Dumping toxic Aβ oligomers onto cultured neurons causes a reduction in synapse number. The assumption, long held but never fully proven, was that Aβ precipitates the destruction of synapses. Now, a study from Zachary Wills’ lab at the University of Pittsburgh proffers a different idea, namely that Aβ oligomers prevent new synapses from forming. In the October 11 Neuron, Wills and colleagues describe how Aβ binds to the Nogo receptor (NgR), a key inhibitor of synapse formation during development. In Wills’ experiments, the Aβ-NgR union kicks off a signaling pathway that shuts down T-type calcium channels, leading to persistent dysregulation of calcium in dendrites. Consequently, new dendritic spines no longer assemble, even in adult mouse brain. “NgR functions as a plasticity restrictor, and our data suggest that Aβ affects plasticity by tapping into NgR signaling,” Wills said.

  • Toxic Aβ oligomers ablate synapses on hippocampal neurons.
  • This may partly stem from inhibition of new synapses.
  • Aβ activates signaling through Nogo receptors that suppress synaptogenesis. 

The study adds to a long list of proposed Aβ receptors. Alas, none of them have drawn robust converging evidence thus far, leaving the field with little consensus on which, if any, might be relevant to Aβ species in the Alzheimer’s brain. Commenters raised questions about how relevant the Nogo receptor pathway might be in old mice or people, and whether synthetic Aβ oligomers as used here are a reasonable proxy for amyloid species present in brain.

Nonetheless, because learning and memory depend on new synapse formation, the findings could help explain why Alzheimer's disease takes an early toll on the ability to make new memories. Supporting this idea, injecting Aβ oligomers into the hippocampus inhibited learning and memory in wild-type mice but not NgR knockouts, report the authors.

The idea that Aβ might put the kibosh on synaptogenesis is not new. Recently, Jochen Herms of Ludwig-Maximilians-Universität Munich and colleagues reported that assembly of new dendritic spines, which bear new synapses, stalls in two different AD mouse models (Zou et al, 2015). However, no one had looked for a mechanism by which Aβ might interfere with spine formation.

To do that, first author Yanjun Zhao imaged spine dynamics in hippocampal slice cultures prepared from postnatal day six rat pups. He exposed the slices to Aβ-derived diffusible ligands (ADDLs), prepared from purified synthetic Aβ (Nicoll et al., 2013). In control cultures, spine density doubled in 96 hours, but cultures exposed to Aβ saw no increase. When he tracked individual spines, Zhao found Aβ did not affect spine elimination compared to control cultures. Instead, in the Aβ-treated cultures fewer new spines appeared during the imaging period.

Wills suspected a role for Nogo receptors, a family of cell surface proteins that bind myelin-associated glycoproteins. NgRs also bind APP and Aβ peptides, and previously have been implicated in APP processing, amyloid deposition, and learning deficits in AD mouse models (Zhou et al., 2011Park et al, 2006Park et al., 2006). Because Wills previously reported that NgR1 signaling inhibited the assembly of new synapses in hippocampal neurons, the researchers asked if NgR1 mediated Aβ’s actions on spine formation (Wills et al., 2012). After determining that Aβ oligomers bound to NgR1 on cells, Zhao knocked down NgR1 expression in the hippocampal neurons from the P6 rat pups. This drove up spine density, unperturbed by Aβ. In addition, while Aβ inhibited synaptic long-term potentiation in hippocampal slices from six- to nine-month-old wild-type mice, it had no such effect on slices from NgR1 knockouts.

Looking more closely into the signaling changes evoked by Aβ, Zhao found that oligomers elicited activation of Rho GTPase, and a reduction of dendritic calcium levels that were dependent on NgR1. In experiments with knockouts and pharmacological inhibitors, the researchers defined a pathway whereby Aβ binding to NgR1 on dendrites led to activation of the Rho GTPase, which in turn unleased a Rho-dependent kinase (ROCK)-mediated phosphorylation of the T-type calcium channel, CaV3.1. Phosphorylation inactivated the channel, lowered baseline calcium, and reduced the calcium transients required for dendritic spine assembly and synaptic plasticity. The researchers saw the same cascade when they used Aβ oligomers derived from 7PA2 cells, which secrete a variety of Aβ species, including oligomers. The results suggest that Aβ peptides may co-opt a developmental function of the NgR, acting as receptor agonists and suppressing spine formation.

Because learning and memory require new spines, Wills wanted to know if inhibiting spine assembly affected behavior in an animal model. They injected Aβ oligomers directly into brains of young wild-type or NgR knockout mice, and measured the animals’ performance on a novel object-recognition task. As in the slice studies, Aβ decreased spine density and calcium currents in wild-type mice and hampered their ability to recognize the novel object. NgR knockouts were unfazed.

Other researchers questioned the relevance to AD of pathways found in young mice. “The perinatal and developmentally young neurons harvested for much of these studies express a distinct population of channels and signaling systems uniquely designed for structural plasticity, harboring the ability to extend, retract, and respond to cues that aren’t fully replicated in adult, or even aged, neuronal populations. It is not yet clear from this study how much of the developmental machinery present only in the perinatal stages contribute to some of the mechanisms proposed for synapse loss in AD,” wrote Grace Stutzmann from the Rosalind Franklin University in North Chicago (see full comment below).

Yet synapse formation associated with learning does occur in adulthood. Robert Sweet, who is also at the University of Pittsburgh but was not involved in the new study, said he hopes the work will call attention to disruption of spine assembly as a potential mechanism for synapse loss and learning deficits in AD. “People have focused on loss of existing spines that occur with neuronal death. This study suggests a more subtle process by which Aβ could impair learning and memory by subtly accelerating the reduced formation of new synapses that occurs with normal aging,” he said.

It also remains to be seen if the NgR pathway triggered by synthetic Aβ or cell-derived Aβ plays a role in AD, where toxic species of Aβ may be different. Even with soluble Aβ isolated from brain, the identity of the toxic species remains a matter of debate (Jan 2017 news). Wills said he’s planning to examine if this mechanism is at play in people with AD, using phosphorylation of Cav3.1 as a signature for NgR-mediated effects. “We’ve developed phospho-specific antibodies for the 3.1 channel, and we hope to use mass spec to quantitate phosphorylation levels in postmortem tissue,” he said.—Pat McCaffrey


  1. Zhao et al. present an interesting study with a clearly presented mechanism proposing how exogenously applied ADDLs can result in synaptic decline via a reduction in Nogo-regulated spine assembly. This involves an inhibition of the low threshold, low conductance, transiently activated T-type voltage-gated calcium channel (VGCC) found in the dendrites and soma of a wide array of neurons throughout the brain. In broad terms, the focus on mechanisms of synaptic loss is strategic and exciting, and the senior author is well-versed in Nogo signaling and synapse formation. In comparison, it was unclear if there was a comparable command of the complexities of AD with regard to calcium signaling and synaptic physiology, as multiple synaptic and calcium signaling cascades that contribute to synaptic decline were omitted from consideration or discussion. The list can be rather long but includes SOCC, ER, lysosomal and mitochondrial stores, MCU, ion exchangers, etc. This was further complicated by questionable assumptions about synaptic plasticity deficits in AD models; for example, the authors presumed postsynaptic causation based on the lack of differences in the paired-pulse ratio (PPR); however, profound and early presynaptic deficits exist, including in PPR, as we and others have demonstrated across many models (e.g., Chakroborty et al., 2009; Chakroborty et al., 2012; de Wilde et al., 2016).When considering the properties of the T-type channel, which can generate inward calcium currents at rest and is highly expressed in the thalamus and other relatively resilient regions, the lack of correlation between its distribution and AD pathology is curious. Clearly, the dynamics are complex, and worthy of additional study.

    The focus on using synthetic ADDLs specifically is an additional source of interest, and one wonders what effects different or endogenous mixtures of amyloid peptide species would exert on this proposed mechanism. While there is little doubt that certain Aβ species disrupt calcium signaling, the mechanisms and conditions are unclear. For example, there is little effect of amyloid plaques on activity-dependent calcium signaling in confirmed hippocampal neurons (Briggs et al., 2013), yet multiple studies demonstrate oligomers can create Ca2+ permeable pores in membranes (e.g., Demuro et al., 2005; Ullah et al., 2015). 

    Investigating if and how Aβ peptides increase synapse elimination or reduce synapse formation is valid and motivating for the field, and studies such as Zhao et al. can provide an essential and needed framework. These and related future questions would be of enhanced interest if probed in mature neurons or in vivo, as the perinatal and developmentally young neurons harvested for much of these studies express a distinct population of channels and signaling systems uniquely designed for structural plasticity, harboring the ability to extend, retract, and respond to cues that aren’t fully replicated in adult, or even aged, neuronal populations. It is not yet clear from this study how much of the developmental machinery present only in the perinatal stages contribute to some of the mechanisms proposed for synapse loss in AD.  


    . Deviant ryanodine receptor-mediated calcium release resets synaptic homeostasis in presymptomatic 3xTg-AD mice. J Neurosci. 2009 Jul 29;29(30):9458-70. PubMed.

    . Early presynaptic and postsynaptic calcium signaling abnormalities mask underlying synaptic depression in presymptomatic Alzheimer's disease mice. J Neurosci. 2012 Jun 13;32(24):8341-53. PubMed.

    . Meta-analysis of synaptic pathology in Alzheimer's disease reveals selective molecular vesicular machinery vulnerability. Alzheimers Dement. 2016 Jun;12(6):633-44. Epub 2016 Jan 14 PubMed.

    . Beta amyloid peptide plaques fail to alter evoked neuronal calcium signals in APP/PS1 Alzheimer's disease mice. Neurobiol Aging. 2013 Jun;34(6):1632-43. PubMed.

    . Calcium dysregulation and membrane disruption as a ubiquitous neurotoxic mechanism of soluble amyloid oligomers. J Biol Chem. 2005 Apr 29;280(17):17294-300. PubMed.

    . Analyzing and Modeling the Kinetics of Amyloid Beta Pores Associated with Alzheimer's Disease Pathology. PLoS One. 2015;10(9):e0137357. Epub 2015 Sep 8 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Sweat the Small Stuff: Teeniest Aβ Oligomers Wreak Most Havoc

Paper Citations

  1. . Intraneuronal APP and extracellular Aβ independently cause dendritic spine pathology in transgenic mouse models of Alzheimer's disease. Acta Neuropathol. 2015 Jun;129(6):909-20. Epub 2015 Apr 11 PubMed.
  2. . Amyloid-β nanotubes are associated with prion protein-dependent synaptotoxicity. Nat Commun. 2013;4:2416. PubMed.
  3. . Interaction between amyloid precursor protein and Nogo receptors regulates amyloid deposition. FASEB J. 2011 Sep;25(9):3146-56. PubMed.
  4. . Subcutaneous Nogo receptor removes brain amyloid-beta and improves spatial memory in Alzheimer's transgenic mice. J Neurosci. 2006 Dec 20;26(51):13279-86. PubMed.
  5. . Alzheimer precursor protein interaction with the Nogo-66 receptor reduces amyloid-beta plaque deposition. J Neurosci. 2006 Feb 1;26(5):1386-95. PubMed.
  6. . The nogo receptor family restricts synapse number in the developing hippocampus. Neuron. 2012 Feb 9;73(3):466-81. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Amyloid Beta Peptides Block New Synapse Assembly by Nogo Receptor-Mediated Inhibition of T-Type Calcium Channels. Neuron. 2017 Oct 11;96(2):355-372.e6. PubMed.