Phongpreecha T, Gajera CR, Liu CC, Vijayaragavan K, Chang AL, Becker M, Fallahzadeh R, Fernandez R, Postupna N, Sherfield E, Tebaykin D, Latimer C, Shively CA, Register TC, Craft S, Montine KS, Fox EJ, Poston KL, Keene CD, Angelo M, Bendall SC, Aghaeepour N, Montine TJ. Single-synapse analyses of Alzheimer's disease implicate pathologic tau, DJ1, CD47, and ApoE. Sci Adv. 2021 Dec 17;7(51):eabk0473. Epub 2021 Dec 15 PubMed.

Recommends

Please login to recommend the paper.

Comments

  1. This work by Tom Montine’s group is incredible in the spatial resolution and identification of protein amounts and co-localization at the synapse in AD. It offers a new orthogonal window to peer into the molecular machinery of AD (and synapses in general) and has interesting findings of candidate proteins that may have mechanistic importance at the synapse. It’s amazing how the group developed and validated the technology to provide deep insights into AD processes—truly cutting-edge!

    View all comments by Randall Bateman

  2. This technique is useful and will certainly provide novel insight into synaptic mechanisms linked to neurodegeneration. To date, many of the proteomic analyses of postmortem brain have been performed on bulk tissue, restricting the ability to characterize cell-type-specific protein alterations. Moreover, it remains extremely challenging to efficiently capture neurons intact from brain for proteomic analysis, especially with single-cell resolution.

    What is impressive about this current study is that it leverages CyTOF technology, combining principles of mass spectrometry and flow cytometry, to resolve single-synaptic events in brain. The authors have termed this technique SynTOF. Thus, analogous to single-cell or single-nuclei RNA-Seq, this allowed the authors to identify unique sub-populations of synaptosomes across disease states.

    This included synaptosomes with increased hippocampal pathologic tau in AD and reduced levels of the dopamine transporters in LBD. Interestingly, they showed some hippocampal synapses with increased levels of the CD47 in AD, which is a marker that would be predicted to protect the synapse from excessive microglial pruning and degradation.

    Limitations of the approach include dependence on high-quality antibodies and that despite creating more than 100 synaptosome preparations, a relatively small number were suitable for SynTOF analyses. Better cryopreservation techniques could potentially be key to advancing this exciting approach.

    View all comments by Nicholas Seyfried

  3. Using a mass-cytometry–based method, synaptometry by time of flight (SynTOF), this study focuses on the synapse to uncover what causes AD. I generally like this elegant approach, but believe it adds an additional variable compared to more classical mass-spec approaches.

    The authors screened 166 antibodies to arrive at 38 with high specificity. Just as the utility of a model depends on the validity of the underlying assumptions, SynTOF used here depends on the quality of the antibodies employed. It seems to me that, ideally, one would use more than one antibody for a given target.

    I did like the fact that two AD-resilient cases were analyzed among the nine ADNC cases.

    I am convinced that this methodology will be widely adopted.

    View all comments by Jürgen Götz

  4. This interesting study by Montine and colleagues examines the composition of synapses using mass cytometry. It is surprising that they do not observe Aβ in human synapses from AD patient brain, because we and others have observed this using multiple methods, including high-resolution imaging and biochemical enrichment of synaptic fractions (see, for example, Koffie et al., 2012; Jackson et al., 2019; Perez-Nievas et al., 2013). Perhaps the Aβ antibodies used for their assay do not detect the oligomeric Aβ that we have observed in human synapses.

    The APOE data are also not consistent with some of our previous work using array tomography, where we saw no increase in the proportion of synapses in AD brain containing APOE, although we cannot rule out that those synapses that did contain APOE had higher levels of the apolipoprotein in our study (Koffie et al., 2012). It would be good to see how the current results differ based on APOE genotype of the donors.

    The tau synapse data are consistent with our observations, both from proteomics and high-resolution imaging (Hesse et al., 2019; Zhou et al., 2017; Pickett et al., 2019). The CD47 data are also consistent with our data using unbiased proteomics mass-spec analysis of human synapses (Hesse et al., 2019). 

    It is super interesting to see how things replicate (or don't) in studies by other groups. I'm very glad to see the growing interest in synaptic composition in AD!

    References:

    Hesse R, Hurtado ML, Jackson RJ, Eaton SL, Herrmann AG, Colom-Cadena M, Tzioras M, King D, Rose J, Tulloch J, McKenzie CA, Smith C, Henstridge CM, Lamont D, Wishart TM, Spires-Jones TL. Comparative profiling of the synaptic proteome from Alzheimer's disease patients with focus on the APOE genotype. Acta Neuropathol Commun. 2019 Dec 20;7(1):214. PubMed.

    Jackson RJ, Rose J, Tulloch J, Henstridge C, Smith C, Spires-Jones TL. Clusterin accumulates in synapses in Alzheimer's disease and is increased in apolipoprotein E4 carriers. Brain Commun. 2019;1(1):fcz003. Epub 2019 Jun 24 PubMed.

    Koffie RM, Hashimoto T, Tai HC, Kay KR, Serrano-Pozo A, Joyner D, Hou S, Kopeikina KJ, Frosch MP, Lee VM, Holtzman DM, Hyman BT, Spires-Jones TL. Apolipoprotein E4 effects in Alzheimer's disease are mediated by synaptotoxic oligomeric amyloid-β. Brain. 2012 Jul;135(Pt 7):2155-68. PubMed.

    Perez-Nievas BG, Stein TD, Tai HC, Dols-Icardo O, Scotton TC, Barroeta-Espar I, Fernandez-Carballo L, de Munain EL, Perez J, Marquie M, Serrano-Pozo A, Frosch MP, Lowe V, Parisi JE, Petersen RC, Ikonomovic MD, López OL, Klunk W, Hyman BT, Gómez-Isla T. Dissecting phenotypic traits linked to human resilience to Alzheimer's pathology. Brain. 2013 Aug;136(Pt 8):2510-26. PubMed.

    Pickett EK, Herrmann AG, McQueen J, Abt K, Dando O, Tulloch J, Jain P, Dunnett S, Sohrabi S, Fjeldstad MP, Calkin W, Murison L, Jackson RJ, Tzioras M, Stevenson A, d'Orange M, Hooley M, Davies C, Colom-Cadena M, Anton-Fernandez A, King D, Oren I, Rose J, McKenzie CA, Allison E, Smith C, Hardt O, Henstridge CM, Hardingham GE, Spires-Jones TL. Amyloid Beta and Tau Cooperate to Cause Reversible Behavioral and Transcriptional Deficits in a Model of Alzheimer's Disease. Cell Rep. 2019 Dec 10;29(11):3592-3604.e5. PubMed.

    Zhou L, McInnes J, Wierda K, Holt M, Herrmann AG, Jackson RJ, Wang YC, Swerts J, Beyens J, Miskiewicz K, Vilain S, Dewachter I, Moechars D, De Strooper B, Spires-Jones TL, De Wit J, Verstreken P. Tau association with synaptic vesicles causes presynaptic dysfunction. Nat Commun. 2017 May 11;8:15295. PubMed.

    View all comments by Tara Spires-Jones

  5. This is an interesting technology with the potential for deep interrogation of protein content in synapses and how they change with age or disease. It would be interesting to see if the technology could be adapted to interrogate other parts of the neuron, such as axons, dendrites, or organelles, including mitochondria, lysosomes, and the nucleus.

    I don’t see the results as pointing to a shortcoming of the mouse models. Postmortem human AD brain represents end-stage disease, consisting of burned-out plaques and tangles—lesions that are tombstones. Therefore, it is not surprising that the human AD synapses did not contain Aβ, because they were end-stage, highly degenerated, and dysfunctional.

    In contrast, the AD mouse models represent early, preclinical AD, wherein the pathogenic processes are active and progressing, but are not burned out, as is the case in postmortem human AD. Hence, the PS/APP mouse synapses were actively producing Aβ.

    I think this technology will be very informative for earlier-stage human neurodegenerative diseases, but we should be cautious interpreting the results from end-stage AD or other neurodegenerations. 

    View all comments by Robert Vassar

  6. This is certainly an exceptional study on synaptosomes prepared from fresh human brain. Most studies on “synaptosomes” from human tissue are done with frozen material and that does not lead to “real” synaptosomes. I am indeed fascinated by the data obtained in control patients.

    However, as a neuropathologist, I have difficulty judging the results obtained in humans with Aβ or/and α-synuclein pathology when the ApoE genotype is not given. LBD cases of old age without significant Aβ pathology are very rare—if it occurs these are often PD cases without any ApoE4 allele. So, I assume that the ApoE genotypes are significantly different between the groups compared in this study. But the authors could tell us. To me the statement comparing “pure AD” with “LBD” in the abstract is problematic since Aβ pathology is most often found in both.

    In nearly all mouse amyloid models, the synaptic pathology is restricted to the area close to amyloid plaques, but in humans with AD there is severe tau pathology and synapse loss distant to plaques (Peters et al., 2018). So, I would not expect similar results when comparing synaptosomes from mouse amyloid models and humans with AD-related pathology. The amyloid models are only models for some aspects of AD related to Aβ accumulation—indeed, not for those that are attributed to p-tau accumulation, for example.

    References:

    Peters F, Salihoglu H, Rodrigues E, Herzog E, Blume T, Filser S, Dorostkar M, Shimshek DR, Brose N, Neumann U, Herms J. BACE1 inhibition more effectively suppresses initiation than progression of β-amyloid pathology. Acta Neuropathol. 2018 May;135(5):695-710. Epub 2018 Jan 11 PubMed.

    View all comments by Jochen Herms

  7. This technique is very interesting and could be more widely applied across neurodegenerative diseases. In particular, it would be really interesting to see which changes in disease are specific to AD, and which are due to damaged synapses reacting to disease more generally. For example, I’d like to see if DJ-1 levels are lower in synapses in dopamine neurons in Parkinson’s early in disease.

    To this point, it is probably going to be really tough to work in studies of early disease in people, i.e., before frank pathology has set in, so I wonder if one interpretation of discrepancies between the human and mouse data is that the latter are due to the very earliest of events.

    View all comments by Mark Cookson

  8. I think this method has potentially important applications.

    However, it is important to note that the analyses in this study were limited to presynaptic structures. Most synaptic studies in AD models that I’m aware of, but then I’m likely biased by my area of expertise, have observed postsynaptic changes, for example in synaptic glutamate receptor number. Other types of studies on human AD brain samples have seen reductions in PSD-95, a postsynaptic protein that is important in synaptic function. As a side note, elevated PSD-95 can protect synapses from Aβ (Dore et al., 2021). 

    It may be potentially more relevant to AD to analyze postsynaptic protein composition with this type of method. Unfortunately, the synaptosome preparation used in this study is largely limited to presynaptic proteins. I don’t know of a biochemical preparation akin to synaptosomes that contains primarily postsynaptic proteins.

    References:

    Dore K, Carrico Z, Alfonso S, Marino M, Koymans K, Kessels HW, Malinow R. PSD-95 protects synapses from β-amyloid. Cell Rep. 2021 Jun 1;35(9):109194. PubMed.

    View all comments by Roberto Malinow

  9. This is certainly a tour de force of proteomic analyses. In combination with machine learning, the authors found several differences in the protein expressions of key proteins, such as CD47, DJ-1, and ApoE in AD brains. They also implied that, for the first time, ApoE might be an important factor for resilience against AD pathology.

    Further comparisons might reveal the neurodegenerative process at a molecular level in situ.

    View all comments by Taisuke Tomita

Make a Comment

To make a comment you must login or register.

This paper appears in the following:

News

  1. Single Synapse Mass Spec Snags CD47 as Alzheimer’s Resilience Factor