large proportion of individuals develop early-onset AD, which is likely attributed to the increased gene dosage of APP in this disease (Wiseman et al. 2009), although this relationship is not yet fully understood (Choi et al. 2009).

In addition, TSEs appear to be more efficiently transmitted than the induced Aβ deposition in experimental models. For example, most attempts to accelerate cerebral β-amyloidosis in AD transgenic mice by inoculation of exogenous Aβ via multiple peripheral routes failed (Eisele et al. 2009). By contrast, TSEs are reproducibly transmitted via all these routes of infection in a process strongly influenced by glycosylphosphatidylinositol anchoring of PrP (Klingeborn et al. 2011). These findings raise the possibility that the unique glycosylphosphatidylinositol anchoring of PrP among amyloidogenic proteins might contribute to the differences in transmissibility of TSEs versus other protein misfolding diseases, especially via peripheral routes of infection (Speare et al. 2010).

Advances in understanding the process of Aβ-induced misfolding and further testing of prion-like mechanisms in AD will depend on the rigorous biochemical characterization of the seeding activity present in brain extracts. A number of initial experiments investigating the biochemical identity of the seeding activity have provided evidence in favor of Aβ peptides, likely present in multiple conformational states (Meyer-Luehmann et al. 2006; Eisele et al. 2010; Langer et al. 2011). Unfortunately, synthetic or recombinant Aβ produced with current protocols have not been able to induce formation of plaques when injected in the brains of transgenic mice, suggesting that either the correct Aβ conformation is not generated ‘in vitro’, or that other factors in the brain extracts may also have a critical role in seeding Aβ plaques. A recent attempt to biochemically identify the Aβ species involved in this process has been published, suggesting that multiple Aβ forms may be responsible for the seeding activity. Interestingly, soluble forms of Aβ seemed to be more efficient at inducing plaque formation (Langer et al. 2011), as has been observed for TSEs (Silveira et al. 2005).

Future breakthroughs will certainly come from reconstituting the identified components for the ‘in vivo’ seeding activity. Moreover, titration of seeding activity via different routes of infection is critical to begin to assess the potential public health risks (if any) for transmission of β-amyloidosis under practical circumstances. It will also be important to further understand the relationship between neuronal activity, resting brain connectivity (the default network) and diffusion of soluble Aβ species in the brain. Understanding these processes in detail at the molecular level will be required to develop strategies to restrict the diffusion of Aβ oligomers in brain pathways.


  1. Top of page
  2. Acknowledgements
  3. References

We express our gratitude to Laura Hausmann for preparation of the figure. GB was supported by the Intramural Research Program of the NIH, NIAID. MAMP is supported by PrioNet-Canada, CIHR and The Alzheimer’s Association (USA). The authors have no conflict of interest to declare.


  1. Top of page
  2. Acknowledgements
  3. References