To stop the destruction of neurons in Alzheimer disease requires an understanding of how the two protein perpetrators, amyloid-β (Aβ) and tau do their dirty work. Two papers out in the January 23 Journal of Neuroscience address that issue, with one exonerating activation of the pro-apoptotic caspases and neuronal death in cells that contain tau aggregated into neurofibrillary tangles. The other paper reports on synaptic inhibition by Aβ oligomers, now regarded as the initial insult of AD, and implicates voltage-gated calcium channels as Aβ's target.

The case for neurofibrillary tangles (NFTs) as initiators of apoptotic cell death rests mainly on circumstantial evidence. Neurons producing Aβ show caspase activation and tau cleavage, which precedes and could initiate tangle formation (Gamblin et al., 2003; Rissman et al., 2004), but some studies suggest that tangle formation is not the trigger for cell death, and may even serve a protective function (see ARF related news story) and ARF news story). Indeed, tau researchers have recently taken a page from the Aβ book, looking to tau oligomers or small aggregates as potential killers, rather than large tangles.

To look directly at the relationship among tangles, caspase activation, and neuronal death, Bradley Hyman and colleagues at the Massachusetts General Hospital in Charlestown used their imaging expertise to observe tangles and caspase activation in neurons in living mice. First author Tara Spires-Jones used multiphoton in vivo imaging to detect both NFT pathology and caspase activation in brains of living rTg4510 mice, which reversibly express human tau with the pathogenic P301L mutation. Imaging was done through a cranial window after thioflavin S was applied to stain tau tangles. Caspases were detected using fluorescent inhibitors that covalently bind only to activated enzyme. Using this method, Spires-Jones observed only a small minority of neurons with evidence of caspase activation, but saw that most of the caspase-positive cells had an NFT. Cells with NFTs were more likely to have activated caspases: in all, 6 percent of NFT-positive cells displayed caspase activation, while only 0.1 percent of tangle-negative cells were caspase-positive.

Nonetheless, the cells with activated caspase showed no signs of degenerating over 4 hours of observation, nor did they show any signs of terminal apoptosis despite activation of both initiator and executioner caspases. When tau transgene expression was turned off, cell death stopped, but caspase activation persisted. The results indicated that while aggregated tau was associated with activated caspases, activation did not lead to cell death, at least in the short term. Thus, the authors conclude, “Although NFTs may be associated with toxic phenomenon [sic], they do not appear to be coupled with acute neurodegeneration in this tauopathy model.” Another recent report also supports the idea that caspase cleavage of tau has little to do with filament formation or toxicity (Delobel et al., 2008).

The second paper, from Volker Nimmrich and colleagues at the Abbot research labs in Ludwigshafen, Germany, shows that Aβ oligomers acutely impair synaptic transmission via an inhibitory effect on calcium currents. The Abbot researchers described their Aβ oligomers, or “globulomers,” 2 years ago (see ARF related news story), and showed that, like naturally occurring oligomers, the globulomers inhibit long-term potentiation in hippocampal slices. Antibodies to the globulomers recognize epitopes in the brains of AD patients and in mice expressing human Aβ, suggesting that these globulomers are physiologically relevant. But how do they affect synaptic plasticity?

To answer that question, Nimmrich and fellow first author Christiane Grimm measured spontaneous synaptic currents in cultures of hippocampal neurons. When they added oligomers (8 nM), they saw reduced frequency of currents. The changes were not due to alterations in the cells’ excitability, since the investigators found action potentials unaffected. The effects, which were seen in both stimulatory glutamatergic and inhibitory GABAergic synapses, were only elicited by preformed oligomers, and not when monomer was added to the cells. By measuring a variety of pre- and postsynaptic currents, the researchers narrowed the locus of oligomer action to the presynaptic transmitter release machinery.

Neurotransmitter release is triggered by calcium influx into the presynaptic compartment, flowing via voltage-gated calcium channels in case of glutamate and GABA release. By using a variety of channel blockers, the researchers found that Aβ oligomers suppressed currents through P/Q-type channels, but not other types.

If Aβ oligomers are blocking calcium channels, then enhancers of those channels might offer a novel therapeutic strategy. Along these lines, the investigators show that roscovitine, an enhancer of P/Q-type calcium channel activity, partially reversed the effects of Aβ. Studies in animals will be needed to confirm if the action of P/Q channel agonists translates to improvements in cognitive defects as well.

The current study does not prove that Aβ acts directly on the calcium channel, but the speed with which the channel is modulated suggests the oligomers act either directly at the channel or on a closely linked partner. Voltage-gated calcium channels were recently identified as a target of another neurodegenerative protein, huntingtin, which activates the channels and leads to synaptic hyperactivity (see ARF related news story).

Whether the globulomers are identical to the cell-generated Aβ oligomers studied by Dominic Walsh when he was at Dennis Selkoe’s lab at Harvard Medical School (see ARF related news story), the ADDLs of Bill Klein at Northwestern University, Chicago (see ARF related news story), or the Aβ*56 of Karen Ashe and colleagues at the University of Minnesota, Minneapolis (see ARF related news story), is not clear. They appear to be 12-mers, similar to some other described oligomers. Recent reports from other labs showed that some oligomers act by modulating NMDA receptor activity (see ARF related news story). If the current study is replicated with other oligomers, that would place potential Aβ targets on both sides of the synapse.—Pat McCaffrey


  1. The Spires-Jones paper is a very interesting contribution that is pushing live imaging in more and more fascinating directions. The results in these transgenic mice certainly have a corollary in human studies where, for example, most neocortical neurofibrillary tangles in Alzheimer disease are contained within living neurons (Vickers et al., 2003) and that tangle-bearing neurons demonstrate no specific evidence of apoptotic degeneration (Woodhouse et al., 2006). This is not to say that the abnormalities in tau do not cause neurodegeneration, but that the detectable pathological hallmark may be associated with neuronal dysfunction which can be tolerated for long periods of time and may not necessarily result in cell death.


    . Direct determination of the proportion of intra- and extra-cellular neocortical neurofibrillary tangles in Alzheimer's disease. Brain Res. 2003 May 2;971(1):135-7. PubMed.

    . No difference in expression of apoptosis-related proteins and apoptotic morphology in control, pathologically aged and Alzheimer's disease cases. Neurobiol Dis. 2006 May;22(2):323-33. PubMed.

    View all comments by James Vickers
  2. Aβ1-42 globulomers (1) are synthetic soluble Aβ42 oligomers, which are interesting entities because they are relatively homogeneous. The major species runs at 48kDa by SDS-PAGE. There are minor species at 38kDa and 20kDa. A monoclonal antibody with relative specificity for globulomers, 8F5, stains in a halo-like pattern around Tg2576 plaques, and tends to stain the periphery of AD plaques. This is different from the staining pattern of the A11 antibody developed in the Glabe lab, which does not stain thioflavin S-positive plaques (2). In addition, Aβ1-42 globulomers bind to cultured neurons in a punctuate manner, similar in appearance to the pattern observed with ADDLs (Aβ-derived diffusible ligands, a different synthetic Aβ oligomer preparation), which have been shown to colocalize with dendritic spine proteins (3). However, Aβ globulomers differ from Aβ*56, an endogenous high-n Aβ oligomeric assembly of 56 kDa (4). For instance, differences reside in the electrophoretic and chromatographic profile using SDS-PAGE and gel filtration, and in their relative levels in Tg2576 mice at 2, 10, and 12 months of age. The levels of Aβ1-42 globulomers are low and approximately the same at 2 and 10 months, then rise ~fourfold at 12 months (1). These findings appear to mirror Aβ deposition occurring in Tg2576 mice (5,6). In contrast, the levels of Aβ*56 are undetectable at 2 months, and are the same at 10 and 12 months (4), except for a ~2-week interval around 12 months when Aβ*56 levels decrease as the rate of amyloid deposition increases in the brain of Tg2576 mice (6). Very importantly, the levels of Aβ*56 negatively correlate with spatial memory function in Tg2576 mice. Since the measured levels of Aβ1-42 globulomers do not correspond with poorer cognitive performance in Tg2576 mice, their relevance to memory loss is unclear, at least as can be assessed in this transgenic model of AD. Instead, the staining pattern of Aβ1-42 globulomers around plaques is reminiscent of the toxic halo that has been described in the Hyman lab (7-10).

    In this latest report, the authors demonstrate that Aβ1-42 globulomers suppress spontaneous synaptic activity in hippocampal primary neurons. They further refined their studies using pharmacological approaches and showed that in-vitro application of Aβ1-42 globulomers altered calcium influx through presynaptic P/Q-type calcium channels (Cav2.1). The globulomer-induced inhibition of calcium influx was rescued by roscovitine, recently proposed to positively regulate P/Q-type Ca2+ channels by acting on their extracellular domain, independently of its ability to inhibit the neuron-specific cyclin-dependent kinase Cdk5. These electrophysiological findings concerning globulomers could potentially explain the asynchrony of electrical activity that researchers in the Hyman lab described around plaques (8-10). Unlike ADDLs and secreted Aβ oligomers from APP-overexpressing cells (3,11,12), the punctuate staining pattern in cultured neurons suggests that the Aβ globulomers do not bind to postsynaptic spines, since the electrophysiological findings point to presynaptic P/Q channels being their site of action. This observation would not support the authors’ hypothesis that “all Aβ oligomers target the same synaptic mechanism.” On the contrary, if true, the collective data on various forms of Aβ to date indicate that they can target different synaptic elements.


    . Globular amyloid beta-peptide oligomer - a homogenous and stable neuropathological protein in Alzheimer's disease. J Neurochem. 2005 Nov;95(3):834-47. PubMed.

    . Common structure of soluble amyloid oligomers implies common mechanism of pathogenesis. Science. 2003 Apr 18;300(5618):486-9. PubMed.

    . Synaptic targeting by Alzheimer's-related amyloid beta oligomers. J Neurosci. 2004 Nov 10;24(45):10191-200. PubMed.

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

    . Age-dependent changes in brain, CSF, and plasma amyloid (beta) protein in the Tg2576 transgenic mouse model of Alzheimer's disease. J Neurosci. 2001 Jan 15;21(2):372-81. PubMed.

    . Plaque-bearing mice with reduced levels of oligomeric amyloid-beta assemblies have intact memory function. Neuroscience. 2008 Feb 6;151(3):745-9. PubMed.

    . Neurotoxic effects of thioflavin S-positive amyloid deposits in transgenic mice and Alzheimer's disease. Proc Natl Acad Sci U S A. 2002 Oct 29;99(22):13990-5. PubMed.

    . Cortical synaptic integration in vivo is disrupted by amyloid-beta plaques. J Neurosci. 2004 May 12;24(19):4535-40. PubMed.

    . Dendritic spine abnormalities in amyloid precursor protein transgenic mice demonstrated by gene transfer and intravital multiphoton microscopy. J Neurosci. 2005 Aug 3;25(31):7278-87. PubMed.

    . Impaired spine stability underlies plaque-related spine loss in an Alzheimer's disease mouse model. Am J Pathol. 2007 Oct;171(4):1304-11. PubMed.

    . Abeta oligomers induce neuronal oxidative stress through an N-methyl-D-aspartate receptor-dependent mechanism that is blocked by the Alzheimer drug memantine. J Biol Chem. 2007 Apr 13;282(15):11590-601. PubMed.

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

    View all comments by Sylvain Lesne
  3. The link among activated caspases, tangles, and death of neurons has been proposed based on snapshot types of analyses of patient brain, mouse brain, and cells. Experimental evidence for a causal relation was lacking, as most data did not surpass the “chicken-egg” level. It is unclear what is cause, consequence, or correlation.

    The Spires-Jones et al. study takes care of that, although only on a short time scale. The most recent study by the group of Michel Goedert goes in the same direction, and does so on a longer time scale (Delobel et al., 2008). Both groups conclude that caspases are unlikely to contribute to tauopathy.

    We have assessed in the past, and occasionally still do, in our different transgenic AD models how activated caspases relate to amyloid and tau pathology. We were unable to find a close correlation, also not in the p25 mice in which neurons degenerate “in droves” (Muyllaert et al., 2008). Activated caspase is also not part of the picture in our most recent “combined” Alzheimer model, i.e., bigenic APPxTau mice with amyloid and tau pathology. Moreover, in that same study, we compare with TauxGSK3 bigenic mice (biGT mice) that develop tangles in almost every forebrain neuron with aging, but without killing the neurons (Terwel D, Muyllaert D, Dewachter I, Borghgraef P., Croes S., Devijver H., Van Leuven F., Amyloid activates GSK-3 to aggravate neuronal tauopathy in bigenic mice Am J Pathol. 2008, March issue).

    The central issue in AD of “Do tangles kill neurons?,” raised again by Spires-Jones et al., can now be answered with a firm no. On the contrary, the hypothesis that “tangles are protective” gains in weight and might be true, after all! Our upcoming paper on the bigenic mice reinforces this view further.

    In summary: tangle formation in itself does not activate caspases. Contributions of activated caspases to neuronal cell death does not involve truncation of tau. Tau truncation in itself is not involved in promoting tangle formation, and tangle formation in itself is not causing cell death. Clearly, other factors, yet to be identified, must exist in the complex chain of events leading to neuronal death.

    Negative results are important to exclude “candidates” and stop hypotheses. The new data, even if “they come only out of mouse models,” should inspire us to rethink the AD problem, and invite new hypotheses and new candidates to come forward.


    . Analysis of tau phosphorylation and truncation in a mouse model of human tauopathy. Am J Pathol. 2008 Jan;172(1):123-31. PubMed.

    . Neurodegeneration and neuroinflammation in cdk5/p25-inducible mice: a model for hippocampal sclerosis and neocortical degeneration. Am J Pathol. 2008 Feb;172(2):470-85. Epub 2008 Jan 17 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Tau Roundup: Inducible Mice Accentuate Aggregation and More
  2. No Toxicity in Tau’s Tangles?
  3. SfN: Amyloid Oligomers—Not So Elusive, After All? Part 1
  4. Huntingtin Protein’s First Act: Overexciting Synapses
  5. Earliest Amyloid Aggregates Fingered As Culprits, Disrupt Synapse Function in Rats
  6. Orlando: Adding to ADDLs: Where Are They; What Might They Be Doing?
  7. Aβ Star is Born? Memory Loss in APP Mice Blamed on Oligomer
  8. Aβ Oligomers and NMDA Receptors—One Target, Two Toxicities

Paper Citations

  1. . Caspase cleavage of tau: linking amyloid and neurofibrillary tangles in Alzheimer's disease. Proc Natl Acad Sci U S A. 2003 Aug 19;100(17):10032-7. PubMed.
  2. . Caspase-cleavage of tau is an early event in Alzheimer disease tangle pathology. J Clin Invest. 2004 Jul;114(1):121-30. PubMed.
  3. . Analysis of tau phosphorylation and truncation in a mouse model of human tauopathy. Am J Pathol. 2008 Jan;172(1):123-31. PubMed.

Other Citations

  1. rTg4510 mice

Further Reading

Primary Papers

  1. . In vivo imaging reveals dissociation between caspase activation and acute neuronal death in tangle-bearing neurons. J Neurosci. 2008 Jan 23;28(4):862-7. PubMed.
  2. . Amyloid beta oligomers (A beta(1-42) globulomer) suppress spontaneous synaptic activity by inhibition of P/Q-type calcium currents. J Neurosci. 2008 Jan 23;28(4):788-97. PubMed.