. The viral protein corona directs viral pathogenesis and amyloid aggregation. Nat Commun. 2019 May 27;10(1):2331. PubMed.


Please login to recommend the paper.


  1. Ezzat et al. show that HSV1 accumulates a multiprotein corona, which they propose enhances infectivity. The authors present data showing Aβ binds HSV1 and RSV, and that the binding induces fibrillization of the peptide. We have previously shown that Aβ binding to HSV1 mediates rapid fibril generation on the surface of HSV1, as well as similar results for HHV6A and 6B (Eimer et al., 2018). This, and our finding in the same paper that HSV1 infection accelerates amyloidosis in 5XFAD mice, is nicely confirmed by the Ezzat et al. data.

    The authors state “... evidence of a direct role of HSV-1 in the process of amyloid nucleation and subsequent fibril growth is currently lacking” (p. 7, 2nd last para). We respectfully point out that a central finding of our previous study is that HSV1 (and other pathogens) directly seed amyloid deposition as a mechanism to trap and neutralize invading microbes. We also showed the same for Aβ-mediated entrapment of Candida and bacteria (Kumar et al., 2016). 

    The authors propose that Aβ is among the proteins that enhance HSV1 infectivity. This argument is at odds with data from at least three independent studies showing Aβ fibrillization mediates host protective activities against viral infection. Bourgade et al. showed Aβ protects against HSV1 in cell culture (Bourgade et al., 2015). White et al. showed Aβ protects cultured cells from infection by influenza virus (White et al., 2014). We showed Aβ protects cultured cells against HSV1, HHV6A, and 6B, and lengthens survival of AD mice (5XFAD) with acute HSV1 neuroinfection (Eimer et al., 2018). Other investigators have confirmed Aβ fibrillization protects against bacteria and fungal pathogens (Spitzer et al., 2016). Independent studies have shown HSV1 co-localizes with Aβ plaques, consistent with Aβ fibril-mediated capture and entrapment (Wozniak et al., 2009; Harris et al., 2018). We refer to this as the antimicrobial protection hypotheses of AD (Moir et al., 2018), which is more consistent with the emerging data on the AMP actions of the peptide and the protective microbial entrapment role of Aβ in brain.

    All three virus studies support an antiviral activity for Aβ that involves viral agglutination. Microbial agglutination is part of the protective activities of a wide range of antimicrobial peptides (AMPs), consistent with the recent identification of Aβ as an AMP. AMP fibrillization is recognized as a key mechanism mediating AMP agglutination pathways (Torrent et al., 2012; Tsai et al., 2011a, 2011b; Chu et al., 2012; Yount et al., 2006). Fibrillization is a classical AMP mechanism, not something unique to Aβ.

    The finding that HSV1 accumulates a protein corona is potentially an important advance in understanding viral infectivity. However, Ezzat et al. provide no data to support their assertion that Aβ fibrils increase the infectivity of HSV1 or RSV. They assume that Aβ is recruited by HSV1 to be part of the protein corona that enhances the virus’s pathogenic actions. Our previous data do not support this interpretation.

    In conclusion, previously published data support the hypothesis that Aβ inhibits viral infections via an ancient, evolutionarily conserved, AMP agglutination pathway. The notion that Aβ fibrils enhance HSV1 infection is not well-supported by the literature and is lacking direct evidence. The available data continue to support microbial seeding and entrapment as a means for antimicrobial protection and consequent amyloidosis in the brain following HSV1 infection.


    . β-Amyloid peptides display protective activity against the human Alzheimer's disease-associated herpes simplex virus-1. Biogerontology. 2015 Feb;16(1):85-98. Epub 2014 Nov 7 PubMed.

    . Human α-defensin 6 promotes mucosal innate immunity through self-assembled peptide nanonets. Science. 2012 Jul 27;337(6093):477-81. Epub 2012 Jun 21 PubMed.

    . Molecular Mechanisms for Herpes Simplex Virus Type 1 Pathogenesis in Alzheimer's Disease. Front Aging Neurosci. 2018;10:48. Epub 2018 Mar 6 PubMed.

    . Amyloid-β peptide protects against microbial infection in mouse and worm models of Alzheimer's disease. Sci Transl Med. 2016 May 25;8(340):340ra72. PubMed.

    . The antimicrobial protection hypothesis of Alzheimer's disease. Alzheimers Dement. 2018 Dec;14(12):1602-1614. Epub 2018 Oct 9 PubMed.

    . Amyloidogenic amyloid-β-peptide variants induce microbial agglutination and exert antimicrobial activity. Sci Rep. 2016 Sep 14;6:32228. PubMed.

    . Exploring new biological functions of amyloids: bacteria cell agglutination mediated by host protein aggregation. PLoS Pathog. 2012;8(11):e1003005. Epub 2012 Nov 1 PubMed.

    . Characterizing the role of cell-wall β-1,3-exoglucanase Xog1p in Candida albicans adhesion by the human antimicrobial peptide LL-37. PLoS One. 2011;6(6):e21394. Epub 2011 Jun 21 PubMed.

    . Human antimicrobial peptide LL-37 inhibits adhesion of Candida albicans by interacting with yeast cell-wall carbohydrates. PLoS One. 2011 Mar 14;6(3):e17755. PubMed.

    . Alzheimer's associated β-amyloid protein inhibits influenza A virus and modulates viral interactions with phagocytes. PLoS One. 2014;9(7):e101364. Epub 2014 Jul 2 PubMed.

    . Herpes simplex virus type 1 DNA is located within Alzheimer's disease amyloid plaques. J Pathol. 2009 Jan;217(1):131-8. PubMed.

    . Advances in antimicrobial peptide immunobiology. Biopolymers. 2006;84(5):435-58. PubMed.

    View all comments by Rudy Tanzi
  2. This paper provides further detail on the interaction between herpes simplex type 1 and Aβ. The results are in line with those presented by Robert Moir's group last year (Eimer et al., 2018). They confirm that the presence of viral particles accelerates Aβ fibril formation, and that growing Aβ fibrils are radiating from the viral particle surface.

    The Aβ peptide seems to have potent antimicrobial properties, and fibril formation is the mechanism by which it entraps and incapacitates pathogens. This also provides a mechanistic link between persistent infections within the central nervous system, Aβ plaque formation, and the development of Alzheimer's disease.

    Herpes simplex type 1 infection is common, with the potential to establish persistent infections within the central nervous system, and a link has been confirmed between this specific pathogen and Alzheimer's disease in prospective population-based cohort studies. The National Institutes of Health (NIH) recently announced that infectious etiology of Alzheimer's disease is now regarded as a high-priority topic of interest, and, in my opinion, the possibility to affect the risk of Alzheimer's disease with treatments targeting herpes simplex infection should now be investigated.


    . Alzheimer's Disease-Associated β-Amyloid Is Rapidly Seeded by Herpesviridae to Protect against Brain Infection. Neuron. 2018 Jul 11;99(1):56-63.e3. PubMed.

    View all comments by Hugo Lövheim
  3. This study is important because it provides a mechanism of action that rationalizes previous epidemiological observations of a correlation between herpes and Alzheimer's disease.

    Alzheimer's is characterized by the deposition of the Aβ peptide in plaques in the brain. The peptide itself, however, is present at low concentrations, and therefore its aggregation is very slow and can be effectively regulated by our natural defense mechanisms throughout most of our lives, at least until we age. Although this is somewhat reassuring, the presence of triggers can speed up the formation of aggregates and eventually overwhelm our bodies’ capacity to eliminate them.

    Some triggers, for example cholesterol-containing lipid membranes, have been already identified, and others are likely to exist. This study shows that protein-coated herpes viruses of the HSV-1 type can indeed act in this way.

    When our natural defenses are intact, HSV1-catalyzed Aβ aggregation is kept at bay, but upon aging the situation changes and the balance can become compromised, especially in the presence of other contributing factors, for example mutations in ApoE or certain other genes.

    From a therapeutic point of view, although anti-HSV-1 treatments may be beneficial to prevent Alzheimer's disease, I would expect that drugs that act in a different way, by inhibiting Aβ aggregation in any form, would be most effective. The reason is that this process may be catalyzed in a variety of different manners, and therefore trying to directly block the majority, if not all, of them is likely to be very challenging.

    View all comments by Michele Vendruscolo
  4. In this work, we investigated whether viral corona interactions with host factors involve surface-assisted nucleation of amyloidogenic peptides. Similar to other phase-transition processes (such as crystallization, for example), the process of amyloid formation is biphasic, with an initial rate-limiting nucleation phase followed by a rapid growth or elongation phase (Eisenberg and Jucker, 2012; Arosio et al., 2015; Adamcik and Mezzenga, 2018). 

    According to classical nucleation theory, small molecular assemblies (nuclei) need to be formed within a supersaturated metastable phase (soluble peptide at high concentration) for phase transition to take place (Vekilov, 2010; Buell, 2017). The process of nucleation represents the energy barrier that the system has to overcome to transform from a supersaturated metastable phase to a more stable phase; amyloid fibrils in this case (Buell, 2017; Adamcik and Mezzenga, 2018). Nucleation can take place due to the spontaneous formation of nuclei within the bulk of a supersaturated phase, and in this case, it is termed homogeneous nucleation (Vekilov, 2010; Kamano et al., 2014). However, homogenous nucleation is a rare and slow process, as it depends on the unlikely event of the spontaneous formation of a stable nucleus (Eisenberg and Jucker, 2012). 

    Adding preformed amyloid fragments can circumvent the nucleation barrier leading to accelerated amyloid formation in a process termed “seeding” (Buell, 2017). However, the more common mechanism of phase transition is via nucleation catalyzed by the presence of a foreign surface, i.e., heterogenous nucleation (Kamano et al., 2014; Buell, 2017). The surface acts as a scaffold that facilitates the assembly of the nucleus, which logarithmically lowers the nucleation energy barrier and accelerates the process of phase transition, usually eradicating the lag phase completely from the kinetics. In this regard, nanoparticles have been shown to act as catalytic surfaces that facilitate heterogenous nucleation of amyloid fibrils via binding, concentrating and enabling conformational changes of amyloidogenic peptides (Linse et al., 2007; Mahmoudi et al., 2013; Buell, 2017). 

    In our paper, we showed that viruses, as natural pathogenic nanoparticles, are of capable of the same catalytic process. We respectfully point out that nucleation and seeding are two terms that refer to two well-defined and distinct physicochemical processes as described above, and that in our paper we describe the physical mechanism by which viruses catalyze the process of nucleation via binding, concentrating, and enabling conformational changes of amyloidogenic peptides.

    This explanation is different from the work reported by Eimer et al., 2018, in two ways:

    1. Eimer et al. describe an Aβ that is already oligomerized and targets viral glycoproteins via a heparin-binding domain. In our work, which is based on the classical nucleation theory, oligomerization to form the initial nucleus is the rate-limiting step that is catalyzed by the viral surface as a whole.
    2. We describe a physical mechanism that does not invoke a teleological explanation of a particular biological function of the aggregated peptide, which means that the mechanism by itself neither proves nor disproves the antimicrobial hypothesis or any other functional hypotheses.

    As such, we describe a general phenomenon by which viruses bind proteins in their coronas, leading to protein concentration and conformational changes that can have pathogenic consequences. One consequence can be triggering phase transition into amyloid forms via heterogenous nucleation. Another consequence can be inducing conformational changes without phase transition; however, exposing cryptic epitopes can lead to autoimmune responses, and there are several autoimmune diseases that are known to follow viral infections (Münz et al., 2009). 

    The antimicrobial hypothesis cannot be excluded; however, there are at least 37 peptides and proteins that form amyloids in different human diseases (Chiti and Dobson, 2017), and many of them have other known functions that are not related to innate immunity or antimicrobial response. In addition, it has been reported that amyloids can enhance the infectivity of viruses such as HSV and HIV (Wojtowicz et al., 2002; Münch et al., 2007). Moreover, when animals are deficient in orthodox antimicrobial peptides such as LL37 they become extremely susceptible for infection (Nizet et al., 2001). However, Kumar et al., 2016, reported that APP knockout animals were not significantly more prone to infection than wild-type. Another explanation for the extra protection demonstrated in animals overexpressing Aβ might be due to the overlap of certain structural features between amyloids and antimicrobial peptides that is magnified by overexpression, which does not necessarily indicate that amyloids are antimicrobial peptides in physiological conditions (Landreh et al., 2014). 

    Our physical model has some predictions that can be falsified or verified experimentally by studying the phenomenon of amyloid polymorphism. I quote from the paper: “… [W]e hypothesize that the viral-catalyzed heterogeneously nucleated amyloids would be polymorphically distinct from homogeneously nucleated amyloids that result from aberrant protein expression/overexpression. This hypothesis can be tested in future studies on the Aβ polymorphism in familial and sporadic Alzheimer’s disease and whether this structural polymorphism can be traced back to different nucleation mechanisms related to different etiologies."


    . On the lag phase in amyloid fibril formation. Phys Chem Chem Phys. 2015 Mar 28;17(12):7606-18. PubMed.

    . The Nucleation of Protein Aggregates - From Crystals to Amyloid Fibrils. Int Rev Cell Mol Biol. 2017;329:187-226. Epub 2016 Oct 22 PubMed.

    . Protein Misfolding, Amyloid Formation, and Human Disease: A Summary of Progress Over the Last Decade. Annu Rev Biochem. 2017 Jun 20;86:27-68. Epub 2017 May 12 PubMed.

    . Alzheimer's Disease-Associated β-Amyloid Is Rapidly Seeded by Herpesviridae to Protect against Brain Infection. Neuron. 2018 Jul 11;99(1):56-63.e3. PubMed.

    . The amyloid state of proteins in human diseases. Cell. 2012 Mar 16;148(6):1188-203. PubMed.

    . Quantitative evaluation on the heterogeneous nucleation of amino acid by a thermodynamic analysis. Journal of Molecular Liquids 2014

    . Amyloid-β peptide protects against microbial infection in mouse and worm models of Alzheimer's disease. Sci Transl Med. 2016 May 25;8(340):340ra72. PubMed.

    . Separate molecular determinants in amyloidogenic and antimicrobial peptides. J Mol Biol. 2014 May 29;426(11):2159-66. Epub 2014 Mar 17 PubMed.

    . Nucleation of protein fibrillation by nanoparticles. Proc Natl Acad Sci U S A. 2007 May 22;104(21):8691-6. PubMed.

    . Protein fibrillation and nanoparticle interactions: opportunities and challenges. Nanoscale. 2013 Apr 7;5(7):2570-88. PubMed.

    . Amyloid Polymorphism in the Protein Folding and Aggregation Energy Landscape. Angew Chem Int Ed Engl. 2018 Jul 9;57(28):8370-8382. Epub 2018 Jun 8 PubMed.

    . Semen-derived amyloid fibrils drastically enhance HIV infection. Cell. 2007 Dec 14;131(6):1059-71. PubMed.

    . Antiviral immune responses: triggers of or triggered by autoimmunity?. Nat Rev Immunol. 2009 Apr;9(4):246-58. PubMed.

    . Innate antimicrobial peptide protects the skin from invasive bacterial infection. Nature. 2001 Nov 22;414(6862):454-7. PubMed.

    . Nucleation. Cryst Growth Des. 2010 Nov 15;10(12):5007-5019. PubMed.

    . Stimulation of enveloped virus infection by beta-amyloid fibrils. J Biol Chem. 2002 Sep 20;277(38):35019-24. PubMed.

    View all comments by Kariem Ezzat
  5. This is an intriguing study. The ex-vivo “test tube” studies are quite convincing that HSV and other virus particles can nucleate Aβ fibril formation.  

    However, we find the report in Figure 7 that inoculation of 3-month-old 5XFAD mice with HSV1 supernatant results in a tripling of plaque load in 48 hours implausible. In 3-month-old 5XFAD mice (Oakley et al., 2006), the total Aβ accumulated can be estimated to range from 500 to 1,000 picomoles per brain. In this study, reported plaque load measured by IHC increases almost three- to fourfold. As IHC amyloid loads and biochemical loads are quite proportional at this age, this means that an additional 1,500 to 3,000 picomoles of Aβ deposited in only two days. 

    Where could this come from? This amount would far exceed the amount of Aβ produced in the 5XFAD mouse brain in 48 hours, which we estimate to be ~100 pmoles per day and likely less (see note regarding this calculation below). This means that, in two days, the maximum production could account for roughly a 20 percent increase in amyloid deposition if every single molecule of Aβ (<400 pmoles) produced deposited. Indeed, in the 5XFAD mice, to increase plaque load by three- to fourfold normally takes two or more months.

    It is possible, but not very likely, that the inoculum dramatically increases Aβ production in this timeframe. It is also possible that some other confound is present that alters the immunohistochemical assessment (it would have been valuable to provide biochemical data). Unless the authors can account for a pool of “dark” Aβ, it’s hard to account for the outcome of a three- to fourfold increases in plaque load based solely on seeding induced by HSV particles.

    In Levites et al., 2006, using rigorous biochemical measurements, we estimated that ~250 pmoles of Aβ turn over in the Tg2576 mouse brain every day. As APP levels are about threefold lower in the 5XFAD mice, we would suggest that the turnover is probably around 100 pmoles per day. Given the rapid onset of plaque deposition in 5XFAD mice, we are unaware of a report of total predeposition steady-state level of Aβ in these mice. We have therefore used the Tg2576 as the comparison and made a conservative estimate of a maximal amount of Aβ produced. Reports of brain Aβ half-lives in mice are very consistent, and thus we are confident that these quantitative estimates are quite valid.


    . Intraneuronal beta-amyloid aggregates, neurodegeneration, and neuron loss in transgenic mice with five familial Alzheimer's disease mutations: potential factors in amyloid plaque formation. J Neurosci. 2006 Oct 4;26(40):10129-40. PubMed.

    . Insights into the mechanisms of action of anti-Abeta antibodies in Alzheimer's disease mouse models. FASEB J. 2006 Dec;20(14):2576-8. Epub 2006 Oct 26 PubMed.

    View all comments by David R. Borchelt
  6. In Figure 7, we report a significant net increase in the percentage of Aβ-positive areas by approximately 4 percent in the hippocampus and approximately 3 percent in the cortex. We respectfully point out that this is not the same thing as a fourfold increase in the total amyloid load in the brain, which is the basis of the calculations made in the comment by Drs. Golde and Borchelt, and we have not implied that in the results nor in the discussion.

    To quote Levites et al., “After immunization of PDAPP and other APP mice with abundant diffuse plaques, much larger reductions in immunohistochemical amyloid loads are reported than biochemical loads as measured by ELISA, suggesting a preferential reduction of diffuse Aβ deposits” (Levites et al., 2006). This indicates that IHC amyloid loads and biochemical loads are not always proportional. Since this is the case, would it be accurate to make assumptive calculations, especially ones that involve extrapolation from one model to another where these measurements have not been performed experimentally?

    Our data is in accordance with the time frame reported for increased Aβ accumulation 48 hours post-infection (Eimer et al., 2018). As a catalytic process that involves conformational change and phase transition, it is expected that much of the soluble Aβ that was not detectable by the antibodies before viral inoculation becomes more available for antibody binding after the phase change into amyloid fibrils. In this regard, the ability of viruses to replicate and get released in large quantities could be a particularly potent trigger for amyloid precipitation in a certain microenvironment due to the sudden availability of vast areas of exogenous nanosurfaces.


    . Insights into the mechanisms of action of anti-Abeta antibodies in Alzheimer's disease mouse models. FASEB J. 2006 Dec;20(14):2576-8. Epub 2006 Oct 26 PubMed.

    . Alzheimer's Disease-Associated β-Amyloid Is Rapidly Seeded by Herpesviridae to Protect against Brain Infection. Neuron. 2018 Jul 11;99(1):56-63.e3. PubMed.

    View all comments by Kariem Ezzat
  7. In response to Kariem Ezzat, we'd like to note that, having looked at thousands of Aβ IHCs, we would suggest that the "plaques" shown look pretty compact and not diffuse. This is what prompted our original critique. As biochemical data was not provided, the way to settle this is with data and reproduction. Our laboratories would be happy to have blinded samples sent that have been injected with the supernatent or control to see if such an effect is reproducible, using both biochemical and histochemical assessments.

    View all comments by David R. Borchelt
  8. Scientific reproducibility is very important, and we would be very happy to share any material from our research.

    View all comments by Kariem Ezzat
  9. I would like to stress how novel and intriguing the findings by Ezzat et al. are, introducing a completely new aspect of viral interaction with host cells. It will be most interesting if the authors find how the corona influences extracellular spread of HSV1, as well as the effects it might have on axonal transport, especially vis-à-vis the studies of Bearer’s group on the transport of APP associated with HSV1 along axons (Cheng et al., 2011). 

    Also, the fact that the proteins in the corona do not reflect their abundance in the surrounding medium, as is also the case for nanoparticles, is another fascinating discovery and presumably might be  involved in determining viral infectivity.


    . Herpes simplex virus dances with amyloid precursor protein while exiting the cell. PLoS One. 2011;6(3):e17966. PubMed.

    View all comments by Ruth Itzhaki

Make a Comment

To make a comment you must login or register.

This paper appears in the following:


  1. When Host Proteins Coat Virus, Amyloid Fibrils Form