. Reversing EphB2 depletion rescues cognitive functions in Alzheimer model. Nature. 2011 Jan 6;469(7328):47-52. PubMed.


Please login to recommend the paper.


Make a Comment

To make a comment you must login or register.

Comments on this content

  1. I find this paper interesting, as it reports evidence that Aβ oligomers may cause dementia by ultimately inhibiting functions of NMDA receptor, a factor that functions in memory. Importantly, the described work predicts that additional Alzheimer’s disease (AD)-linked abnormalities in NMDA receptor functions may be found. A potential reservation relates to the animal model used and its relevance to AD: Both recent (Mawuenyega et al., 2010) and older (reviewed by Robakis, 2010) evidence indicates that, in contrast to animal models based on overproduction of APP and Αβ, there is no increased production of either APP or Aβ in AD. In addition, animal models lack important hallmarks of the disease. Thus, further work will determine whether NMDA receptor abnormalities contribute to Aβ oligomers-induced dementia.


    . Decreased clearance of CNS beta-amyloid in Alzheimer's disease. Science. 2010 Dec 24;330(6012):1774. PubMed.

    . Mechanisms of AD neurodegeneration may be independent of Aβ and its derivatives. Neurobiol Aging. 2011 Mar;32(3):372-9. PubMed.

  2. Reply to comment by Nikolaos K. Robakis
    I much appreciate Nick’s comments, and fully agree that all experimental models must be put into proper perspective. In regards to our paper, this perspective should include the fact that Mother Nature herself has carried out some informative overproduction experiments in humans. Overexpression of APP in people with duplication of the APP gene or with trisomy 21 causes syndromes that closely resemble both sporadic and autosomal-dominant familial AD (FAD). What all of these conditions have in common with FAD-mutant hAPP transgenic mice is abnormally elevated cerebral levels of Aβ. In my opinion, these mice are good models not only of amyloid deposition, but also of Aβ oligomer-induced synaptic dysfunction, a feature we have taken advantage of in this and other studies. In addition, we explored the effects of pathogenic Aβ assemblies in cultures of primary neurons from wild-type rats that did not overproduce APP. The results we obtained in these cultures (depletion of EphB2 and NMDA receptors and suppression of NMDA receptor-dependent gene expression [see paper], and reversal of these effects by increased expression of EphB2 [unpublished]) were completely consistent with the results we obtained in vivo in hAPP transgenic mice. As pointed out in our paper, we and others have also found decreased levels of EphB2 in brains of humans with sporadic AD. All of these points underline the potential relevance of our findings to the human condition, which—I agree—does require further corroboration.

  3. The study by Mucke et al. supporting the role of EphB2 in Aβ oligomer binding and synaptotoxicity is interesting and has been conducted with elegant approaches using both knockdown and overexpression of EphB2 in in vitro and animal models of AD. The study first confirms the withdrawal of surface EphB2 resulting from cultured neurons exposed to Aβ oligomeric species and demonstrates that loss of EphB2 is due to an increase of the proteosomal degradation of the receptor. Additionally, they demonstrate the interaction between EphB2 and oligomers, first by using EphB2 constructs to demonstrate that, in test tubes, EphB2 binds Aβ oligomers in the extracellular fibronectin type III repeats domain, and also by doing co-immunoprecipitation experiments using cultured neurons.

    Surface EphB2 withdrawal induced by Aβ oligomers was originally reported by us (see Lacor et al., 2007 and ARF related news story). A different interpretation is made as surface EphB2, but not total EphB2, is largely decreased after six hours’ incubation with ADDLs while NR1 and NR2B decrease earlier (three hours), whereas Mucke et al. see loss over several days. Timing for NR1 loss was also in the order of hours for Paul Greengard’s studies (see Snyder et al., 2005). Mucke's study and ours also differ in the type of Aβ oligomer preparations used. Mucke uses alternately Aβ from 7PA2-conditioned media and ADDL preparations. Plus, changes in the ADDL protocol may have generated different Aβ oligomeric species. It cannot be ruled out that the extended exposure time of the Aβ oligomeric preparations in culture would generate oligomeric species that have different characteristics than what was initially added. In this regard, we recently demonstrated that addition of low-molecular-weight Aβ oligomers for a long period of time most probably trigger similar events to those induced by high-molecular-weight Aβ oligomers (MW >50kDa), as the smaller species will accumulate and cluster at synapses and ultimately also get immobilized there where it can trigger synaptotoxicity (Renner et al., 2010).

    Another difference between our study and Mucke’s is cell culture age. We conducted our experiments when synapses are fully mature, whereas experiments here might be looking at a developmental regulation of NMDAR by EphB2. While they conclude that EphB2 is responsible for NMDAR withdrawal, that prediction cannot be drawn from our study, as NMDARs are massively lost prior to EphB2 decrease.

    Co-IP of Aβ oligomers with EphB2 alongside many other receptors was also previously shown by our team (Renner et al., 2010). The co-immunoprecipitation experiments are only an indirect method to determine ligand receptor interaction. The number of potential receptors described to date, such as nAchR, PrP, and mGluR5, and their intricate regulation make it difficult to pinpoint which receptor(s) (and maybe we can expect different receptors for different oligomeric species) are the first to interact with Aβ and trigger synaptotoxicity. In fact, we have also found that competition using antibodies against the extracellular portions of NMDAR1, mGluR5, PrP, or EphB2 are unable to fully abolish the binding of ADDLs, yet prevent different aspects of synaptotoxicity. This strongly suggests that more than one receptor is involved in the Aβ oligomer-receptor complex, and that matters are complicated by different species binding different receptors. In the present study, there seems to be a contradiction between the test tube experiment and the cell experiment, for example, where it is noticeable that in one case monomers are pulled down, while in the other, remarkable amounts of oligomeric species larger than trimers are pulled down.

    The rest of the study very nicely demonstrates that EphB2 overexpression reverses specifically Aβ-triggered synaptic deficits and restores memory function originating in the dentate gyrus of an animal model of AD without a major change in Aβ expression. This report increases the evidence that Aβ oligomers are synaptotoxic ligands, and that major changes occur at synaptic levels prior to amyloid deposition, as initially proposed by Mucke's group.


    . Abeta oligomer-induced aberrations in synapse composition, shape, and density provide a molecular basis for loss of connectivity in Alzheimer's disease. J Neurosci. 2007 Jan 24;27(4):796-807. PubMed.

    . Regulation of NMDA receptor trafficking by amyloid-beta. Nat Neurosci. 2005 Aug;8(8):1051-8. PubMed.

    . Deleterious effects of amyloid beta oligomers acting as an extracellular scaffold for mGluR5. Neuron. 2010 Jun 10;66(5):739-54. PubMed.

  4. Synapse loss in the hippocampus and other brain regions is today considered the best neuropathological correlate of cognitive decline in Alzheimer’s disease. Eph receptors, the largest family of tyrosine-kinase receptors, stabilize synaptic structure, regulate glutamatergic neurotransmission, and influence synaptic plasticity and memory (Murai and Pasquale, 2004; Klein, 2009). Little attention has been paid, however, to the role of Eph receptors in Alzheimer’s disease synaptic dysfunction. EphB2 is one of the Eph receptors more intensively studied in the adult brain, where it is located in dendritic shafts in the hippocampus and cerebral cortex (Bouvier et al., 2008). In the current paper by Mucke’s group, an early reduction of EphB2 mRNA and protein levels was found in the hippocampus of transgenic hAPP mice, a result consistent with a previous report (Simon et al., 2009).

    In a series of elegant experiments, the authors next found that EphB2 depletion in non-transgenic mice by anti-EphB2 short hairpin RNA induced marked long-term potentiation (LTP) deficits similar to those seen in untreated hAPP mice, probably by impairing NMDA receptor function. Conversely, increasing EphB2 levels in the dentate gyrus of hAPP mice reversed LTP and NMDA-receptor function deficits. The increase in EphB2 expression also ameliorated learning and memory deficits in transgenic hAPP mice. As to the mechanisms involved in EphB2 depletion in hAPP mice, the authors report that Aβ oligomers bind to EphB2 and enhance its proteasomal degradation in primary neuronal cultures. It would, no doubt, be of much interest to assess in future studies the importance of this mechanism in transgenic hAPP mice.

    The study by Mucke’s group is of great value, as it points to a novel target for therapeutic approaches in Alzheimer’s disease. Interestingly, EphB2 receptor levels were also reduced in postmortem hippocampal tissue from patients at an incipient stage of Alzheimer’s disease (Simon et al., 2009). Thus, strategies to manipulate EphB2 receptors, either by increasing receptor levels or by designing drugs able to activate EphB2 signaling pathways, may help in the treatment of synaptic dysfunction associated to Alzheimer’s disease.


    . Pre-synaptic and post-synaptic localization of EphA4 and EphB2 in adult mouse forebrain. J Neurochem. 2008 Jul;106(2):682-95. PubMed.

    . Bidirectional modulation of synaptic functions by Eph/ephrin signaling. Nat Neurosci. 2009 Jan;12(1):15-20. PubMed.

    . Eph receptors, ephrins, and synaptic function. Neuroscientist. 2004 Aug;10(4):304-14. PubMed.

    . Early changes in hippocampal Eph receptors precede the onset of memory decline in mouse models of Alzheimer's disease. J Alzheimers Dis. 2009;17(4):773-86. PubMed.

  5. Reply to comment by Pascale Lacor

    We much appreciate Pascale Lacor’s interest and comments on our study. As she points out, Aβ binds to multiple proteins and has diverse effects in cell culture models, depending on a plethora of variables. We find it difficult to assess the importance of such findings until the impact of each interaction on brain function has been examined in animal models that more closely simulate the complexity of the intracerebral environment and the chronicity of the human condition. We therefore rely more heavily on the electrophysiological and behavioral analysis of animal models than on the histological or biochemical analysis of cell cultures. Because EphB2 knockdown caused deficits in NMDA receptor function and synaptic plasticity in wild-type mice, while normalizing EphB2 levels prevented such deficits in hAPP transgenic mice, we consider it likely that, at least in the presence of chronically elevated Aβ levels and in vivo, EphB2 depletion is upstream of and mediates NMDA receptor impairments.

    We agree with Dr. Lacor that methodological details could also account for some of the differences observed between pure cell-free conditions and cell cultures, and among cell cultures that differed with respect to source and days in vitro of primary neurons, preparation and concentration of Aβ, and length of Aβ exposure. It is quite possible that the mechanisms underlying acute ADDL-induced loss of NMDA receptors differ from those causing chronic Aβ-dependent NMDA receptor impairments. Additional studies are needed to assess this possibility and the other interesting points Dr. Lacor raised.