The crimes of soluble Aβ continue to grow. Already implicated in synaptic dysfunction and loss, the peptide also scrambles the brain’s wiring, a new study suggests. In the August 21 Nature Communications, researchers led by Mark Albers at Massachusetts General Hospital, Boston, report that olfactory neurons connect to the wrong targets in mice that express mutant human amyloid precursor protein (APP). This happens in the absence of any amyloid plaques, and impairs the animals’ sense of smell. The same miswiring occurs in mice that specifically express Aβ40 or Aβ42 in the olfactory system, suggesting that these peptides are to blame, the authors note. Because new olfactory neurons continually emerge and make connections in the adult brain, the data hint that olfactory circuitry could also become jumbled in elderly people who have excess brain amyloid. Intriguingly, olfactory dysfunction is a common early feature of many neurodegenerative diseases, especially Parkinson’s and Alzheimer’s (see, e.g., Murphy, 1999; Doty, 2012).

“What is compelling about this study is the magnitude of the aberration in axonal targeting,” said Daniel Wesson at Case Western Reserve University, Cleveland, Ohio. He was not involved in the study. “Based on this work, it will be critical to explore how amyloid-β and possibly other APP metabolites are involved in regulating the organization of other highly plastic neural circuits, such as the hippocampus.”

Previously, Wesson and colleagues showed that olfactory deficits in Tg2576 AD mice correlate with plaque load, indicating that amyloid pathology damages the olfactory system (see ARF related news story on Wesson et al., 2010). But what about soluble, pre-plaque forms of Aβ? Soluble Aβ dampens synaptic transmission and prunes spines in mice that express a variety of human APP mutations (see ARF related news story; ARF news story on Hsieh et al., 2006; Hsia et al., 1999; Mucke et al., 2000; and Perez-Cruz et al., 2011). Recent evidence suggests soluble Aβ harms the olfactory system as well. Researchers led by Leonardo Belluscio at the National Institute of Neurological Disorders and Stroke, Bethesda, Maryland, reported that overexpressing mutant APP in the olfactory epithelium of mice caused olfactory neurons to die off, even though plaques did not form (see ARF related news story).

Albers and colleagues turned to the olfactory system to investigate the effect of Aβ on neuronal connectivity. Mouse olfactory circuitry provides a convenient, well-understood model system for examining neuronal wiring, with many genetic tools available to study it, Albers noted. Each mouse olfactory neuron expresses only one olfactory receptor out of a suite of more than 1,000. Based on this specific receptor expression, olfactory neurons in the epithelium of the nose must extend their axons to a single correct target, or glomerulus, in the olfactory bulb. Accurate wiring is essential for the animals to correctly discriminate smells. To visualize these connections, first author Luxiang Cao crossed Tg2576 mice, which carry human APP with the Swedish mutation, with a strain in which one type of odorant receptor was fluorescently labeled. Instead of the labeled axons converging on a single glomerulus, as they do in normal mice, the fibers scattered across multiple targets, the authors found. The same thing occurred with a different olfactory receptor. In both cases, the neural connections miswired before plaques formed.

The authors then generated mice that expressed Swedish APP only in olfactory neurons. These animals never make plaques. Again, they saw miswiring. These CORMAP mice (the name stands for “conditional olfactory sensory neuron-restricted mosaic expression of APPswe and PLAP”) were less able to avoid a predator odor or find a food odor than were littermate controls, demonstrating a functional consequence of the jumbled circuitry.

These experiments could not determine whether APP or one of its metabolites was the culprit. To narrow down the field, the authors examined mice that overexpressed human wild-type APP, as well as mice carrying a synthetic APP mutant with reduced BACE1 cleavage. Both lines had normal olfactory connections. Since Swedish APP is more easily cut by BACE1, this suggested that a BACE1 cleavage product causes the faulty wiring. The most infamous products of BACE1 processing are Aβ40 and Aβ42. To directly test their effect, the authors infected one side of the mouse olfactory epithelium with a virus that expressed one of these peptides. On the infected side, neurons mistargeted in the olfactory bulb. Interestingly, the Aβ peptides were mostly expressed by non-neuronal cells, again implying that soluble, secreted Aβ does the dirty work. However, this finding does not rule out the possibility that other BACE1 cleavage products could also contribute to miswiring, the authors note.

In ongoing work, Albers is digging into the mechanism by which Aβ disrupts neural circuitry, using longitudinal multiphoton in-vivo imaging to follow the dynamics of axons over time. He will investigate how turning the APP gene on or off changes targeting of new axons, and will correlate that with behavioral outcomes, Albers told Alzforum. He also plans to examine genetically encoded indicators of calcium flux and synaptic vesicle release to look for differences in the physiology of correctly targeted and mistargeted axons, and to examine synapses by electron microscopy.

One tantalizing question is how these findings relate to human olfactory dysfunction in neurodegenerative disease. “We’re pursuing that question, but we don’t have conclusive data yet,” Albers said. One problem is that the human olfactory wiring diagram appears to be more complicated than that of the mouse, and it has not yet been fully traced. To clarify the picture, Albers has generated antibodies to human olfactory receptors, and is using them to map out normal connections in human postmortem brains. Later, he will look at whether this map is perturbed in AD brain, he said.

In addition, Albers is interested in how miswired neurons affect the next neurons in the circuit. In other words, is pathology propagated through circuits? He saw some hints of this in the current study. Dopaminergic and tufted neurons in the olfactory bulb receive inputs from olfactory neurons. In CORMAP mice, dopaminergic and tufted neurons had lower levels of key proteins compared to neurons in wild-type littermates, implying depressed neuronal activity.

Wesson suggested that another important issue will be to determine the functional consequences of miswiring. “If amyloid-β is able to modulate the formation of the olfactory sensory map, what does this entail for the processing of odors? Does it change the animal’s ability to detect odors, or to discriminate them from background smells, or both?” he asked. More precise behavioral assays might help answer this question, Wesson said. He believes that Albers’ study and Belluscio’s previous cell death work complement each other. “By studying the peripheral olfactory system, we see two new factors that could contribute to olfactory system dysfunction in AD. Both increase our understanding of the mechanisms of dysfunction,” he said.

For his part, Belluscio told Alzforum he is intrigued by the new findings. “[The paper] demonstrates very nicely how an olfactory-based model can be used to study AD,” he noted. Belluscio suggested that in the future, the human olfactory system might be useful for early diagnosis, as a way to follow the progression of AD, or as a place to test new therapeutics. “All of these things are on the horizon,” he predicted.

Gordon Sun at the University of Michigan, Ann Arbor, agreed that taking the work into humans would be a logical step. He wrote to Alzforum, “I think the findings of this study are quite promising…. I hope this discovery will generate additional interest in identifying ways to detect and treat Alzheimer's disease in its early stages.” (See full comment below.)–Madolyn Bowman Rogers


  1. I think the findings of this study are quite promising. The study demonstrates that human amyloid-β peptides can disrupt olfactory neuronal pathways in mice, even without forming the plaques that characterize Alzheimer's disease. Furthermore, these disruptions led to a measurable decline in the animals' ability to smell.

    Hopefully, the authors of the study will be taking the next step in determining whether similar changes in olfactory neurons and a resultant decline in smell are also seen in people expressing amyloid-β peptides, but not necessarily the plaques.

    I also hope that this discovery will generate additional interest in identifying ways to detect and treat Alzheimer's disease in its early stages, even before the accumulation of amyloid plaques begins.

    View all comments by Gordon H. Sun
  2. The authors demonstrate convincingly an Aβ-induced alteration of the olfactory sensory system. Among the various transgenic animal models presented, CORMAP mice might represent a valuable model for screening of potential drugs to restore the network function/connectivity.

    It would be interesting to see, in future studies, whether the observed phenotype is reversible in a tau-null background, i.e., whether tau is an important mediator of Aβ-induced changes in connectivity (1,2). Also, it might be interesting to evaluate the influence of different, potentially more amyloidogenic forms of Aβ (3,4) applying the AAV injection technique.


    . Tau reduction prevents Abeta-induced defects in axonal transport. Science. 2010 Oct 8;330(6001):198. PubMed.

    . Reducing endogenous tau ameliorates amyloid beta-induced deficits in an Alzheimer's disease mouse model. Science. 2007 May 4;316(5825):750-4. PubMed.

    . Selective hippocampal neurodegeneration in transgenic mice expressing small amounts of truncated Aβ is induced by pyroglutamate-Aβ formation. J Neurosci. 2011 Sep 7;31(36):12790-801. PubMed.

    . Prion-like behaviour and tau-dependent cytotoxicity of pyroglutamylated amyloid-β. Nature. 2012 May 31;485(7400):651-5. PubMed.

    View all comments by Stephan Schilling

Make a Comment

To make a comment you must login or register.


News Citations

  1. Blunted Sense of Smell Parallels Pathology in AD, PD
  2. Does Aβ Normally Rein in Excited Synapses?
  3. AMPA Receptors: Going, Going, Gone in Aβ-exposed Synapses, PSD95 Knockouts
  4. Wake Up and Smell the...Neurodegeneration? AD Model Causes Cell Loss

Paper Citations

  1. . Loss of olfactory function in dementing disease. Physiol Behav. 1999 Apr;66(2):177-82. PubMed.
  2. . Olfactory dysfunction in Parkinson disease. Nat Rev Neurol. 2012 Jun;8(6):329-39. PubMed.
  3. . Olfactory dysfunction correlates with amyloid-beta burden in an Alzheimer's disease mouse model. J Neurosci. 2010 Jan 13;30(2):505-14. PubMed.
  4. . AMPAR removal underlies Abeta-induced synaptic depression and dendritic spine loss. Neuron. 2006 Dec 7;52(5):831-43. PubMed.
  5. . Plaque-independent disruption of neural circuits in Alzheimer's disease mouse models. Proc Natl Acad Sci U S A. 1999 Mar 16;96(6):3228-33. PubMed.
  6. . High-level neuronal expression of abeta 1-42 in wild-type human amyloid protein precursor transgenic mice: synaptotoxicity without plaque formation. J Neurosci. 2000 Jun 1;20(11):4050-8. PubMed.
  7. . Reduced spine density in specific regions of CA1 pyramidal neurons in two transgenic mouse models of Alzheimer's disease. J Neurosci. 2011 Mar 9;31(10):3926-34. PubMed.

Other Citations

  1. Tg2576 AD mice

Further Reading

Primary Papers

  1. . Aβ alters the connectivity of olfactory neurons in the absence of amyloid plaques in vivo. Nat Commun. 2012;3:1009. PubMed.