REVIEW published: 27 November 2020 doi: 10.3389/fnmol.2020.582488 MIRRAGGE – Minimum Information Required for Reproducible AGGregation Experiments Pedro M. Martins 1,2† , Susanna Navarro 3† , Alexandra Silva 1 , Maria F. Pinto 1 , Zsuzsa Sárkány 1 , Francisco Figueiredo 1,2,4 , Pedro José Barbosa Pereira 1 , Francisca Pinheiro 3 , Zuzana Bednarikova 5 , Michał Burdukiewicz 6 , Oxana V. Galzitskaya 7,8 , Zuzana Gazova 5 , Cláudio M. Gomes 9 , Annalisa Pastore 10 , Louise C.Serpell 11 , Rostislav Skrabana 12,13 , Vytautas Smirnovas 14 , Mantas Ziaunys 14 , Daniel E. Otzen 15 , Salvador Ventura 3* and Sandra Macedo-Ribeiro 1* 1 Instituto de Biologia Molecular e Celular and Instituto de Investigação e Inovação em Saúde, Universidade do Porto, Porto, Portugal, 2 Instituto de Ciências Biomédicas Abel Salazar, Universidade do Porto, Porto, Portugal, 3 Institut de Biotecnologia i Biomedicina – Departament de Bioquímica i Biologia Molecular, Universitat Autònoma de Barcelona, Bellaterra, Spain, 4 International Iberian Nanotechnology Laboratory – Department of Atomic Structure – Composition of Materials, Braga, Portugal, 5 Department of Biophysics, Institute of Experimental Physics, Slovak Academy of Sciences, Kosice, Slovakia, 6 Faculty of Mathematics and Information Science, Warsaw University of Technology, Warsaw, Poland, 7 Institute of Protein Research, Russian Academy of Sciences, Pushchino, Russia, 8 Institute of Theoretical and Experimental Biophysics, Russian Edited by: Academy of Sciences, Pushchino, Russia, 9 Biosystems and Integrative Sciences Institute and Departamento de Química e Serena Carra, Bioquímica, Faculdade de Ciências, Universidade de Lisboa, Lisbon, Portugal, 10 UK-DRI Centre at King’s College London, University of Modena and Reggio the Maurice Wohl Clinical Neuroscience Institute, London, United Kingdom, 11 Sussex Neuroscience, School of Life Emilia, Italy Sciences, University of Sussex, Brighton, United Kingdom, 12 Department of Neuroimmunology, Axon Neuroscience R&D Reviewed by: Services SE, Bratislava, Slovakia, 13 Institute of Neuroimmunology, Slovak Academy of Sciences, Bratislava, Slovakia, 14 Fabrizio Chiti, Institute of Biotechnology, Life Sciences Center, Vilnius University, Vilnius, Lithuania, 15 Interdisciplinary Nanoscience Center University of Florence, Italy (iNANO) and Department of Molecular Biology and Genetics, Aarhus University, Aarhus, Denmark Ehud Cohen, Hebrew University of Jerusalem, Israel Reports on phase separation and amyloid formation for multiple proteins and *Correspondence: Salvador Ventura aggregation-prone peptides are recurrently used to explore the molecular mechanisms salvador.ventura@uab.cat associated with several human diseases. The information conveyed by these reports can Sandra Macedo-Ribeiro be used directly in translational investigation, e.g., for the design of better drug screening sribeiro@ibmc.up.pt † These strategies, or be compiled in databases for benchmarking novel aggregation-predicting authors share first authorship algorithms. Given that minute protocol variations determine different outcomes of Received: 12 July 2020 protein aggregation assays, there is a strong urge for standardized descriptions of Accepted: 23 October 2020 Published: 27 November 2020 the different types of aggregates and the detailed methods used in their production. Citation: In an attempt to address this need, we assembled the Minimum Information Required Martins PM, Navarro S, Silva A, for Reproducible Aggregation Experiments (MIRRAGGE) guidelines, considering first- Pinto MF, Sárkány Z, Figueiredo F, principles and the established literature on protein self-assembly and aggregation. Pereira PJB, Pinheiro F, Bednarikova Z, Burdukiewicz M, This consensus information aims to cover the major and subtle determinants of Galzitskaya OV, Gazova Z, experimental reproducibility while avoiding excessive technical details that are of limited Gomes CM, Pastore A, Serpell LC, Skrabana R, Smirnovas V, Ziaunys M, practical interest for non-specialized users. The MIRRAGGE table (template available in Otzen DE, Ventura S and Supplementary Information) is useful as a guide for the design of new studies and as Macedo-Ribeiro S (2020) a checklist during submission of experimental reports for publication. Full disclosure of MIRRAGGE – Minimum Information Required for Reproducible relevant information also enables other researchers to reproduce results correctly and AGGregation Experiments. facilitates systematic data deposition into curated databases. Front. Mol. Neurosci. 13:582488. doi: 10.3389/fnmol.2020.582488 Keywords: amyloid, reproducible data, protein, peptide, phase separation Frontiers in Molecular Neuroscience | www.frontiersin.org 1 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE INTRODUCTION mutants are characterized by a diminished ability to reiteratively shift between dispersed and condensed phases consisting of Aggregation of misfolded proteins and peptides is associated dense liquids and gels (Murakami et al., 2015). The physical with common and rare neurodegenerative disorders and properties of the different polypeptide assemblies (summarized with amyloidoses (Aguzzi and O’Connor, 2010; Chiti and in Figure 1D) are determined by the degrees of molecular and Dobson, 2017; Benson et al., 2018). Classical neuropathological supramolecular order. Although denser liquid phases relax more hallmarks such as neurofibrillary tangles in Alzheimer’s disease slowly in response to shear deformation, they still lack the (AD), Lewy bodies in Parkinson’s disease (PD), or inclusion long-range translational order characteristic of solids (Falahati bodies in Huntington’s disease, have as main constituent and Haji-Akbari, 2019). The maximal organization provided by characteristic polypeptides aggregated in the form of insoluble crystal lattices allows the structure of folded proteins to be solved highly-ordered structures named amyloid fibrils (Westermark using X-ray crystallography. Naturally occurring microcrystals et al., 2007; Gremer et al., 2017). The definition of amyloid are also used by living cells for protein storage, protection and encompasses morphological (Figure 1A), structural (Figure 1B) stabilization (Schönherr et al., 2018), whereas in crystallopathies and histological (Figure 1C) aspects. Amyloid polymorphs such as eosinophilic inflammation, protein crystals have been are associated with different clinical sub-types of the same reported as promising drug targets (Persson et al., 2019). disease (Guo et al., 2013; Qiang et al., 2017; Goedert et al., 2018), and they exhibit structural differences that can be characterized at atomic resolution using solid-state nuclear THERMODYNAMIC AND KINETIC magnetic resonance with magic angle spinning (ss-NMR) and OBSTACLES TO REPRODUCIBILITY cryo-electron microscopy (cryo-EM) techniques (Fitzpatrick et al., 2013; Qiang et al., 2017; Rasmussen et al., 2017; According to the phase rule of thermodynamics, the maximum Iadanza et al., 2018b). When the amyloid fold is exploited by number of stable phases that can coexist within a mixture nature for functional purposes, the structural variability is less is limited to k + 2, with k being the number of non-reactive evident than in disease-related aggregates, thus suggesting that components present in the mixture (Falahati and Haji-Akbari, functional amyloids are the result of naturally evolved amino 2019). The reduced number of accessible microstates due to acid sequences (Otzen and Riek, 2019). The analogy of a low- demixing implies an entropic penalty of −T1S > 0 reflecting energy “black hole” has been used to illustrate the propensity the disorder-to-order transition (Shin and Brangwynne, 2017; of proteins to form amyloid fibrils after incomplete native Falahati and Haji-Akbari, 2019). This entropic cost, which is folding or if subjected to appropriate denaturing conditions; higher at higher temperature (T), can be compensated by under mildly denaturing conditions, native interactions and the enthalpic contribution (1H < 0) provided that the new intrachain interactions are in competition, so that the occurrence intermolecular interactions are sufficiently strong to decrease of transiently stable, non-native conformations may trigger the Gibbs free energy 1G = 1H-T1S. Whether a given protein amyloid fibril formation (Zheng et al., 2013). However, there undergoes a phase separation or not is, therefore, predictable are other energetically stable structures into which proteins self- according to the information of temperature and composition assemble irrespectively of, prior to, or concomitantly with the given in phase diagrams. In practice, however, differences in the formation of amyloid fibrils (Figure 1D). Of these, the soluble protein source and sample manipulation generate aggregation aggregates of amyloidogenic proteins attract a particular interest states that are not readily duplicated among distinct laboratories, justified by the cytotoxic properties attributed to amyloid-β (Aβ) a problem that was highlighted in the editorial entitled “State oligomers in AD (Ahmed et al., 2010; Yang et al., 2017) and of Aggregation” of Nature Neuroscience in April, 2011. The α-synuclein oligomers in PD (Lorenzen et al., 2014; Ingelsson, difficulties in reproducing protein self-assembly experiments are 2016). Liquid-liquid phase separation is now recognized as correlated with the occurrence of metastable states of protein having a central role in cell physiology and disease (Shin and folding (Sohl et al., 1998) and phase separation (Gazit, 2002; Brangwynne, 2017). For example, membraneless compartments Baldwin et al., 2011) consisting of local free energy minima whose of concentrated proteins/nucleic acids are implicated in diverse evolution toward the global minimum takes place over too long processes, including RNA metabolism, ribosome assembly, DNA timescales to be biologically realistic. Partially unfolded states that repair and intracellular signaling (Banani et al., 2017). When are more stable than the native fold (Giri Rao and Gosavi, 2018) the capacity of protein quality control by the proteasome is and the kinetic trapping of condensates into oligomeric (Miti exceeded, misfolded proteins are sequestered into intracellular et al., 2015) and gel-like states (Alberti, 2017) accentuate the need compartments such as the aggresome, a perinuclear deposit for a clear description of the initial state of the protein, its source destined to autophagy (Johnston et al., 1998), or the CytoQ and how it was manipulated. and INQ, which are deposition sites of misfolded proteins The interplay between protein folding, oligomerization in the cytosol and the nucleus, respectively (Miller et al., and metastable phase separation adds uncertainty to the 2015). On the other hand, a neuropathological hallmark of dynamic distribution of the different species during test-tube sporadic and inherited forms of amyotrophic lateral sclerosis experiments. How cells spatiotemporally regulate these processes and frontotemporal dementia is the deposition of poorly soluble is additionally determined by the occurrence of post-translational assemblies of mutated RNA-binding proteins in the nucleus modifications (Boeynaems et al., 2018; González et al., 2019), and cytoplasm of neurons (Patel et al., 2015). The pathogenic chaperone recruitment (Mateju et al., 2017) and macromolecular Frontiers in Molecular Neuroscience | www.frontiersin.org 2 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE FIGURE 1 | Amyloid and non-amyloid aggregates. (A) Amyloids have the characteristic fibrillar appearance illustrated in this transmission electron microscopy (TEM) image of negatively stained fibrils from a peptide derived from the SH3 domain of PI3-kinase (SH3-PI3K) (Ventura et al., 2004). The bar corresponds to 200 nm. (B) The ordered β-sheet structure of amyloid fibrils produces the cross-β X-ray diffraction pattern here exemplified for fibrils of a peptide derived from Mot3 yeast prion (Sant’Anna et al., 2016; Fernandez et al., 2017). The blue and red arrows indicate typical reflections at 4.7 and 10.2 Å, respectively. (C) Thioflavin-S (Th-S) is an histological dye that shows fluorescence in presence of amyloid fibrils (example of SH3-PI3K fibrils stained with Th-S) (Ventura et al., 2004). The bar corresponds to 20 µm. (D) Folded and misfolded monomers are in dynamic equilibrium with soluble oligomers and may coexist in solution with different condensates, such as dense liquids and gels, amyloid fibrils and their protofibrillar precursors, crystals and amorphous solids. For simplicity, the more complex cases of fibril polymorphism, multiprotein assemblies or nucleic acid-protein assemblies are not shown, and amyloid fibrils are represented as composed of four protofilament units. (B) Adapted from (Sant’Anna et al., 2016). All copyright permissions have been secured by the authors. crowding (Aguzzi and Altmeyer, 2016; Rivas and Minton, 2016). oligomerization (Silva et al., 2017), irreversible oligomerization For the molecular-level understanding of protein aggregation, (Silva et al., 2018), and conformational transitions (Ruff et al., kinetic results obtained under tightly controlled conditions and 2019), to cite but a few examples. As a stochastic process, primary in the presence of specific dyes, such as Thioflavin-T (Th- nucleation is itself a source of variability in which stable clusters T), Thioflavin-S (Th-S) and Congo red (CR), are analyzed are formed occasionally as the result of random molecular using different nucleation-and-growth models (Crespo et al., collisions (Vekilov, 2010; Sleutel and Van Driessche, 2018). 2012; Meisl et al., 2016; Chatani and Yamamoto, 2018). When nucleation is a rare event, the emergence of measurable A combination of complementary techniques including (but amounts of protein aggregates is preceded by a lag phase not limited to) chromatography, light scattering, and advanced whose variable duration reflects the probability distribution of microscopy is required to validate complex mechanisms that a successful event (Crespo et al., 2012; Michaels et al., 2016). may comprise the formation of intermediate phases (Ianiro During amyloid fibril formation, prior addition of > 1% of et al., 2019), precursor oligomers (Pieri et al., 2016), reversible preformed fibrils (or seeds) is often sufficient to eliminate the Frontiers in Molecular Neuroscience | www.frontiersin.org 3 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE lag phase and the intrinsic uncertainty associated to primary this free energy barrier is overcome, phase transition will proceed nucleation (Sárkány et al., 2019). Conversely, protein samples at a faster rate if catalyzed by the secondary steps of growth (or containing residual, yet variable, amounts of preformed seeds elongation) and templated nucleation (or secondary nucleation) may not be suitable for “unseeded” assays due to irreproducible (Padrick and Miranker, 2002; Crespo et al., 2012; Meisl et al., lag times (Crespo et al., 2012, 2016). In these cases, a final step 2016; Chatani and Yamamoto, 2018). Alternatively, the initially of sample polishing immediately before the aggregation assay is homogeneous phase A can be further destabilized by increasing recommended to eliminate vestigial assemblies formed during xi and/or by decreasing T to a point of spinodal decomposition storage or upon thawing (Mahler et al., 2009; Silva et al., 2017; where phase separation occurs by spontaneous amplification of Weinbuch et al., 2018). infinitesimal phase fluctuations (Falahati and Haji-Akbari, 2019). The absence of a visible lag phase in aggregation progress curves does not necessarily mean that the critical nucleus has reached the AN ONTOLOGICAL APPROACH TO minimum size of 1 molecule required for spinodal decomposition PROTEIN AGGREGATION (Vekilov, 2010). Instead, it may happen that the presence of a nucleation barrier is concealed by (comparatively) fast elongation As new published data on protein aggregation and phase and secondary nucleation steps, thus originating the hyperbolic separation proliferate, they increase the risk that different curves typical of downhill polymerization (Hurshman et al., 2004; semantics are adopted to characterize the same entities, or Crespo et al., 2012). else, that different entities end up having the same designation. As discussed above, the occurrence of kinetically trapped Motivated by a similar concern, the International Society of species is one of the obstacles for reproducible reporting of Amyloidosis has created a nomenclature committee whose aggregation experiments. While corresponding to an unstable, periodical reports try to keep pace with the ever-increasing high-energy state, the short-lived nucleation embryos are not knowledge of amyloid disorders in humans and animals included in the catalog of such intermediates. Primary nuclei, (Westermark et al., 2007; Benson et al., 2018). Recent efforts on the contrary, may have a fixed, well-ordered structure at toward a controlled vocabulary were also taken in the context the moment of their formation, or pass through local spatial of phase separation in living cells (Alberti, 2017; Shin and ordering and structural annealing until a free energy minimum Brangwynne, 2017; Boeynaems et al., 2018), or while establishing is eventually reached (Murray et al., 2017; Boeynaems et al., guidelines to the use of Th-T in the presence of non-amyloid 2018). The sizes of primary and secondary nuclei can be estimated species (Gade Malmos et al., 2017). We now propose an from the concentration dependences of kinetic parameters such ontological roadmap (Figure 2) and a systematized terminology as the duration of the lag phase and the limit aggregation (Table 1), considering the established nomenclature and the rates (Cohen et al., 2013; Dovidchenko et al., 2014; Silva et al., fundamental concepts of physical-chemical equilibrium. We 2018), but also through direct measurements using dynamic depart from the definition of phase as a region in space with light scattering (Walters and Murphy, 2011; Silva et al., 2017) uniform density and composition at a given pressure and and small-angle X-ray scattering techniques (Vestergaard et al., temperature (Prausnitz et al., 1998). Phase separation, therefore, 2007). In the hierarchy of amyloid fibril assembly, protofilaments refers to a change in density and/or composition that culminates are the primordial insoluble species and the fibrillar subunit of in the formation of a new phase. The transition of a molecule i the β-sheet stacking (Teplow, 1998; Rochet and Lansbury, 2000; from phase A to the new phase B is driven by the difference in Khurana et al., 2003; Makin and Serpell, 2005). Protofilaments chemical potential 1µ = µBi − µA i that results from the different intertwine to form small and flexible protofibrils, which are values of mole fraction xi and activity coefficient γi within the two classified as worm-like, rod-like or fuzzy, according to the phases. Recall that morphological features displayed in, e.g., EM micrographs (Gosal et al., 2005; Gade Malmos et al., 2017). The elongation and µi = µ0i + RTln (xi γi ) (1) intertwining of protofibrils give rise to typically straight and SDS-resistant mature fibrils with length often exceeding 1 µm with µ0i being the standard chemical potential and R the universal and diameter around 10–20 nm (Rochet and Lansbury, 2000; gas constant. The depletion of phase A (and the concomitant Eisenberg and Sawaya, 2017; Gade Malmos et al., 2017). Instead enrichment of phase B) in component i proceeds until the of mature fibrils, the end product of amyloid assembly may equilibrium condition µA i = µi is verified, so that B consist of protofibrils, as in the cases of the L55P mutant of xiA γiA = xiB γiB (2) transthyretin at physiological (instead of acidic) pH (Lashuel et al., 1999), and of ataxin-3 containing a non-expanded (fewer The formation of a new phase can occur instantaneously by than ca. 30 glutamine residues) polyglutamine tract (Carvalho spinodal decomposition, or via the energy-activated process et al., 2018). Equally, dense liquid droplets nucleated after local of nucleation. In the latter case, thermal and compositional enrichment of aggregation-prone proteins can either mediate fluctuations result in the generation of embryos of phase B new disorder-to-order transitions during crystallization and with non-zero surface free energy in the phase boundary. The amyloid fibril formation (Vorontsova et al., 2015; Aguzzi and thermodynamically unstable embryos tend to disintegrate back to Altmeyer, 2016), or develop into gel-like states with reduced phase A, unless they are larger than the critical size of the primary fluidity and protein movement (Aguzzi and Altmeyer, 2016; nucleus (Vekilov, 2010; Falahati and Haji-Akbari, 2019). Once Banani et al., 2017; Boeynaems et al., 2018). Pathway 1 in Frontiers in Molecular Neuroscience | www.frontiersin.org 4 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE FIGURE 2 | Ontological map of protein aggregation. Phase equilibria (solid outlines) are established between the liquid phase (green circle), the dense liquid phase (orange circle) and the solid phase (blue circles). The different species (represented as in Figure 1D) may undergo structural transitions (dashed lines) that originate distinct species within the same stable phase (different shades of color). Visual examples of aggregation pathways are provided, comprising many (arrow 1), some (arrow 2) and no (arrow 3) intermediate states. Off-pathway aggregation occurring in parallel with amyloid fibrillation is represented by a bifurcation (arrow 20 ). The outcome of phase separation (black dots) is contingent on concentration, temperature, pH, ionic strength, amino acid sequence, etc., but also on the initial state of the protein, as it determines the point in the green region from where the experiments depart (white dots). Figure 2 illustrates the formation of mature fibrils preceded Protein unfolding and oligomerization (represented in by the occurrence of several intermediate states, including Figure 2 by green dashed lines) involve the interconversion liquid droplets, gels, and protofibrils. Pathways 2 and 20 show between native, partially unfolded, denatured and oligomeric the possible coexistence of kinetically arrested states that, in states of the protein. The equilibrium composition of these this illustrative example, correspond to protofibrils and liquid structurally distinct entities results, as in phase separation, droplets, respectively. Pathway 3 corresponds to the direct from the balance of chemical potentials; however, each state formation of protein crystals from globular protein. of the protein is now characterized by different values of the Frontiers in Molecular Neuroscience | www.frontiersin.org 5 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE TABLE 1 | Glossary of terms used in protein aggregation phenomena. Amyloid fibril Insoluble, protease resistant aggregate with characteristic (i) fibrillar electron microscopic appearance, (ii) X-ray fiber diffraction pattern and (iii) histological staining reactions, including Th-T fluorescence and green birefringence in the presence of Congo red. Amyloidosis Any disease associated with the formation of amyloid deposits, i.e., deposits that have amyloid fibrils as main constituent. Condensate The denser phase formed upon phase separation. Embryo A metastable cluster formed as the result of random molecular collisions; embryos will tend to disintegrate if smaller than the critical nucleus size. Liquid demixing Separation of a solution into two coexisting liquid phases (compare with phase separation); also known as “liquid-liquid demixing.” Mature amyloid fibril A long, fully stable amyloid fibril consisting of intertwined protofilament units producing a wound structure that is resistant to 2% (w/v) sodium dodecyl sulfate (SDS) treatment. Nucleation The stochastic process by which nuclei are formed; primary nucleation occurs in the homogeneous bulk solution; secondary nucleation occurs on the surface of an already existing aggregate; heterogeneous nucleation occurs on the surface of a foreign substance. Nucleus A cluster formed as the result of random molecular collisions and with the ability to elongate into a fibril (compare with embryo). Off-pathway oligomer Stable oligomer that coexists with amyloid fibrils but is not an amyloid percursor; off-pathway oligomerization and amyloid fibrillation are competitive processes. Oligomer Multimeric species lacking the morphology and properties of amyloid fibrils. On-pathway oligomer A precursor of amyloid fibrils; the addition of pre-formed on-pathway oligomers accelerates the formation of amyloid fibrils without affecting the total amyloid conversion. Phase separation The formation of a new region with uniform density and composition; it can occur instantaneously by spinodal decomposition, or via the energy-activated process of nucleation. Protofibril Beaded chain with ∼5 nm diameter and <150 nm length that matures into amyloid fibrils. Protofilament Subunit of amyloid fibrils; smaller protofilaments with ∼1 nm diameter entwine to form larger protofilament units with 2.5–3.5 nm diameter. Seed Preformed amyloid particle able to accelerate the assembly of amyloid fibrils. standard chemical potential µ0i (Eq. 1) (Tanford, 1970). Although Folding intermediates have been detected for a limited number occurring within the same liquid phase, structural transitions of proteins by measuring the exchange rates of hydrogen atoms might trigger subsequent phase separations as in the general between the main-chain amides and water (Englander et al., case of protein misfolding diseases in which pathologic protein 2016), or by combining 1 H liquid-state NMR and multivariate aggregation arises from the failure of a specific protein or peptide analysis (Malmendal et al., 2010). Figure 2 necessarily simplifies to adopt its native conformation (Chiti and Dobson, 2006). Also, the folding landscape that is associated with pathogenic protein insulin amyloid fibril formation is preceded by the aggregation aggregation as different protein-protein interactions are required of a precursor helical oligomer that later becomes the repeating to maintain proteostasis in living cells. For example, the correct unit of mature fibrils (Vestergaard et al., 2007), whereas several folding of prion protein (PrP) is assured by chaperones of the of the key proteins present in membraneless organelles have endoplasmic reticulum such as calnexin (Wang et al., 2010) oligomerization domains that drive liquid demixing through and the proline cis/trans isomerase cyclophilin B (Ben-Gedalya multiplicative sticky interactions (Boeynaems et al., 2018). In a et al., 2015). Conformational variations between the normal different example, amyloid fibrillation of non-expanded ataxin-3 (PrPC ) and the infectious (PrPSC ) isoforms are responsible for decreases the concentration of soluble monomers, thus causing the recruitment of PrPC by PrPSC during amyloid polymerization the dissociation of off-pathway oligomers to monomeric species (Prusiner, 2001). The self-templating properties of PrPSC are the (Silva et al., 2017). Because the different states of protein basis for ultrasensitive tests for prion infections in biological aggregation are in dynamic equilibrium, a possible rescue fluids and tissues (Saborio et al., 2001; Orru et al., 2012). mechanism includes the sequestration of harmful species into Likewise, the amplification of seeding-competent aggregates less toxic aggregates or in phase separated compartments of the of Aβ peptide (Salvadores et al., 2014), α-synuclein (Fairfoul cell (Gosal et al., 2005; Haass and Selkoe, 2007; Boeynaems et al., et al., 2016) and tau (Saijo et al., 2017) is a promising 2018). Illustrating this protection mechanism, amyloid plaques technological principle for the diagnosis of AD, PD, and isolated from AD cortex do not elicit toxic effects in rodent tauopathies, respectively (Soto and Pritzkow, 2018). Amyloid hippocampus unless they are solubilized to release toxic Aβ seeding capacity can be used to evaluate whether a given dimers (Shankar et al., 2008). intermediate is on-pathway or off-pathway to form amyloid Systematizing protein folding intermediates and oligomeric fibrils. The maximal seeding potency is achieved upon the species into organized categories is a challenging task due to the addition of pre-formed, sonicated fibrils. Conversely, off-pathway heterogeneity and elusive nature of both populations. Moreover, oligomers (Farmer et al., 2017; Hasecke et al., 2018), fibril many proteins naturally occur as an ensemble of more than polymorphs (Kodali and Wetzel, 2007; Falcon et al., 2015; Cao one polypeptide chain folded into a characteristic oligomeric et al., 2019), and seeds pre-treated with aggregation inhibitors conformation (Doyle et al., 2013), while others, known as (Arimon et al., 2008; Oskarsson et al., 2018; Saelices et al., 2018) intrinsically disordered proteins, can sample a continuum of are expected to have lower seeding potencies. Amyloid fibrils conformations (Jarosz and Khurana, 2017; Darling et al., 2018). produced from distinct proteins can be tested for conformational Frontiers in Molecular Neuroscience | www.frontiersin.org 6 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE complementarity in cross-seeding experiments, also known as angle of 90◦ . It reports on the molar mass (weight-average) and heterologous or heterogeneous seeding (Harper and Peterand concentration of the aggregate. Lansbury, 1997; O’Nuallain et al., 2004; Soto and Pritzkow, 2018). Dynamic Light Scattering Dynamic light scattering (DLS), also known as photon correlation spectroscopy or quasi-elastic light scattering, is METHODS TO EXPERIMENTALLY a spectroscopy technique that measures the fluctuation of EVALUATE GENERAL AGGREGATION intensity of scattered light with time and is routinely applied AND AMYLOID FORMATION to detect protein aggregates, as well as other nanoparticles (Zaccai et al., 2017). Significant experimental evidence must be gathered by Dynamic light scattering allows to derive two basic complementary biophysical methods to determine if a protein characteristics from a protein population. First, the mean or peptide does aggregate and if it is specifically able to form hydrodynamic size, assuming a spherical geometry of the amyloid assemblies. Several computational approaches have been particle, and second, the polydispersity index of the solution. developed to predict aggregation-prone regions in (poly)peptide DLS is highly sensitive to the presence of aggregates because sequences (Belli et al., 2011; Pallarès and Ventura, 2019), light scattering intensity scales with the second power of the but often experimental validation of aggregation propensity mass of the light scattering particle. Therefore, low amounts represents a considerable bottleneck (Grishin et al., 2020). of protein aggregates can be detected when the hydrodynamic Protein self-assembly is a complex process that might result in radii are large enough. When the sample contains both the formation of amorphous aggregates, different oligomers or monomeric and aggregated species, the monomer can only amyloid fibrils, heterogeneous species that often coexist during be adequately detected when the polydispersity index is low the aggregation process. In this section, the methodological and the protein concentration is high. In this case, the particle approaches required to identify general protein aggregation size distribution plot shows multiple peaks indicative of a (see section “Monitoring the Presence of Protein Aggregates”) multimodal distribution. and to distinguish amyloid fibril formation specifically (see sections “Methods to Monitor the Conformational Properties Size Exclusion Chromatography of Amyloid Aggregates,” “Morphology of Amyloid Fibrils,” Size-exclusion chromatography (SEC) offers the possibility to and “Fluorescence Techniques and Tinctorial Properties of identify, collect and determine the relative molecular weight of Amyloids”) are summarized. In particular, we focus on the study the different assemblies in an aggregated sample (Striegel et al., of amyloid fibril formation, which can be addressed from the 2009; Burgess, 2018). Extra care must be exerted when dealing structural (see section “Morphology of Amyloid Fibrils”) and with labile oligomers. These can be potentially disrupted during the kinetic point of view (see section “Fluorescence Techniques the fractionation process, shifting the equilibrium between and Tinctorial Properties of Amyloids”). A minimum set of species as a function of the protein concentration. Moreover, experiments is suggested to satisfy the criteria that generally large insoluble aggregates should be filtered out before analysis define amyloid aggregates according to their conformational, to prevent column clogging. morphological and tinctorial properties. Size-exclusion chromatography becomes a versatile technique when coupled to multiangle light scattering (MALS) and Monitoring the Presence of Protein differential refractive index (dRI) detectors, which allow to Aggregates determine protein concentration, molecular weight (MW), size The formation of protein aggregates can be detected by a and conformation (Folta-Stogniew, 2006). The SEC-MALS collection of orthogonal methods that report on different instrument is calibrated independently of the column and properties of these macromolecular assemblies. does not depend on commercial reference standards, becoming the default method for the estimation of native MW on Light Scattering and Turbidimetry heterogeneous samples (Sahin and Roberts, 2012). The aggregated states of proteins or peptides scatter the light passing through the aggregate solution proportionally to the Intrinsic Fluorescence Spectroscopy size of aggregated particles. Therefore, the course of aggregation The sensitivity of the intrinsic fluorescence signal of tyrosine can be monitored by the measurement of an attenuation of and tryptophan residues to their local environment has incident light beam (turbidimetry) or by integration of an been extensively used to study protein conformation by angle-specific scattering [static light scattering, SLS (Zhao et al., fluorescence spectroscopy. A steady-state fluorescence spectrum 2016)]. Turbidimetry is often taken as a linear descriptor of the is obtained by recording the emission fluorescence intensity of kinetics of aggregation reactions; however, caution should be a sample excited at a fixed wavelength. In proteins, tryptophan taken when comparing different proteins or conditions, since fluorescence is owed to the side-chain indole group, displaying light is scattered as a function of both aggregate size and shape. an absorption maximum at 280 nm and a fluorescence peak SLS is able to determine the relative size of an aggregate by that is solvatochromic, ranging from 300 to 350 nm. Depending measuring the time-averaged intensity of scattered light, usually on the polarity of the local environment, the fluorescence employing the same wavelength for excitation and detection at an emission maximum ranges from ∼308 nm for a tryptophan Frontiers in Molecular Neuroscience | www.frontiersin.org 7 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE fully embedded in a hydrophobic pocket to ∼348–352 nm for algorithm – BeStSel – is especially suited for secondary structure a tryptophan fully exposed to solvent (Ghisaidoobe and Chung, determination of proteins with high β-sheet content, such as 2014). Thus, the fluorescence of tryptophan in monomeric and those found in amyloid protein aggregates (Micsonai et al., 2015). aggregated states of proteins is significantly different (Bobone The CD spectrum in the near UV region (260–320 nm) arises and van de Weert, 2014). The fluorescence of tyrosine is due from the contribution from aromatic amino acids and provides to its side-chain phenolic ring, with excitation and emission information on the tertiary structure, complementary to that maxima at 260 nm and 305 nm, respectively. When compared gathered from intrinsic fluorescence measurements. to tryptophan, the fluorescence signal of tyrosine is less sensitive to environmental changes, as a consequence of the lack of Fourier-Transform Infrared Spectroscopy dipole reorientation in the excited phenol. Still, since tyrosine Fourier transform infrared (FTIR) spectroscopy is especially is on average three times more abundant than tryptophan in suitable to determine the existence of β-sheet secondary polypeptides, its signal also differs significantly between the structures in aggregated samples. The presence of a band aggregated and soluble states of proteins. at 1620–1640 cm−1 in the amide I region of the infrared spectrum is a signature of β-strands. Intramolecular (native Methods to Monitor the Conformational β-sheet) signals fall typically in the 1630–1640 cm−1 range, Properties of Amyloid Aggregates whereas the intermolecular β-sheets characteristic of aggregates Although the nature and self-assembly mechanisms of usually peak at 1620–1630 cm−1 or even lower wavenumbers protein amyloids differ, most of them share a final fibrillar (Hiramatsu and Kitagawa, 2005). morphology highly enriched in intermolecular cross-β The FTIR spectrum is particularly sensitive to the presence structure, whose presence can be confirmed by different of additives in the sample, such as TFA, DMSO and reducing biophysical methods. agents; therefore, the signal of the buffer should be subtracted from that of the sample. Because the vibration signal of water Circular Dichroism Spectroscopy maps in the amide I region (1630 cm−1 ), the FTIR spectrum must Circular dichroism (CD) relies on the differential absorption be obtained by drying out the sample to an hydrated protein film of left and right circularly polarized light. Optically active when using Attenuated Total Reflection sampling devices or by chiral molecules and chemical groups absorb preferentially exchanging H2 O with D2 O for transmission experiments when in one direction of the light. Thus, peptide bonds, aromatic using DTGS (deuterated triglycine sulfate) and MCT (mercuric amino acid side chains and disulfide bonds in proteins act cadmium telluride) detectors (Cristovao et al., 2019). as conformation reporters, providing information about the The peaks in the FTIR spectrum are identified by curve secondary structure composition (α-helix, β-sheet, disorder). In fitting using Fourier self-deconvolution, or/and by analysis of the far UV region of the spectrum (190–260 nm) a maximum the inflection points in the second derivative of the spectrum at 196 nm and a single minimum at 218 nm is attributed to curve. Fitting of the FTIR spectrum does not provide a unique a β-sheet conformation; meanwhile, a maximum at 192 nm solution and thus the fitting parameters become critical: a and two minima at 208 nm and 222 nm are indicative of α- minimum number of peaks should be used, the maxima of the helical content, whereas a minimum at 198 nm correlates with curve-fitted peaks should correspond to the evident maxima in the presence of disordered regions (Cristovao et al., 2019). The the raw data, and, generally, curve-fitted peaks should present shift from an initial native conformation to an amyloid β-sheet similar full-width and half-height values. This technique can structure often exacerbates the minimum at 218 nm (Kelly et al., be used to study and compare protein aggregates obtained 2005; Vadukul et al., 2019). The presence of additional molecules in vivo and in vitro (Shivu et al., 2013). Traditional FTIR in the sample, such as reducing agents, organic molecules analysis only reports on bulk composition and even FTIR and excipients, should be taken into account since they might microspectrometers, despite their extensive applications [e.g., in interfere with the measurement, precluding detection of the cell biology (Bouyanfif et al., 2018)], only provide resolution expected transition. down to 2–5 µm, which is insufficient for the analysis of The far UV spectra of aggregated samples are usually complex individual fibrils. However, by combining IR with atomic force and require the use of deconvolution algorithms (SELCON, microscopy in Infrared Nanospectroscopy it is possible to VARSLC, CDSSTR, K2d2, K2d3, BeStSel, DICHROWEB server) acquire both morphological, nanomechanical and nanoscale- (Provencher and Glockner, 1981; Manavalan and JohnsonJr., level (10–20 nm) chemical IR absorption spectra and maps 1987; Andrade et al., 1993; Sreerama and Woody, 2000; from protein aggregates, single cells and liquid-liquid phase Whitmore and Wallace, 2004; Micsonai et al., 2015) to quantify condensates (Ruggeri et al., 2020). This approach is particularly the contribution of the different secondary structure elements to useful in analyzing complex mixtures with co-existing amyloid, the signal. However, care should be taken with these estimates, amorphous aggregates and pre-fibrillar aggregates with different even in the analysis of the variation of secondary structure secondary structure (Ruggeri et al., 2015, 2016; Galante et al., content, which should be favored over absolute determinations. 2016; Waeytens et al., 2020). The presence of aggregates often leads to distortions in the CD spectra (differential absorption flattening) that might lead Proteinase K and Mixture of Proteases Digestion to errors in the estimation of secondary structure content, as Complementary to the techniques mentioned above, the relative most algorithms are optimized for soluble proteins. Of note, one resistance of amyloid fibrils to proteinase K digestion is Frontiers in Molecular Neuroscience | www.frontiersin.org 8 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE commonly used for their characterization, especially in the case of Amyloid Fibrils”), separation of insoluble species is not of prionic proteins. Proteinase K is a serine protease that cleaves straightforward. An increase in particle mass during protein peptide bonds on the carboxy-terminal side of aromatic and aggregation can be exploited for detection, isolation and aliphatic amino acids. Despite the high processivity of proteinase characterization of both high-molecular weight intermediates K in disordered and regular secondary structures, its proteolytic and final filaments. Several ultracentrifugation protocols have activity is rather limited in cross-β-sheet structures, which are been adopted for analyzing protein aggregates in cellular and characteristic of amyloid fibrils (Kushnirov et al., 2020). The animal models of various human proteinopathies including prion identification of proteinase K-resistant amyloid cores can be protein amyloidosis (Vey et al., 1996), a familiar tauopathy achieved by two main approaches: adding different proteinase (Berger et al., 2007) or sporadic AD (Skrabana et al., 2017). K concentrations at the end-times of aggregation reactions or The ability to differentially isolate protein conformers with following the time-course of the digestion at a given proteinase distinct physical properties by sedimentation at high speed can K concentration. The results of the digestion experiments can be used also in in vitro experiments of protein aggregation. be monitored with several techniques, such as SDS-PAGE Insoluble fibrils and larger aggregates pellet quickly around and protein immunoblotting, mass spectrometry, or electron 14,000–16,000 × g in aqueous buffers, but smaller species microscopy. In some cases, a mixture of proteases can be used such as oligomers or shorter fibril fragments sediment more instead of proteinase K to identify the core of amyloid fibrils slowly even under high-speed centrifugation above 50,000 × g (Selivanova et al., 2016; Surin et al., 2020). (Mok and Howlett, 2006). It is sometimes possible to separate different fibrils based on their morphology. Straight fibrils of X-Ray Diffraction and Scattering Techniques human apolipoprotein ApoC-II, which dominate in the presence X-ray diffraction and small-angle scattering techniques, of phospholipid micelles, sediment more rapidly than flexible including small-angle X-ray scattering (SAXS) and small- ribbons formed in the absence of lipids, making it possible to angle neutron scattering (SANS) are considered gold standard pellet the two populations at 14,500 × g and 350,000 × g, techniques for the analysis of the structural properties of respectively (Mok et al., 2015). Similar approaches have separated amyloid fibrils. linear versus closed-loop fibrils of ApoC-II (Yang et al., 2012), The X-ray fiber diffraction pattern of partially aligned amyloid while complex mixtures of huntingtin aggregates have more fibrils displays a characteristic arrangement with a meridional simply been fractionated by pelleting a mixture of inclusions, reflection (vertical axis) at 4.7–4.8 Å and an equatorial reflection soluble but large oligomers and cell debris from lysed cells (horizontal axis) at 10–12 Å (Figure 1B), which report on the (14,000 × g for 10 min) followed by centrifugation of the distances between β-strands that are associated via hydrogen- resuspended pellet through a desalting column, which allows bonding perpendicular to the fibril axis and between the adjacent separating the soluble oligomers from the insoluble material β-sheets running parallel to the axis, respectively. The meridional (Ormsby et al., 2013). Just as importantly, if the relationship signal is generally very sharp and intense due to the repeating between the molecular weight of fibrillar species and their nature of the β-strands along the fiber axis. The equatorial diffusion coefficient is known (MacRaild et al., 2003), it is possible signal is generally weaker and more diffuse and its position can to estimate the size distribution of fibrils using sedimentation vary depending on the amino acid composition of a constituent velocity analysis (Yang et al., 2012), which measures how quickly peptide (e.g., large aromatic side chains will result in a larger sheet fibrils move through a centrifugal field. It is even possible to spacing). This is the reason why the common amyloid fold is selectively monitor fibril sizes in the presence of non-fibrillar known as cross-β (Morris and Serpell, 2012). Low angle signals components (e.g., chaperones) using fluorescence detection and are often also observed, which arise from chain length and/or fluorophore-labeled fibrils; this allows to assess changes in the protofilament packing. size distribution and lateral association of a given protein (Binger Small-angle scattering techniques are one of the best-suited et al., 2013). As an alternative approach, OptiPrep gradients biophysical approaches to study the conformational properties of have been used to isolate and characterize protein aggregates the different oligomeric forms appearing along an amyloidogenic associating with membrane-like domains (Vey et al., 1996). process and how they evolve into mature fibrils. In theory, one can decompose the time-resolved SAXS and SANS data Morphology of Amyloid Fibrils into the scattering intensity profiles of the individual forms Once the β-sheet content has been confirmed by the methods (structural information), along with their relative populations above, the morphology of the amyloid fibrils should be examined (kinetic information), without the need to isolate the transient or by imaging techniques. co-existing species (Langkilde et al., 2015). Nonetheless, the use of this technique is limited to specific cases because it requires Transmission Electron Microscopy highly concentrated and stable samples, containing only a few The visualization of transmission electron microscope (TEM) oligomeric species simultaneously. images at high resolution provides information on the morphology, homogeneity and size of the amyloid fibrils. Sedimentation Techniques: Separation and Size In order to obtain high-quality TEM data, samples deposited on Distribution carbon-coated copper grids should be negatively stained with a While complex mixtures of aggregated protein can be imaged heavy metal solution [typically 2% (w/v) uranyl acetate solution by a variety of different techniques (see section “Morphology in water] to increase contrast. It should, however, be taken into Frontiers in Molecular Neuroscience | www.frontiersin.org 9 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE account that TEM images may display structural artifacts due to change slightly depending on each particular amyloid structure dehydration and staining during the sample preparation process (Sabate et al., 2013). (Gras et al., 2011). Th-S is a mixture of compounds that results from the methylation of dehydrothiotoluidine with sulfonic acid. As Atomic Force Microscopy such, its molar concentration cannot be accurately calculated, Atomic force microscopy (AFM) allows direct visualization of which in addition to its high fluorescence background make amyloid fibrils in aqueous solutions, providing information about this dye sub-optimal for spectroscopic measurements in vitro their structural and mechanical properties under physiological- (Espargaro et al., 2012). However, Th-S presents the advantage like conditions (Ruggeri et al., 2019). AFM images have of being able to permeate through cell membranes, thus nanometer-resolution, allowing to assess the fibril contour allowing to detect intracellular amyloid aggregates, even in length, width, height and periodicity or the high-order assembly living cells. Accordingly, it has been vastly used in the of single protofilaments into mature fibrils (Adamcik et al., histological staining of amyloids and to image purified amyloid 2012). These properties can be measured at final time- material (Figure 1C). points or directly during the assembly reaction. Besides, it is possible to define the packing scheme and polymorphic state of the amyloid fibrils (e.g., twisted, helical ribbon) by Congo Red measuring the height profile of fibrils along their contour The CR dye has been broadly used to detect amyloid aggregates length and the shape of height profile. An advantage of in tissues and in vitro. The absorption spectra of CR alone and AFM, when compared to TEM, is that it allows measuring in the presence of amyloids are recorded in the visible region of forces and elastic properties of amyloid assemblies with the light spectrum (300–700 nm) and compared (Yakupova et al., piconewton resolution. 2019). The binding of CR to amyloids induces a spectral red shift, Several additional methods, such as solid-state NMR and with maximum absorbance change occurring around 540 nm. As cryo-EM, provide valuable information about the morphology in the case of Th-T assays, the CR spectrophotometric assay is and structural features of amyloid fibrils at high resolution. not quantitative. These methods and their applications to uncover the atomic Birefringence originates from the decomposition of a ray of structures of amyloid fibrils have been extensively covered in light into two rays when it passes through certain anisotropic recent reviews (Iadanza et al., 2018a; Fitzpatrick and Saibil, materials, such as crystals. The fixation of CR molecules 2019). Although key to explore correlations between structure along the axis of the amyloid fibrils usually causes apple- and toxicity, they need advanced equipment and go beyond green birefringence when viewed through cross-polarized light, a minimum requirement for determining that a polypeptide providing an assessment of the amyloid nature of protein assembles into amyloid fibrils. aggregates complementary to CR absorbance measurements. Fluorescence Techniques and Tinctorial Other Amyloid-Staining Molecules Properties of Amyloids The use of Th-T and CR for amyloid detection has limitations. It is well established that regular fibrillar structures have the Many non-amyloid molecules can also exhibit birefringence ability to bind small molecules at their surfaces and cavities with under cross-polarized light in the presence of CR, such as a concomitant alteration of the optical properties of the binding phosphate salts, urea, and other types of fibers like hair. compounds. Th-T and CR are two of these molecules, and have Moreover, CR is a pH indicator and, accordingly, its absorbance been extensively used for the detection of amyloid structure in spectrum is strongly dependent on the solution pH, being useless fibrils and deposits. under acidic conditions. The absorption and emission spectra of some molecules, such as flavins and reduced NAD(P)H Thioflavines overlap with that of Th-T. Also, some polyphenolic anti- The Th-T assay measures the increase in emission fluorescence aggregational compounds, such as curcumin or epigallocatechin, signal of Th-T as amyloid fibrils grow. The enhanced fluorescence exhibit fluorescent properties similar to those of Th-T (Hudson can be detected by spectroscopy or visualized by epifluorescence et al., 2009). In a similar way, the auto-fluorescence generated by or confocal microscopy. In all techniques, the molecule is some aldehyde compounds used as tissue fixatives significantly excited at 445 nm and emission fluorescence is recorded in the interferes with the histological detection of amyloids using 470–500 nm range, usually with a maximum around 482 nm this dye. In light of these problems (Viegas et al., 2007), (Biancalana and Koide, 2010). novel fluorescent dyes with improved sensitivity and specificity Th-T fluorescence enhancement is not a quantitative properties have been developed. parameter since it is strongly dependent on the fibril morphology. As Th-T, the ProteoStat dye is a rotor molecule that In some cases, amyloid fibrils may be present, but do not display intercalates into the cross-β structure, leading to a strong red fluorescence because the rotation movement of the Th-T fluorescence with excitation and emission maxima at 500 nm molecule is not sufficiently impeded. On these occasions, Th-T and 600 nm, respectively (Shen et al., 2011; Navarro and fluorescence anisotropy provides an alternative technique for Ventura, 2014). Compared to Th-T, ProteoStat staining produces the study of amyloid aggregation (Sabate and Saupe, 2007). a stronger fluorescent signal with a wider linear dynamic range Also, the maximum excitation and emission wavelengths may in a broad range of pH values (4–10) and avoids spectral Frontiers in Molecular Neuroscience | www.frontiersin.org 10 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE TABLE 2 | Experimental details that should be reported for protein aggregation assays. Reported parameter Information that should be reported Sample stock Sample source and storage Description of protein source – commercial (supplier and reference), recombinant (including conditions expression and purification), or other (e.g., extracted from tissue); description of how the protein or peptide stock was preserved (lyophilized or stored in solution); if stored in solution indicate storage buffer, sample concentration, storage temperature, freezing/thawing conditions, as well as material and supplier of vials used for sample storage. Sample purity and Indicate the methods (SDS-PAGE, HPLC, etc.) used to determine sample purity, as well as those concentration used to determine the exact concentration of the stock sample in the assay. The purity of the sample stock as evaluated by SDS-PAGE should be indicated and higher than 95%. Indicate whether the sample was tested for the presence of “invisible” components such as nucleic acids. Sample concentration Indicate the final concentration of the sample in the assay. Sample preparation prior to Mention if the protein/peptide stocks were pre-treated, filtered or centrifuged; these procedures are assay recommended to remove pre-aggregated forms. Assay buffer Indicate the composition, concentrations and pH value of the buffer or solvent stock. Assay volume The final volume of the sample, as well as the volume of the tube/well in which the aggregation assay is performed should be reported since interfaces (liquid-solid and liquid-air) influence protein aggregation. The method used to prevent sample evaporation should also be mentioned when using small sample volumes. Additives List ALL components that could be found in the aggregation assay (even those that may exist in minute amounts). For example, if an aggregation modulator is dissolved in DMSO, it is important that all samples (including control) contain the same final concentration of DMSO. Experimental procedures, Plate or vial Report the type of material and geometry of the vial or microplate well (binding versus non-binding type of equipment used and surface, bottom versus top-reading, square versus round bottom, etc.) used in the assay. Include system-dependent parameters information on the microplate/vial material, including supplier and reference code. Temperature Report the assay temperature. In particular, mention if samples/buffer are preincubated at the assay temperature: fluorescence drifts may be observed at the beginning of the experiment resulting from temperature shifts. Agitation Indicate whether an orbital shaker, a thermomixer or magnetic stirring was used (describing shape, size and material of stirrers), the type of shaking (orbital, linear, etc.) and speed. If applicable, indicate if beads were included (material, size, number of beads). Measurement cycles and the pre-shaking agitation procedures should be clearly specified. Time The total duration of the aggregation reaction should be reported. Equipment for measuring Indicate the device make and model, control software, and general settings used (e.g., filter aggregation kinetics and its bandpass and bandwidth/monochromator settings). settings Reporters Provide the exact amount of reporter (if used) employed for measurement and any pre-treatment of the sample. Data analysis and raw data Indicate the software (including version) used for image and data analysis and specify the equation that should be preserved applied for fitting kinetic data obtained from aggregation assays. for publication overlap with the autofluorescence signal of membranes and co- maxima occur, due to their binding to the hydrophobic factors. ProteoStat is particularly indicated if working in the clusters exposed in such assemblies. ANS and bis-ANS are presence of RNA, when the use of Th-T is unsuited, because it particularly useful in the detection of the low concentrations of produces huge distortions of the baseline (Sugimoto et al., 2015; protein aggregates populating the early stages of the reaction Liu et al., 2017). (de Groot et al., 2007). Heptamer-formyl thiophene acetic acid (hFTAA) is yet A Th-T derivative, called Pittsburgh compound B (PiB) is another amyloid detection dye. hFTAA exhibits distinct shifts able to enter the brain and is used as a radioactive tracer in its emission spectra when bound to different amyloid species for in vivo PET imaging of amyloid beta pathology in AD and polymorphs, finding application in the differentiation of (Klunk et al., 2004). amyloid (sub)types in vivo and in monitoring changes of amyloid structure and composition over time (Klingstedt et al., 2011, 2013; Sjolander et al., 2016). GUIDELINES FOR REPORTING PROTEIN Molecules such as 1-anilinonaphthalene-8-sulfonate (ANS) AGGREGATION EXPERIMENTS: THE and its dimeric analog 4,40 -bis-1-anilinonaphthalene-8-sulfonate MIRRAGGE TABLE (Bis-ANS) are barely fluorescent in polar solvent, but become highly fluorescent in an apolar environment. In the presence Given the diversity of oligomeric species and folding states, of aggregates a blue shift and an increase of their emission the choice of protein purification and handling procedures Frontiers in Molecular Neuroscience | www.frontiersin.org 11 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE FIGURE 3 | Aggregation of human ataxin-3 (UniProt accession code: P54252-2). (A) Schematic representation of the structural organization of ataxin-3, with the N-terminal globular Josephin domain (JD) and the flexible C-terminal tail encompassing the ubiquitin-interacting motifs (UIMs) and the polyQ tract containing 13 glutamine residues (13Q); (B) Representative graph of human ataxin-3 aggregation at 37◦ C monitored by Th-T fluorescence. (C) TEM image of ataxin-3 protofibrils at the end of the aggregation assay (∼60 h). plays a crucial role in determining where in the aggregation information or experimental detail found to be important for map (Figure 2) are the departing and arrival points located, the outcome of aggregation. This is acchieved by including a and by what pathways are they connected. Therefore, we flexible field of “additional key information” at the end of the propose a set of guidelines for presentation and publication tabular descriptions of the purification protocol and of the of data related to polypeptide aggregation that ensure an aggregation assay, wherein relevant remarks concerning the adequate and accurate description of the experimental results. aggregation state of the protein, sample collection procedures in The relevant information that we propose to be reported gel filtration chromatography, the occurrence of contaminants is divided into two parts: sample preparation and quality or co-solvents, the procedure adopted for removal of air bubbles, and incubation conditions. Ideally, the methods section a critical sequence of reagent addition, etc., can be emphasized. of each publication should contain the information briefly A comprehensive characterization of the aggregation assay described in Table 2. in terms of the total volume of reaction, plate/cuvette/vial A pre-filled MIRRAGGE table using the aggregation of geometry and material, method of evaporation control, size the protein ataxin-3 (Figure 3) (Gales et al., 2005; Almeida and material of beads (if present), and type of agitation is et al., 2015) and of a peptide from the Mot3 yeast prion required on account of the effects of interfaces and shear (Sant’Anna et al., 2016; Fernandez et al., 2017) as examples flow on protein aggregation (Giehm and Otzen, 2010; Bekard is provided as Supplementary Information (Supplementary et al., 2011; Yoshimura et al., 2012; Ferreira et al., 2016; Koepf File MIRRAGGE_template.xls) illustrating the minimum et al., 2018). As a major determinant of phase separation, information that we suggest should be included to ensure protein concentration is discriminated (i) before storage of reproducible (poly)peptide aggregation experiments. A large the purified protein, (ii) immediately before aggregation, e.g., part of the minimum information required in the MIRRAGGE after the final filtration step, and (iii) during aggregation. The table concerns the sample preparation steps that precede concentration of protein or peptide is provided, together with the aggregation assay itself. Starting by the unambiguous the information about MW, extinction coefficient and the identification of the protein molecule, it is particularly quantification method adopted. Freeze-thawing stress can lead important for curators of public databases that the UniProt to the cold denaturation of the protein, its adsorption at the accession number, the species of origin and, if applicable, the ice-liquid interface with subsequent partial denaturation and presence of affinity tags are provided for the molecules under aggregation, or to drastic pH changes in buffers like sodium investigation (Orchard et al., 2007; Trewhella et al., 2017). phosphate or succinate (Hawe et al., 2009). It is recommended For synthetic peptides, the chemical nature of the N and C to control the quality of thawed samples to judge whether termini (whether capped or uncapped) should be made clear additional polishing steps are required before the aggregation as well as the buffering salt (e.g., TFA). Besides the essential assay. Cohen et al. (2013) reported two rounds of gel filtration information about the sample source, namely the protein post-thawing on purified Aβ42 peptide to ensure pure monomer expression system, the purification and polishing protocols, at the beginning of the kinetic assay. In the case of ataxin-3, or, in the case of samples obtained from commercial sources, increased reproducibility is achieved by adding a re-polishing the catalog number and the name and location of the supplier, step immediately before the aggregation assay, separating the MIRRAGGE table can also encompass any additional the predominant monomer from contaminant oligomeric Frontiers in Molecular Neuroscience | www.frontiersin.org 12 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE species putatively formed during the freezing-thawing process FUNDING (Silva et al., 2017). In conclusion, this review aims to provide an overview This work was supported by (i) the European Regional of the biophysical principles underlying protein self-assembly Development Fund (ERDF) through the COMPETE in its multiple shapes, and compiles detailed information 2020—Operacional Programme for Competitiveness and on the experimental validation of polypeptide aggregation, Internationalisation (POCI), Portugal 2020, and by Portuguese with a focus on amyloid fibril formation. It results from funds through FCT—Fundação para a Ciência e a Tecnologia several discussions on this topic in the context of the COST (FCT/MCTES) in the framework of grants POCI-01-0145- action BM1405 [Non-globular proteins – from sequence FEDER-031173, POCI-01-0145-FEDER-007274, POCI-01-0145- to structure, to applications in molecular physiopathology FEDER-031323 (“Institute for Research and Innovation in Health (NGP-net)]. We aim to endow both experienced researchers Sciences”), UID/Multi/04046/2013 (BioISI) and PTDC/NEU- and newcomers to the field with a set of guidelines NMC/2138/2014 (to CMG). SV was funded by the Spanish to enhance reproducibility in (poly)peptide aggregation Ministry of Economy and Competitiveness (BIO2016-78310-R) experiments. The MIRRAGGE table template compiles a and by ICREA (ICREA-Academia 2015). ZG and ZB were funded recommended reporting framework that is expected to provide by Slovak research agentures VEGA 02/0145/17, 02/0030/18 and a minimum set of information required for replicating APVV-18-0284. RS was funded by VEGA 02/0163/19. DEO was protein aggregation experiments and swiftly discriminate funded by the Lundbeck Foundation (grant no. R276-2018-671) protein assembly into amyloid fibrils from amorphous and the Independent Research Foundation Denmark | Natural protein aggregation. Sciences (grant no. 8021-00208B). AP research was supported by UK Dementia Research Institute (RE1 3556) and by ARUK (ARUK-PG2019B-020). AUTHOR CONTRIBUTIONS PMM and SN revised the literature and wrote the manuscript. ACKNOWLEDGMENTS AS, ZS, FF, and MFP designed the template MIRRAGGE table, with contributions from all authors. PJBP, FP, ZB, The authors acknowledge the fruitful discussions with the MB, OVG, ZG, CMG, AP, LCS, RS, VS, MZ, and DEO scientists that participated in NGP-net (COST Action BM1405) participated in the discussion of the minimum requirements scientific activities. for publication of experimental data on (poly)peptide aggregation, and contributed for the preparation of the list of experimental methods and details that should be SUPPLEMENTARY MATERIAL reported for protein aggregation assays. SM-R and SV designed the review focus and finalized the manuscript. The Supplementary Material for this article can be found All authors contributed to the article and approved the online at: https://www.frontiersin.org/articles/10.3389/fnmol. submitted version. 2020.582488/full#supplementary-material REFERENCES unsupervised learning neural network. Protein Eng. 6, 383–390. doi: 10.1093/ protein/6.4.383 Adamcik, J., Lara, C., Usov, I., Jeong, J. S., Ruggeri, F. S., Dietler, G., et al. (2012). Arimon, M., Grimminger, V., Sanz, F., and Lashuel, H. A. (2008). Hsp104 targets Measurement of intrinsic properties of amyloid fibrils by the peak force QNM multiple intermediates on the amyloid pathway and suppresses the seeding method. Nanoscale 4, 4426–4429. doi: 10.1039/c2nr30768e capacity of Aβ fibrils and protofibrils. J. Mol. Biol. 384, 1157–1173. doi: 10.1016/ Aguzzi, A., and Altmeyer, M. (2016). Phase separation: linking cellular j.jmb.2008.09.063 compartmentalization to disease. Trends Cell Biol. 26, 547–558. doi: 10.1016/ Baldwin, A. J., Knowles, T. P. J., Tartaglia, G. G., Fitzpatrick, A. W., Devlin, j.tcb.2016.03.004 G. L., Shammas, S. L., et al. (2011). Metastability of native proteins and the Aguzzi, A., and O’Connor, T. (2010). Protein aggregation diseases: pathogenicity phenomenon of amyloid formation. J. Am. Chem. Soc. 133, 14160–14163. doi: and therapeutic perspectives. Nat. Rev. Drug Discov. 9, 237–248. doi: 10.1038/ 10.1021/ja2017703 nrd3050 Banani, S. F., Lee, H. O., Hyman, A. A., and Rosen, M. K. (2017). Biomolecular Ahmed, M., Davis, J., Aucoin, D., Sato, T., Ahuja, S., Aimoto, S., et al. condensates: organizers of cellular biochemistry. Nat. Rev. Mol. Cell Biol. 18, (2010). Structural conversion of neurotoxic amyloid-β1–42 oligomers 285–298. doi: 10.1038/nrm.2017.7 to fibrils. Nat. Struct. Mol. Biol. 17, 561–567. doi: 10.1038/nsmb. Bekard, I. B., Asimakis, P., Bertolini, J., and Dunstan, D. E. (2011). The effects 1799 of shear flow on protein structure and function. Biopolymers 95, 733–745. Alberti, S. (2017). The wisdom of crowds: regulating cell function through doi: 10.1002/bip.21646 condensed states of living matter. J. Cell Sci. 130, 2789–2796. doi: 10.1242/jcs. Belli, M., Ramazzotti, M., and Chiti, F. (2011). Prediction of amyloid aggregation 200295 in vivo. EMBO Rep. 12, 657–663. doi: 10.1038/embor.2011.116 Almeida, B., Abreu, I. A., Matos, C. A., Fraga, J. S., Fernandes, S., Macedo, Ben-Gedalya, T., Moll, L., Bejerano-Sagie, M., Frere, S., Cabral, W. A., Friedmann- M. G., et al. (2015). SUMOylation of the brain-predominant Ataxin-3 isoform Morvinski, D., et al. (2015). Alzheimer’s disease-causing proline substitutions modulates its interaction with p97. Biochim. Biophys. Acta 1852, 1950–1959. lead to presenilin 1 aggregation and malfunction. EMBO J. 34, 2820–2839. doi: 10.1016/j.bbadis.2015.06.010 doi: 10.15252/embj.201592042 Andrade, M. A., Chacon, P., Merelo, J. J., and Moran, F. (1993). Evaluation of Benson, M. D., Buxbaum, J. N., Eisenberg, D. S., Merlini, G., Saraiva, secondary structure of proteins from UV circular dichroism spectra using an M. J. M., Sekijima, Y., et al. (2018). Amyloid nomenclature 2018: Frontiers in Molecular Neuroscience | www.frontiersin.org 13 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE recommendations by the International Society of Amyloidosis (ISA) protofibril formation. J. Phys. Chem. B 118, 1189–1197. doi: 10.1021/jp4 nomenclature committee. Amyloid 25, 215–219. doi: 10.1080/13506129.2018.15 083294 49825 Doyle, C. M., Rumfeldt, J. A., Broom, H. R., Broom, A., Stathopulos, P. B., Vassall, Berger, Z., Roder, H., Hanna, A., Carlson, A., Rangachari, V., Yue, M., et al. (2007). K. A., et al. (2013). Energetics of oligomeric protein folding and association. Accumulation of pathological tau species and memory loss in a conditional Arch. Biochem. Biophys. 531, 44–64. doi: 10.1016/j.abb.2012.12.005 model of tauopathy. J. Neurosci. 27, 3650–3662. doi: 10.1523/jneurosci.0587- Eisenberg, D. S., and Sawaya, M. R. (2017). Structural studies of amyloid proteins 07.2007 at the molecular level. Annu. Rev. Biochem. 86, 69–95. doi: 10.1146/annurev- Biancalana, M., and Koide, S. (2010). Molecular mechanism of Thioflavin-T biochem-061516-045104 binding to amyloid fibrils. Biochim. Biophys. Acta 1804, 1405–1412. doi: 10. Englander, S. W., Mayne, L., Kan, Z.-Y., and Hu, W. (2016). Protein folding—how 1016/j.bbapap.2010.04.001 and why: by hydrogen exchange, fragment separation, and mass spectrometry. Binger, K. J., Ecroyd, H., Yang, S., Carver, J. A., Howlett, G. J., and Griffin, M. D. W. Annu. Rev. Biophys. 45, 135–152. doi: 10.1146/annurev-biophys-062215- (2013). Avoiding the oligomeric state: αB-crystallin inhibits fragmentation 011121 and induces dissociation of apolipoprotein C-II amyloid fibrils. FASEB J. 27, Espargaro, A., Sabate, R., and Ventura, S. (2012). Thioflavin-S staining 1214–1222. doi: 10.1096/fj.12-220657 coupled to flow cytometry. A screening tool to detect in vivo Bobone, S., and van de Weert, M. (2014). A reassessment of synchronous protein aggregation. Mol. Biosyst. 8, 2839–2844. doi: 10.1039/c2mb fluorescence in the separation of Trp and Tyr contributions in protein emission 25214g and in the determination of conformational changes. J. Mol. Struct. 1077, 68–76. Fairfoul, G., Mcguire, L. I., Pal, S., Ironside, J. W., Neumann, J., Christie, S., doi: 10.1016/j.molstruc.2014.01.004 et al. (2016). Alpha-synuclein RT-QuIC in the CSF of patients with alpha- Boeynaems, S., Alberti, S., Fawzi, N. L., Mittag, T., Polymenidou, M., Rousseau, F., synucleinopathies. Ann. Clin. Transl. Neurol. 3, 812–818. doi: 10.1002/acn3. et al. (2018). Protein phase separation: a new phase in cell biology. Trends Cell 338 Biol. 28, 420–435. doi: 10.1016/j.tcb.2018.02.004 Falahati, H., and Haji-Akbari, A. (2019). Thermodynamically driven assemblies Bouyanfif, A., Liyanage, S., Hequet, E., Moustaid-Moussa, N., and Abidi, N. (2018). and liquid–liquid phase separations in biology. Soft Matter 15, 1135–1154. Review of FTIR microspectroscopy applications to investigate biochemical doi: 10.1039/c8sm02285b changes in C. elegans. Vib. Spectrosc. 96, 74–82. doi: 10.1016/j.vibspec.2018.03. Falcon, B., Cavallini, A., Angers, R., Glover, S., Murray, T. K., Barnham, L., 001 et al. (2015). Conformation determines the seeding potencies of native and Burgess, R. R. (2018). A brief practical review of size exclusion chromatography: recombinant tau aggregates. J. Biol. Chem. 290, 1049–1065. doi: 10.1074/jbc. rules of thumb, limitations, and troubleshooting. Protein Expr. Purif. 150, m114.589309 81–85. doi: 10.1016/j.pep.2018.05.007 Farmer, K., Gerson, J. E., and Kayed, R. (2017). Oligomer formation and cross- Cao, Q., Boyer, D. R., Sawaya, M. R., Ge, P., and Eisenberg, D. S. (2019). Cryo-EM seeding: the new frontier. Isr. J. Chem. 57, 665–673. doi: 10.1002/ijch.201600103 structures of four polymorphic TDP-43 amyloid cores. Nat. Struct. Mol. Biol. Fernandez, M. R., Batlle, C., Gil-Garcia, M., and Ventura, S. (2017). Amyloid cores 26, 619–627. doi: 10.1038/s41594-019-0248-4 in prion domains: key regulators for prion conformational conversion. Prion Carvalho, A.L., Silva, A., and Macedo-Ribeiro, S. (2018). “Polyglutamine- 11, 31–39. doi: 10.1080/19336896.2017.1282020 Independent Features in Ataxin-3 Aggregation and Pathogenesis of Machado- Ferreira, C., Rocha, F. A., Damas, A. M., and Martins, P. M. (2016). The finding Joseph Disease,” in Polyglutamine Disorders, eds C. Nóbrega and L. Pereira De of nondissolving lysozyme crystals and its significance for the study of hard- Almeida. (Cham: Springer), 275-288. doi: 10.3389/fphar.2020.01311 to-crystallize biological macromolecules. Cryst. Growth Des. 16, 4285–4291. Chatani, E., and Yamamoto, N. (2018). Recent progress on understanding the doi: 10.1021/acs.cgd.6b00334 mechanisms of amyloid nucleation. Biophys. Rev. 10, 527–534. doi: 10.1007/ Fitzpatrick, A. W., and Saibil, H. R. (2019). Cryo-EM of amyloid fibrils and cellular s12551-017-0353-8 aggregates. Curr. Opin. Struct. Biol. 58, 34–42. doi: 10.1016/j.sbi.2019.05.003 Chiti, F., and Dobson, C. M. (2006). Protein misfolding, functional amyloid, Fitzpatrick, A. W. P., Debelouchina, G. T., Bayro, M. J., Clare, D. K., Caporini, and human disease. Annu. Rev. Biochem. 75, 333–366. doi: 10.1146/annurev. M. A., Bajaj, V. S., et al. (2013). Atomic structure and hierarchical assembly biochem.75.101304.123901 of a cross-β amyloid fibril. Proc. Natl. Acad. Sci. U.S.A. 110, 5468–5473. doi: Chiti, F., and Dobson, C. M. (2017). Protein misfolding, amyloid formation, 10.1073/pnas.1219476110 and human disease: a summary of progress over the last decade. Annu. Rev. Folta-Stogniew, E. (2006). Oligomeric states of proteins determined by size- Biochem. 86, 27–68. doi: 10.1146/annurev-biochem-061516-045115 exclusion chromatography coupled with light scattering, absorbance, and Cohen, S. I. A., Linse, S., Luheshi, L. M., Hellstrand, E., White, D. A., Rajah, refractive index detectors. Methods Mol. Biol. 328, 97–112. doi: 10.1385/1- L., et al. (2013). Proliferation of amyloid-β42 aggregates occurs through a 59745-026-x:97 secondary nucleation mechanism. Proc. Natl. Acad. Sci. U.S.A. 110, 9758–9763. Gade Malmos, K., Blancas-Mejia, L. M., Weber, B., Buchner, J., Ramirez-Alvarado, doi: 10.1073/pnas.1218402110 M., Naiki, H., et al. (2017). ThT 101: a primer on the use of thioflavin T to Crespo, R., Rocha, F. A., Damas, A. M., and Martins, P. M. (2012). A investigate amyloid formation. Amyloid 24, 1–16. doi: 10.1080/13506129.2017. generic crystallization-like model that describes the kinetics of amyloid 1304905 fibril formation. J. Biol. Chem. 287, 30585–30594. doi: 10.1074/jbc.m112. Galante, D., Ruggeri, F. S., Dietler, G., Pellistri, F., Gatta, E., Corsaro, A., et al. 375345 (2016). A critical concentration of N-terminal pyroglutamylated amyloid beta Crespo, R., Villar-Alvarez, E., Taboada, P., Rocha, F. A., Damas, A. M., and Martins, drives the misfolding of Ab1-42 into more toxic aggregates. Int. J. Biochem. Cell P. M. (2016). What can the kinetics of amyloid fibril formation tell about off- Biol. 79, 261–270. doi: 10.1016/j.biocel.2016.08.037 pathway aggregation? J. Biol. Chem. 291, 2018–2032. doi: 10.1074/jbc.m115. Gales, L., Cortes, L., Almeida, C., Melo, C. V., Costa, M. C., Maciel, P., et al. (2005). 699348 Towards a structural understanding of the fibrillization pathway in Machado- Cristovao, J. S., Henriques, B. J., and Gomes, C. M. (2019). Biophysical Joseph’s disease: trapping early oligomers of non-expanded ataxin-3. J. Mol. and spectroscopic methods for monitoring protein misfolding and amyloid Biol. 353, 642–654. doi: 10.1016/j.jmb.2005.08.061 aggregation. Methods Mol. Biol. 1873, 3–18. doi: 10.1007/978-1-4939-8820-4_1 Gazit, E. (2002). The “Correctly Folded” state of proteins: Is it a metastable state? Darling, A. L., Liu, Y., Oldfield, C. J., and Uversky, V. N. (2018). Intrinsically Angew. Chem. Int. Ed. Engl. 41, 257–259. doi: 10.1002/1521-3773(20020118)41: disordered proteome of human membrane-less organelles. Proteomics 2<257::aid-anie257>3.0.co;2-m 18:e1700193. doi: 10.1002/pmic.201700193 Ghisaidoobe, A. B., and Chung, S. J. (2014). Intrinsic tryptophan fluorescence de Groot, N. S., Parella, T., Aviles, F. X., Vendrell, J., and Ventura, S. (2007). Ile-phe in the detection and analysis of proteins: a focus on Forster resonance dipeptide self-assembly: clues to amyloid formation. Biophys. J. 92, 1732–1741. energy transfer techniques. Int. J. Mol. Sci. 15, 22518–22538. doi: 10.3390/ doi: 10.1529/biophysj.106.096677 ijms151222518 Dovidchenko, N. V., Finkelstein, A. V., and Galzitskaya, O. V. (2014). How to Giehm, L., and Otzen, D. E. (2010). Strategies to increase the reproducibility determine the size of folding nuclei of protofibrils from the concentration of protein fibrillization in plate reader assays. Anal. Biochem. 400, 270–281. dependence of the rate and lag-time of aggregation. I. Modeling the amyloid doi: 10.1016/j.ab.2010.02.001 Frontiers in Molecular Neuroscience | www.frontiersin.org 14 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE Giri Rao, V. V. H., and Gosavi, S. (2018). On the folding of a structurally Johnston, J. A., Ward, C. L., and Kopito, R. R. (1998). Aggresomes: a cellular complex protein to its metastable active state. Proc. Natl. Acad. Sci. U.S.A. 115, response to misfolded proteins. J. Cell Biol. 143, 1883–1898. doi: 10.1083/jcb. 1998–2003. doi: 10.1073/pnas.1708173115 143.7.1883 Goedert, M., Falcon, B., Zhang, W., Ghetti, B., and Scheres, S. H. W. (2018). Kelly, S. M., Jess, T. J., and Price, N. C. (2005). How to study proteins by circular Distinct conformers of assembled Tau in Alzheimer’s and Pick’s diseases. dichroism. Biochim. Biophys. Acta 1751, 119–139. doi: 10.1016/j.bbapap.2005. Cold Spring Harb. Symp. Quant. Biol. 83, 163–171. doi: 10.1101/sqb.2018.83. 06.005 037580 Khurana, R., Ionescu-Zanetti, C., Pope, M., Li, J., Nielson, L., Ramírez-Alvarado, González, N., Arcos-López, T., König, A., Quintanar, L., Menacho Márquez, M., et al. (2003). A general model for amyloid fibril assembly based on M., Outeiro, T. F., et al. (2019). Effects of alpha-synuclein posttranslational morphological studies using atomic force microscopy. Biophys. J. 85, 1135– modifications on metal binding. J. Neurochem. 150, 507–521. doi: 10.1111/jnc. 1144. doi: 10.1016/s0006-3495(03)74550-0 14721 Klingstedt, T., Aslund, A., Simon, R. A., Johansson, L. B., Mason, J. J., Nystrom, Gosal, W. S., Morten, I. J., Hewitt, E. W., Smith, D. A., Thomson, N. H., and S., et al. (2011). Synthesis of a library of oligothiophenes and their utilization as Radford, S. E. (2005). Competing pathways determine fibril morphology in fluorescent ligands for spectral assignment of protein aggregates. Org. Biomol. the self-assembly of β2-microglobulin into amyloid. J. Mol. Biol. 351, 850–864. Chem. 9, 8356–8370. doi: 10.1039/c1ob05637a doi: 10.1016/j.jmb.2005.06.040 Klingstedt, T., Blechschmidt, C., Nogalska, A., Prokop, S., Haggqvist, B., Gras, S. L., Waddington, L. J., and Goldie, K. N. (2011). Transmission electron Danielsson, O., et al. (2013). Luminescent conjugated oligothiophenes for microscopy of amyloid fibrils. Methods Mol. Biol. 752, 197–214. doi: 10.1007/ sensitive fluorescent assignment of protein inclusion bodies. Chembiochem 14, 978-1-60327-223-0_13 607–616. doi: 10.1002/cbic.201200731 Gremer, L., Schölzel, D., Schenk, C., Reinartz, E., Labahn, J., Ravelli, R. B. G., Klunk, W. E., Engler, H., Nordberg, A., Wang, Y., Blomqvist, G., Holt, D. P., et al. (2017). Fibril structure of amyloid-β(1–42) by cryo–electron microscopy. et al. (2004). Imaging brain amyloid in Alzheimer’s disease with Pittsburgh Science 358, 116–119. doi: 10.1126/science.aao2825 Compound-B. Ann. Neurol. 55, 306–319. doi: 10.1002/ana.20009 Grishin, S. Y., Deryusheva, E. I., Machulin, A. V., Selivanova, O. M., Glyakina, Kodali, R., and Wetzel, R. (2007). Polymorphism in the intermediates and products A. V., Gorbunova, E. Y., et al. (2020). Amyloidogenic propensities of ribosomal of amyloid assembly. Curr. Opin. Struct. Biol. 17, 48–57. doi: 10.1016/j.sbi.2007. S1 proteins: bioinformatics screening and experimental checking. Int. J. Mol. 01.007 Sci. 21:5199. doi: 10.3390/ijms21155199 Koepf, E., Eisele, S., Schroeder, R., Brezesinski, G., and Friess, W. (2018). Guo, J. L., Covell, D. J., Daniels, J. P., Iba, M., Stieber, A., Zhang, B., et al. Notorious but not understood: how liquid-air interfacial stress triggers protein (2013). Distinct alpha-synuclein strains differentially promote tau inclusions in aggregation. Int. J. Pharm. 537, 202–212. doi: 10.1016/j.ijpharm.2017.12.043 neurons. Cell 154, 103–117. doi: 10.1016/j.cell.2013.05.057 Kushnirov, V. V., Dergalev, A. A., and Alexandrov, A. I. (2020). Proteinase K Haass, C., and Selkoe, D. J. (2007). Soluble protein oligomers in neurodegeneration: resistant cores of prions and amyloids. Prion 14, 11–19. doi: 10.1080/19336896. lessons from the Alzheimer’s amyloid β-peptide. Nat. Rev. Mol. Cell Biol. 8, 2019.1704612 101–112. doi: 10.1038/nrm2101 Langkilde, A. E., Morris, K. L., Serpell, L. C., Svergun, D. I., and Vestergaard, Harper, J. D., and Peter, T., and Lansbury, J. (1997). Models of amyloid seeding B. (2015). The architecture of amyloid-like peptide fibrils revealed by in Alzheimer’s disease and scrapie: mechanistic truths and physiological X-ray scattering, diffraction and electron microscopy. Acta Crystallogr. consequences of the time-dependent solubility of amyloid proteins. Annu. Rev. Sect. D Biol. Crystallogr. 71, 882–895. doi: 10.1107/s1399004715 Biochem. 66, 385–407. doi: 10.1146/annurev.biochem.66.1.385 001674 Hasecke, F., Miti, T., Perez, C., Barton, J., Schölzel, D., Gremer, L., et al. (2018). Lashuel, H. A., Wurth, C., Woo, L., and Kelly, J. W. (1999). The most pathogenic Origin of metastable oligomers and their effects on amyloid fibril self-assembly. transthyretin variant, L55P, forms amyloid fibrils under acidic conditions and Chem. Sci. 9, 5937–5948. doi: 10.1039/c8sc01479e protofilaments under physiological conditions. Biochemistry 38, 13560–13573. Hawe, A., Kasper, J. C., Friess, W., and Jiskoot, W. (2009). Structural properties of doi: 10.1021/bi991021c monoclonal antibody aggregates induced by freeze–thawing and thermal stress. Liu, S., Peng, P., Wang, H., Shi, L., and Li, T. (2017). Thioflavin T binds dimeric Eur. J. Pharm. Sci. 38, 79–87. doi: 10.1016/j.ejps.2009.06.001 parallel-stranded GA-containing non-G-quadruplex DNAs: a general approach Hiramatsu, H., and Kitagawa, T. (2005). FT-IR approaches on amyloid fibril to lighting up double-stranded scaffolds. Nucleic Acids Res. 45, 12080–12089. structure. Biochim. Biophys. Acta 1753, 100–107. doi: 10.1016/j.bbapap.2005.07. doi: 10.1093/nar/gkx942 008 Lorenzen, N., Nielsen, S. B., Buell, A. K., Kaspersen, J. D., Arosio, P., Vad, B. S., Hudson, S. A., Ecroyd, H., Kee, T. W., and Carver, J. A. (2009). The thioflavin T et al. (2014). The role of stable α-synuclein oligomers in the molecular events fluorescence assay for amyloid fibril detection can be biased by the presence of underlying amyloid formation. J. Am. Chem. Soc. 136, 3859–3868. doi: 10.1021/ exogenous compounds. FEBS J. 276, 5960–5972. doi: 10.1111/j.1742-4658.2009. ja411577t 07307.x MacRaild, C. A., Hatters, D. M., Lawrence, L. J., and Howlett, G. J. (2003). Hurshman, A. R., White, J. T., Powers, E. T., and Kelly, J. W. (2004). Transthyretin Sedimentation velocity analysis of flexible macromolecules: self-association aggregation under partially denaturing conditions is a downhill polymerization. and tangling of amyloid fibrils. Biophys. J. 84, 2562–2569. doi: 10.1016/s0006- Biochemistry 43, 7365–7381. doi: 10.1021/bi049621l 3495(03)75061-9 Iadanza, M. G., Jackson, M. P., Hewitt, E. W., Ranson, N. A., and Radford, Mahler, H.-C., Friess, W., Grauschopf, U., and Kiese, S. (2009). Protein aggregation: S. E. (2018a). A new era for understanding amyloid structures and pathways, induction factors and analysis. J. Pharm. Sci. 98, 2909–2934. doi: disease. Nat. Rev. Mol. Cell. Biol. 19, 755–773. doi: 10.1038/s41580-01 10.1002/jps.21566 8-0060-8 Makin, O. S., and Serpell, L. C. (2005). Structures for amyloid fibrils. FEBS J. 272, Iadanza, M. G., Silvers, R., Boardman, J., Smith, H. I., Karamanos, T. K., 5950–5961. doi: 10.1111/j.1742-4658.2005.05025.x Debelouchina, G. T., et al. (2018b). The structure of a β2-microglobulin fibril Malmendal, A., Underhaug, J., Otzen, D. E., and Nielsen, N. C. (2010). Fast suggests a molecular basis for its amyloid polymorphism. Nat. Commun. mapping of global protein folding states by multivariate NMR: a GPS for 9:4517. doi: 10.1038/s41467-018-06761-6 proteins. PLoS One 5:e10262. doi: 10.1371/journal.pone.0010262 Ianiro, A., Wu, H., Van Rijt, M. M. J., Vena, M. P., Keizer, A. D. A., Esteves, A. C. C., Manavalan, P., and Johnson, W. C. Jr. (1987). Variable selection method improves et al. (2019). Liquid–liquid phase separation during amphiphilic self-assembly. the prediction of protein secondary structure from circular dichroism spectra. Nat. Chem. 11, 320–328. doi: 10.1038/s41557-019-0210-4 Anal. Biochem. 167, 76–85. doi: 10.1016/0003-2697(87)90135-7 Ingelsson, M. (2016). Alpha-synuclein oligomers—neurotoxic molecules in Mateju, D., Franzmann, T. M., Patel, A., Kopach, A., Boczek, E. E., Maharana, Parkinson’s disease and other lewy body disorders. Front. Neurosci. 10:408. S., et al. (2017). An aberrant phase transition of stress granules triggered doi: 10.3389/fnins.2016.00408 by misfolded protein and prevented by chaperone function. EMBO J. 36, Jarosz, D. F., and Khurana, V. (2017). Specification of physiologic and disease 1669–1687. doi: 10.15252/embj.201695957 states by distinct proteins and protein conformations. Cell 171, 1001–1014. Meisl, G., Kirkegaard, J. B., Arosio, P., Michaels, T. C. T., Vendruscolo, M., doi: 10.1016/j.cell.2017.10.047 Dobson, C. M., et al. (2016). Molecular mechanisms of protein aggregation from Frontiers in Molecular Neuroscience | www.frontiersin.org 15 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE global fitting of kinetic models. Nat. Protoc. 11, 252–272. doi: 10.1038/nprot. Persson, E. K., Verstraete, K., Heyndrickx, I., Gevaert, E., Aegerter, H., Percier, 2016.010 J.-M., et al. (2019). Protein crystallization promotes type 2 immunity and is Michaels, T. C. T., Dear, A. J., Kirkegaard, J. B., Saar, K. L., reversible by antibody treatment. Science 364:eaaw4295. doi: 10.1126/science. Weitz, D. A., and Knowles, T. P. J. (2016). Fluctuations in aaw4295 the kinetics of linear protein self-assembly. Phys. Rev. Lett. Pieri, L., Madiona, K., and Melki, R. (2016). Structural and functional properties of 116:258103. doi: 10.1103/PhysRevLett.116.258103 prefibrillar α-synuclein oligomers. Sci. Rep. 6:24526. doi: 10.1038/srep24526 Micsonai, A., Wien, F., Kernya, L., Lee, Y. H., Goto, Y., Refregiers, M., et al. (2015). Prausnitz, J. M., Lichtenthaler, R. N., and De Azevedo, E. G. (1998). Molecular Accurate secondary structure prediction and fold recognition for circular Thermodynamics of Fluid-Phase Equilibria. London: Pearson Education. dichroism spectroscopy. Proc. Natl. Acad. Sci. U.S.A. 112, E3095–E3103. doi: Provencher, S. W., and Glockner, J. (1981). Estimation of globular protein 10.1073/pnas.1500851112 secondary structure from circular dichroism. Biochemistry 20, 33–37. doi: 10. Miller, S. B., Ho, C.-T., Winkler, J., Khokhrina, M., Neuner, A., Mohamed, M. Y., 1021/bi00504a006 et al. (2015). Compartment-specific aggregases direct distinct nuclear and Prusiner, S. B. (2001). Neurodegenerative Diseases and Prions. N. Engl. J. Med. 344, cytoplasmic aggregate deposition. EMBO J. 34, 778–797. doi: 10.15252/embj. 1516–1526. doi: 10.1056/NEJM200105173442006 201489524 Qiang, W., Yau, W.-M., Lu, J.-X., Collinge, J., and Tycko, R. Miti, T., Mulaj, M., Schmit, J. D., and Muschol, M. (2015). Stable, metastable, and (2017). Structural variation in amyloid-β fibrils from Alzheimer’s kinetically trapped amyloid aggregate phases. Biomacromolecules 16, 326–335. disease clinical subtypes. Nature 541, 217–221. doi: 10.1038/nature doi: 10.1021/bm501521r 20814 Mok, Y.-F., and Howlett, G. J. (2006). Sedimentation velocity analysis of amyloid Rasmussen, J., Mahler, J., Beschorner, N., Kaeser, S. A., Häsler, L. M., Baumann, oligomers and fibrils. Methods Enzymol. 413, 199–217. doi: 10.1016/s0076- F., et al. (2017). Amyloid polymorphisms constitute distinct clouds of 6879(06)13011-6 conformational variants in different etiological subtypes of Alzheimer’s disease. Mok, Y.-F., Howlett, G. J., and Griffin, M. D. W. (2015). Sedimentation Proc. Natl. Acad. Sci. U.S.A. 114, 13018–13023. doi: 10.1073/pnas.17132 velocity analysis of the size distribution of amyloid oligomers and 15114 fibrils. Methods Enzymol. 562, 241–256. doi: 10.1016/bs.mie.2015. Rivas, G., and Minton, A. P. (2016). Macromolecular crowding in vitro, in vivo, and 06.024 in between. Trends Biochem. Sci. 41, 970–981. doi: 10.1016/j.tibs.2016.08.013 Morris, K. L., and Serpell, L. C. (2012). X-ray fibre diffraction studies of amyloid Rochet, J.-C., and Lansbury, P. T. (2000). Amyloid fibrillogenesis: themes and fibrils. Methods Mol. Biol. 849, 121–135. doi: 10.1007/978-1-61779-551-0_9 variations. Curr. Opin. Struct. Biol. 10, 60–68. doi: 10.1016/s0959-440x(99) Murakami, T., Qamar, S., Lin, J Q., Schierle, G. S. K., Rees, E., Miyashita, 00049-4 A., et al. (2015). ALS/FTD mutation-induced phase transition of FUS Ruff, K. M., Pappu, R. V., and Holehouse, A. S. (2019). Conformational preferences liquid droplets and reversible hydrogels into irreversible hydrogels impairs and phase behavior of intrinsically disordered low complexity sequences: RNP granule function. Neuron 88, 678–690. doi: 10.1016/j.neuron.2015. insights from multiscale simulations. Curr. Opin. Struct. Biol. 56, 1–10. doi: 10.030 10.1016/j.sbi.2018.10.003 Murray, D. T., Kato, M., Lin, Y., Thurber, K. R., Hung, I., Mcknight, S. L., et al. Ruggeri, F. S., Longo, G., Faggiano, S., Lipiec, E., Pastore, A., and Dietler, (2017). Structure of FUS protein fibrils and its relevance to self-assembly and G. (2015). Infrared nanospectroscopy characterization of oligomeric phase separation of low-complexity domains. Cell 171, 615–627.e16. doi: 10. and fibrillar aggregates during amyloid formation. Nat. Commun. 1016/j.cell.2017.08.048 6:7831. doi: 10.1038/ncomms8831 Navarro, S., and Ventura, S. (2014). Fluorescent dye ProteoStat to detect Ruggeri, F. S., Mannini, B., Schmid, R., Vendruscolo, M., and Knowles, T. P. J. and discriminate intracellular amyloid-like aggregates in Escherichia coli. (2020). Single molecule secondary structure determination of proteins through Biotechnol. J. 9, 1259–1266. doi: 10.1002/biot.201400291 infrared absorption nanospectroscopy. Nat. Commun. 11:2945. doi: 10.1038/ O’Nuallain, B., Williams, A. D., Westermark, P., and Wetzel, R. (2004). Seeding s41467-020-16728-1 specificity in amyloid growth induced by heterologous fibrils. J. Biol. Chem. 279, Ruggeri, F. S., Sneideris, T., Vendruscolo, M., and Knowles, T. P. J. (2019). 17490–17499. doi: 10.1074/jbc.m311300200 Atomic force microscopy for single molecule characterisation of protein Orchard, S., Salwinski, L., Kerrien, S., Montecchi-Palazzi, L., Oesterheld, M., aggregation. Arch. Biochem. Biophys. 664, 134–148. doi: 10.1016/j.abb.2019. Stümpflen, V., et al. (2007). The minimum information required for 02.001 reporting a molecular interaction experiment (MIMIx). Nat. Biotechnol. 25, Ruggeri, F. S., Vieweg, S., Cendrowska, U., Longo, G., Chiki, A., Lashuel, H. A., et al. 894–898. doi: 10.1038/nbt1324 (2016). Nanoscale studies link amyloid maturity with polyglutamine diseases Ormsby, A. R., Ramdzan, Y. M., Mok, Y.-F., Jovanoski, K. D., and Hatters, D. M. onset. Sci. Rep. 6:31155. doi: 10.1038/srep31155 (2013). A platform to view huntingtin Exon 1 aggregation flux in the cell reveals Sabate, R., Rodriguez-Santiago, L., Sodupe, M., Saupe, S. J., and Ventura, S. (2013). divergent influences from chaperones hsp40 and hsp70. J. Biol. Chem. 288, Thioflavin-T excimer formation upon interaction with amyloid fibers. Chem. 37192–37203. doi: 10.1074/jbc.m113.486944 Commun. 49, 5745–5747. doi: 10.1039/c3cc42040j Orru, C. D., Wilham, J. M., Vascellari, S., Hughson, A. G., and Caughey, B. (2012). Sabate, R., and Saupe, S. J. (2007). Thioflavin T fluorescence anisotropy: an New generation QuIC assays for prion seeding activity. Prion 6, 147–152. doi: alternative technique for the study of amyloid aggregation. Biochem. Biophys. 10.4161/pri.19430 Res. Commun. 360, 135–138. doi: 10.1016/j.bbrc.2007.06.063 Oskarsson, M. E., Hermansson, E., Wang, Y., Welsh, N., Presto, J., Johansson, Saborio, G. P., Permanne, B., and Soto, C. (2001). Sensitive detection of J., et al. (2018). BRICHOS domain of Bri2 inhibits islet amyloid polypeptide pathological prion protein by cyclic amplification of protein misfolding. Nature (IAPP) fibril formation and toxicity in human beta cells. Proc. Natl. Acad. Sci. 411, 810–813. doi: 10.1038/35081095 U.S.A. 115, E2752–E2761. doi: 10.1073/pnas.1715951115 Saelices, L., Chung, K., Lee, J. H., Cohn, W., Whitelegge, J. P., Benson, Otzen, D., and Riek, R. (2019). Functional amyloids. Cold Spring Harb. Perspect. M. D., et al. (2018). Amyloid seeding of transthyretin by ex vivo Biol. 11:a033860. doi: 10.1101/cshperspect.a033860 cardiac fibrils and its inhibition. Proc. Natl. Acad. Sci. U.S.A. 115, Padrick, S. B., and Miranker, A. D. (2002). Islet amyloid: phase partitioning E6741–E6750. doi: 10.1073/pnas.1805131115 and secondary nucleation are central to the mechanism of fibrillogenesis. Sahin, E., and Roberts, C. J. (2012). Size-exclusion chromatography with multi- Biochemistry 41, 4694–4703. doi: 10.1021/bi0160462 angle light scattering for elucidating protein aggregation mechanisms. Methods Pallarès, I., and Ventura, S. (2019). Advances in the prediction of protein Mol. Biol. 899, 403–423. doi: 10.1007/978-1-61779-921-1_25 aggregation propensity. Curr. Med. Chem. 26, 3911–3920. doi: 10.2174/ Saijo, E., Ghetti, B., Zanusso, G., Oblak, A., Furman, J. L., Diamond, M. I., et al. 0929867324666170705121754 (2017). Ultrasensitive and selective detection of 3-repeat tau seeding activity Patel, A., Lee, H. O., Jawerth, L., Maharana, S., Jahnel, M., Hein, M. Y., in Pick disease brain and cerebrospinal fluid. Acta Neuropathol. 133, 751–765. et al. (2015). A liquid-to-solid phase transition of the ALS protein FUS doi: 10.1007/s00401-017-1692-z accelerated by disease mutation. Cell 162, 1066–1077. doi: 10.1016/j.cell.2015. Salvadores, N., Shahnawaz, M., Scarpini, E., Tagliavini, F., and Soto, C. (2014). 07.047 Detection of misfolded Aβ oligomers for sensitive biochemical diagnosis Frontiers in Molecular Neuroscience | www.frontiersin.org 16 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE of Alzheimer’s disease. Cell Rep. 7, 261–268. doi: 10.1016/j.celrep.2014. at molecular and cellular levels. Nucleic Acids Res. 43:e92. doi: 10.1093/nar/ 02.031 gkv338 Sant’Anna, R., Fernandez, M. R., Batlle, C., Navarro, S., De Groot, N. S., Serpell, Surin, A. K., Grishin, S. Y., and Galzitskaya, O. V. (2020). L., et al. (2016). Characterization of amyloid cores in prion domains. Sci. Rep. Determination of amyloid core regions of insulin analogues 6:34274. doi: 10.1038/srep34274 fibrils. Prion 14, 149–162. doi: 10.1080/19336896.2020.17 Sárkány, Z., Rocha, F., Damas, A. M., Macedo-Ribeiro, S., and Martins, P. M. 76062 (2019). Chemical kinetic strategies for high-throughput screening of protein Tanford, C. (1970). “Protein Denaturation: Part C.∗∗ Parts A and B were published aggregation modulators. Chem. Asian J. 14, 500–508. doi: 10.1002/asia. in Volume 23 of Advances in Protein Chemistry (1968), starting on p. 121 201801703 Theoretical Models for The Mechanism of Denaturation,” in Advances in Schönherr, R., Rudolph Janine, M., and Redecke, L. (2018). Protein crystallization Protein Chemistry, eds C. B. Anfinsen, J. T. Edsall, and F. M. Richards in living cells. Biol. Chem. 399, 751–772. doi: 10.1515/hsz-2018-0158 (Cambridge, MA: Academic Press), 1–95. doi: 10.1016/s0065-3233(08) Selivanova, O. M., Surin, A. K., Marchenkov, V. V., Dzhus, U. F., 60241-7 Grigorashvili, E. I., Suvorina, M. Y., et al. (2016). The mechanism Teplow, D. B. (1998). Structural and kinetic features of amyloid β- underlying amyloid polymorphism is opened for Alzheimer’s disease protein fibrillogenesis. Amyloid 5, 121–142. doi: 10.3109/13506129808 amyloid-β peptide. J. Alzheimers Dis. 54, 821–830. doi: 10.3233/jad- 995290 160405 Trewhella, J., Duff, A. P., Durand, D., Gabel, F., Guss, J. M., Hendrickson, W. A., Shankar, G. M., Li, S., Mehta, T. H., Garcia-Munoz, A., Shepardson, N. E., Smith, et al. (2017). 2017 publication guidelines for structural modelling of small-angle I., et al. (2008). Amyloid-β protein dimers isolated directly from Alzheimer’s scattering data from biomolecules in solution: an update. Acta Crystallogr. Sect. brains impair synaptic plasticity and memory. Nat. Med. 14, 837–842. doi: D Biol. Crystallogr. 73, 710–728. doi: 10.1107/S2059798317011597 10.1038/nm1782 Vadukul, D. M., Al-Hilaly, Y. K., and Serpell, L. C. (2019). Methods for structural Shen, D., Coleman, J., Chan, E., Nicholson, T. P., Dai, L., Sheppard, analysis of amyloid fibrils in misfolding diseases. Methods Mol. Biol. 1873, P. W., et al. (2011). Novel cell- and tissue-based assays for detecting 109–122. doi: 10.1007/978-1-4939-8820-4_7 misfolded and aggregated protein accumulation within aggresomes and Vekilov, P. G. (2010). Phase transitions of folded proteins. Soft Matter 6, 5254– inclusion bodies. Cell Biochem. Biophys. 60, 173–185. doi: 10.1007/s12013-01 5272. doi: 10.1039/c0sm00215a 0-9138-4 Ventura, S., Zurdo, J., Narayanan, S., Parreno, M., Mangues, R., Reif, Shin, Y., and Brangwynne, C. P. (2017). Liquid phase condensation in B., et al. (2004). Short amino acid stretches can mediate amyloid cell physiology and disease. Science 357:eaaf4382. doi: 10.1126/science. formation in globular proteins: the Src homology 3 (SH3) case. aaf4382 Proc. Natl. Acad. Sci. U.S.A. 101, 7258–7263. doi: 10.1073/pnas.0308 Shivu, B., Seshadri, S., Li, J., Oberg, K. A., Uversky, V. N., and Fink, 249101 A. L. (2013). Distinct beta-sheet structure in protein aggregates determined Vestergaard, B., Groenning, M., Roessle, M., Kastrup, J. S., De Weert, M. V., by ATR-FTIR spectroscopy. Biochemistry 52, 5176–5183. doi: 10.1021/bi Flink, J. M., et al. (2007). A helical structural nucleus is the primary elongating 400625v unit of insulin amyloid fibrils. PLoS Biol. 5:e134. doi: 10.1371/journal.pbio.00 Silva, A., Almeida, B., Fraga, J. S., Taboada, P., Martins, P. M., and Macedo-Ribeiro, 50134 S. (2017). Distribution of amyloid-like and oligomeric species from protein Vey, M., Pilkuhn, S., Wille, H., Nixon, R., Dearmond, S. J., Smart, E. J., et al. (1996). aggregation kinetics. Angew. Chem. Int. Ed. 56, 14042–14045. doi: 10.1002/anie. Subcellular colocalization of the cellular and scrapie prion proteins in caveolae- 201707345 like membranous domains. Proc. Natl. Acad. Sci. U.S.A. 93, 14945–14949. doi: Silva, A., Sárkány, Z., Fraga, J. S., Taboada, P., Macedo-Ribeiro, S., and 10.1073/pnas.93.25.14945 Martins, P. M. (2018). Probing the occurrence of soluble oligomers through Viegas, M. S., Martins, T. C., Seco, F., and Do Carmo, A. (2007). An improved amyloid aggregation scaling laws. Biomolecules 8:108. doi: 10.3390/biom and cost-effective methodology for the reduction of autofluorescence in direct 8040108 immunofluorescence studies on formalin-fixed paraffin-embedded tissues. Eur. Sjolander, D., Rocken, C., Westermark, P., Westermark, G. T., Nilsson, J. Histochem. 51, 59–66. doi: 10.4081/1013 K. P., and Hammarstrom, P. (2016). Establishing the fluorescent Vorontsova, M. A., Maes, D., and Vekilov, P. G. (2015). Recent advances in the amyloid ligand h-FTAA for studying human tissues with systemic and understanding of two-step nucleation of protein crystals. Faraday Discuss. 179, localized amyloid. Amyloid 23, 98–108. doi: 10.3109/13506129.2016.11 27–40. doi: 10.1039/c4fd00217b 58159 Waeytens, J., Van Hemelryck, V., Deniset-Besseau, A., Ruysschaert, J.-M., Dazzi, Skrabana, R., Kovacech, B., Filipcik, P., Zilka, N., Jadhav, S., Smolek, T., et al. (2017). A., and Raussens, V. (2020). Characterization by nano-infrared spectroscopy Neuronal expression of truncated Tau efficiently promotes neurodegeneration of individual aggregated species of amyloid proteins. Molecules 25:2899. doi: in animal models: pitfalls of toxic oligomer analysis. J. Alzheimers Dis. 58, 10.3390/molecules25122899 1017–1025. doi: 10.3233/jad-161124 Walters, R. H., and Murphy, R. M. (2011). aggregation kinetics of interrupted Sleutel, M., and Van Driessche, A. E. (2018). Nucleation of protein crystals– polyglutamine peptides. J. Mol. Biol. 412, 505–519. doi: 10.1016/j.jmb.2011.07. a nanoscopic perspective. Nanoscale 10, 12256–12267. doi: 10.1039/c8nr 003 02867b Wang, W., Chen, R., Luo, K., Wu, D., Huang, L., Huang, T., et al. (2010). Sohl, J. L., Jaswal, S. S., and Agard, D. A. (1998). Unfolded conformations of Calnexin inhibits thermal aggregation and neurotoxicity of prion protein. α-lytic protease are more stable than its native state. Nature 395, 817–819. J. Cell. Biochem. 111, 343–349. doi: 10.1002/jcb.22698 doi: 10.1038/27470 Weinbuch, D., Hawe, A., Jiskoot, W., and Friess, W. (2018). “Introduction Soto, C., and Pritzkow, S. (2018). Protein misfolding, aggregation, and into formulation development of biologics,” in Challenges in Protein Product conformational strains in neurodegenerative diseases. Nat. Neurosci. 21, 1332– Development, eds N. W. Warne and H.-C. Mahler (Cham: Springer), 3–22. 1340. doi: 10.1038/s41593-018-0235-9 doi: 10.1007/978-3-319-90603-4_1 Sreerama, N., and Woody, R. W. (2000). Estimation of protein secondary structure Westermark, P., Benson, M. D., Buxbaum, J. N., Cohen, A. S., Frangione, B., Ikeda, from circular dichroism spectra: comparison of CONTIN, SELCON, and S.-I., et al. (2007). A primer of amyloid nomenclature. Amyloid 14, 179–183. CDSSTR methods with an expanded reference set. Anal. Biochem. 287, 252–260. doi: 10.1080/13506120701460923 doi: 10.1006/abio.2000.4880 Whitmore, L., and Wallace, B. A. (2004). DICHROWEB, an online server Striegel, A. M., Yau, W. W., Kirkland, J. J., and Bly, D. D. (2009). Modern for protein secondary structure analyses from circular dichroism Size-Exclusion Liquid Chromatography: Practice of Gel Permeation and Gel spectroscopic data. Nucleic Acids Res. 32, W668–W673. doi: 10.1093/nar/ Filtration Chromatography. Hoboken, NJ: John Wiley & Sons, Inc. doi: 10.1002/ gkh371 9780470442876 Yakupova, E. I., Bobyleva, L. G., Vikhlyantsev, I. M., and Bobylev, A. G. Sugimoto, S., Arita-Morioka, K., Mizunoe, Y., Yamanaka, K., and Ogura, T. (2019). Congo Red and amyloids: history and relationship. Biosci. Rep. (2015). Thioflavin T as a fluorescence probe for monitoring RNA metabolism 39:BSR20181415. doi: 10.1042/BSR20181415 Frontiers in Molecular Neuroscience | www.frontiersin.org 17 November 2020 | Volume 13 | Article 582488 Martins et al. MIRRAGGE Yang, S., Griffin, M. D. W., Binger, K. J., Schuck, P., and Howlett, G. J. (2012). An Zheng, W., Schafer, N. P., and Wolynes, P. G. (2013). Frustration in the energy equilibrium model for linear and closed-loop amyloid fibril formation. J. Mol. landscapes of multidomain protein misfolding. Proc. Natl. Acad. Sci. U.S.A. 110, Biol. 421, 364–377. doi: 10.1016/j.jmb.2012.02.026 1680–1685. doi: 10.1073/pnas.1222130110 Yang, T., Li, S., Xu, H., Walsh, D. M., and Selkoe, D. J. (2017). Large soluble oligomers of amyloid β-protein from Alzheimer brain are far less neuroactive Conflict of Interest: The authors declare that the research was conducted in the than the smaller oligomers to which they dissociate. J. Neurosci. 37, 152–163. absence of any commercial or financial relationships that could be construed as a doi: 10.1523/jneurosci.1698-16.2017 potential conflict of interest. Yoshimura, Y., Lin, Y., Yagi, H., Lee, Y.-H., Kitayama, H., Sakurai, K., et al. (2012). Distinguishing crystal-like amyloid fibrils and glass-like amorphous aggregates Copyright © 2020 Martins, Navarro, Silva, Pinto, Sárkány, Figueiredo, Pereira, from their kinetics of formation. Proc. Natl. Acad. Sci. U.S.A. 109, 14446–14451. Pinheiro, Bednarikova, Burdukiewicz, Galzitskaya, Gazova, Gomes, Pastore, Serpell, doi: 10.1073/pnas.1208228109 Skrabana, Smirnovas, Ziaunys, Otzen, Ventura and Macedo-Ribeiro. This is an Zaccai, N. R., Serdyuk, I. N., and Zaccai, J. (2017). Methods in Molecular open-access article distributed under the terms of the Creative Commons Attribution Biophysics: Structure, Dynamics, Function for Biology and Medicine. Cambridge: License (CC BY). The use, distribution or reproduction in other forums is permitted, Cambridge University Press. doi: 10.1017/9781107297227 provided the original author(s) and the copyright owner(s) are credited and that the Zhao, R., So, M., Maat, H., Ray, N. J., Arisaka, F., Goto, Y., et al. (2016). original publication in this journal is cited, in accordance with accepted academic Measurement of amyloid formation by turbidity assay-seeing through the practice. No use, distribution or reproduction is permitted which does not comply cloud. Biophys. Rev. 8, 445–471. doi: 10.1007/s12551-016-0233-7 with these terms. Frontiers in Molecular Neuroscience | www.frontiersin.org 18 November 2020 | Volume 13 | Article 582488