Impact of nanoparticles on amyloid peptide and protein aggregation: a review with a focus on gold nanoparticles

Torsten John† ^abc, Anika Gladytz† ^ab, Clemens Kubeil‡ ^c, Lisandra L. Martin ^c, Herre Jelger Risselada ^ad and Bernd Abel *^ab
aLeibniz Institute of Surface Engineering (IOM), Permoserstraße 15, 04318 Leipzig, Germany. E-mail: bernd.abel@iom-leipzig.de
^bWilhelm-Ostwald-Institute for Physical and Theoretical Chemistry, Leipzig University, Linnéstraße 3, 04103 Leipzig, Germany
^cSchool of Chemistry, Monash University, Clayton, Victoria 3800, Australia
^dInstitute for Theoretical Physics, Georg-August-Universität Göttingen, Friedrich-Hund-Platz 1, 37077 Göttingen, Germany

First published on 3rd September 2018

---

---

Torsten John

Torsten John graduated with a M.Sc. degree in Chemistry from Leipzig University (Germany) in 2015 and he is now a Ph.D. Candidate in Biophysical and Computational Chemistry. He performs his research jointly at Leipzig University (Germany), the Leibniz Institute of Surface Engineering (IOM) (Germany) and Monash University (Australia). His work is focused on understanding the aggregation behaviour of amyloidogenic peptides near interfaces. As such interfaces, biological membranes and nanoparticles are studied using both experimental and computational approaches. Torsten John is a recipient of a Ph.D. fellowship from the Friedrich-Ebert-Foundation and an Endeavour Research Fellowship from the Australian Government.

Anika Gladytz

Anika Gladytz holds a Joint M.Sc. degree in Advanced Spectroscopy in Chemistry from the University of Bergen (Norway), Université Lille 1 (France) and Leipzig University (Germany). She received a Ph.D. in Chemistry from Leipzig University for her research on “Peptides near interfaces” in 2016. Her main systems, namely amyloid peptides near (gold) nanoparticles and mineral-binding peptides on mica, were investigated experimentally and using molecular dynamics simulations. Anika Gladytz qualified as an environmental manager at the FernUniversität in Hagen (Germany) and is now responsible for the enforcement of national and European regulations on chemicals at the federal state authority in Brandenburg.

Clemens Kubeil

Clemens Kubeil was awarded his doctorate degree in Chemistry from the University of Technology Dresden (Germany) in 2016. His research is focused on transport phenomena in nanoscaled electrochemical systems. Understanding the fundamental mechanisms in small systems and exploiting their unique properties for analytical applications and in energy conversion and storage systems are essential to his work. He currently performs his research at the Helmholtz-Zentrum Dresden-Rossendorf HZDR (Germany). Before, he worked in the groups of Andreas Bund at TU Ilmenau (Germany), Henry S. White at the University of Utah (USA) and Lisa Martin at Monash University (Australia).

Lisandra L. Martin

Lisa Martin was awarded her Ph.D. in Chemistry from The Australian National University (ANU). After several post-doctoral years in Germany and the USA, as a recipient of an Alexander von Humboldt and Fulbright Fellowships, she was appointed at Flinders University (Australia). She moved to Monash University (Australia) in 2003 where she built her research interests in bioanalytical and medicinal chemistry. Lisa Martin's current research includes bioactive molecules that interact with membrane surfaces, the mechanisms of steroid hormone biosynthesis by membrane anchored cytochrome P450 enzymes, the membrane action of antimicrobial and amyloidogenic peptides and the incorporation of biomolecules into novel materials.

Herre Jelger Risselada

Herre Jelger Risselada works as an independent Research Associate at both the University of Leiden (The Netherlands) and the Georg-August University of Goettingen (Germany). He is a Guest Scientist at the Leibniz Institute of Surface Engineering (IOM) (Leipzig, Germany). Risselada combines a theoretical background in polymer sciences (M.Sc., Groningen, The Netherlands) with biophysical chemistry (Ph.D., Groningen) to study the behaviour and organisation of complex systems via computer simulations and theory. In particular, he focuses on the interface and relationships between bio- and material sciences. His work was recently awarded with the NWO Vidi grant (The Netherlands, 2017).

Bernd Abel

Bernd Abel received his Ph.D. in Chemistry from the University of Goettingen (Germany). After a postdoctoral stay at the MIT in Boston (USA), he became Associate Professor at the University of Goettingen and the MPI for Biophysical Chemistry (Germany). In 2008, Bernd Abel received a full professor position (W3) at the Wilhelm-Ostwald-Institute for Physical and Theoretical Chemistry at Leipzig University (Germany), and since 2012, he is Head of the Department Functional Surfaces and Deputy Director of the Leibniz Institute of Surface Engineering (IOM). The Abel groups study molecular physical and analytical chemistry, smart materials, ultrafast spectroscopy, and nanoscale imaging near surfaces.

---

1. Introduction

The encounter of non-biological interfaces, such as inorganic solid surfaces or nanoparticles, with biological media typically leads to conformational changes and ultimately denaturation of the biological species.^1–4 Cells and biomolecules can adsorb onto the inorganic solid usually with a loss of their native structure that can often result in a ‘foreign body’ reaction in the organism.⁵ Surfaces covered with proteins, lipids, other biomolecules or cells result in a biocoating.⁶ In the case of nanoparticles, the formed layer is commonly termed ‘corona’ (as shown in Fig. 1).⁷ This ‘corona’ is dynamic in the sense that depending on the composition of the surrounding medium, the kinetics of the adsorption and desorption events can determine the physical, chemical and even biochemical properties of the covered surface.^8,9

The biomolecule binding to a surface is due to an equilibrium between the gain in adsorption energy and the loss in entropy.¹ Physisorption is the main mechanism that drives the coating of inorganic surfaces by biomolecules in biological media.^10–12 It relies on the present attractive forces between the protein or peptide and the surface. The primary driving forces that lead to physisorption are electrostatic (Coulomb) interactions between opposite charges, hydrogen bond formation and van-der-Waals interactions.^12–15 Some amino acid side chains (e.g. thiols) can also be chemisorbed to surfaces leading to an essentially covalent binding of the peptide.^14,16 Tuning of the peptide properties by changing its sequence can consequently lead to very high affinities towards a specific surface. Such highly optimised peptides are, for example, used to coat the surface of implants in order to enhance their biocompatibility.^17–19 Highly-biocompatible surface coatings prevent the denaturation of native protein structures.²⁰ Hence, the design of inorganic surfaces or nanoparticles aims to control the effects on the biological medium.²¹

The exposure of nanoparticles and nanomaterials to organisms has increased over the last decade. For instance, nanoparticles are part of vehicle exhaust fumes,^22,23 paints²⁴ and cosmetic products.^25–27 Furthermore, the emerging research fields of nanomedicine,^28–33 nanodiagnostics^34,35 and nanoengineering^24,36 expose the body to these materials. Every (nano)particle that enters a biological medium is instantaneously coated by a corona.^8,9 The corona is the functional surface that cells and biomolecules detect and interact with.⁷

In recent years, the influence of nanoparticles on the folding and possible aggregation of proteins and peptides has gained increasing interest.^{11,37,38–45,46,47} Especially, the role of surfaces in the misfolding of different amyloid proteins and peptides into β-sheet rich fibrils is widely discussed.⁴⁸ These fibrillar structures can form amyloid plaques that are associated with numerous diseases including Alzheimer's disease^49–53 (Aβ peptide, tau protein), spongiform encephalitis^54,55 (prion proteins), type 2 diabetes^15,56–58 (human islet amyloid polypeptide, hIAPP) and Parkinson's disease^59–61 (α-synuclein). Amyloid peptides are soluble in their native state and only aggregate under specific circumstances. Whether the aggregated fibrils or soluble intermediates are toxic in the development of the associated diseases is still investigated.^62–64

Depending on the nanoparticle type,^9,38,43,65 peptide sequence,^3,37 concentration ratio between nanoparticles and peptides^{3,37,38,43,66,67} and the physical and chemical conditions^68,69 (pH, temperature, shaking, ionic strength), the effect of the nanoparticle ranges from complete inhibition^{38,41,58,66,70–72} to an enhanced acceleration^{11,37,43,68,73,74} of the misfolding of the peptides into amyloid fibrils (Fig. 1). Although enormous efforts have been made to identify a general pattern, a comprehensive mechanism that explains and predicts the diverse effects of nanoparticles on fibril formation remains elusive.

To what extent a nanoparticle surface is covered by a biomolecular corona depends not only on the composition of the surrounding medium but also on the properties of the nanoparticle.^9,65 Very hydrophilic surfaces are largely resistant to protein adsorption, as there is only little gain in interfacial energy if a biomolecule attaches to the hydrophilic surface. The interfacial energy describes the energy between nanoparticle surface and water that is minimised by the binding of a peptide or protein to the nanoparticle surface under partial release of its solvation shell.⁷⁵ In fact, this effect is commonly exploited by covering surfaces with sterically demanding or polar polymers, such as polyethylene glycol (PEG).^76,77 Vácha et al. recently reported that the same surface can either accelerate or inhibit amyloid peptide aggregation depending on the intrinsic aggregation propensity of the peptide in bulk solution. This means that the aggregation process of peptides that have a high intrinsic aggregation propensity, may be inhibited or retarded, whereas the same surface may accelerate aggregation for a peptide with low intrinsic aggregation propensity.⁶⁹

In this article, we review recent observations on the influence of nanoparticles on amyloid fibril formation and discuss the biophysical models that are used to explain these observations. In order to understand these effects, we first introduce the known mechanisms of amyloid fibril formation in the absence of nanoparticles. This is followed by an overview of the structural role of nanoparticles in amyloid peptide aggregation with a focus on the molecular insights of the peptide corona formation. We incorporate unpublished recent preliminary results from our groups in this work. Using surface sensing methods (QCM-D, SPR) and large-scale molecular dynamics simulations (MD), the described models for amyloid peptide aggregation near gold nanoparticle surfaces are discussed and evaluated. These insights on the interplay between nanoparticles and amyloid peptides may pave the road towards the future design of functional nanomaterials, especially in nanomedicine.

2. Amyloid peptide aggregation

2.1 Structural characteristics

Amyloidosis has been known as a clinical disorder resulting from extracellular amyloid fibril deposits in vivo.^78,79 The term amyloid is further used by biophysicists to describe any peptide fibrils with characteristic cross-β-sheet structure, independent of the location, extracellular or intracellular, or disease association.^80–83 Amyloid fibrils are highly symmetric elongated protein aggregates that share a characteristic X-ray pattern for their quaternary structure. The general fibril structure consists of two neighbouring β-sheets, a common secondary structure motif in peptides and proteins. A β-sheet is formed from at least two β-strands, i.e. peptide chains in a specific conformation, which laterally interact via backbone hydrogen bonds. Two neighbouring β-sheets in a fibril are typically 8–11 Å apart and have an interstrand distance within one sheet of 4.7 Å.^80–84 Amyloid fibrils form larger collective networks, so-called amyloid plaques, through hydrophobic interaction and H-bonding.

Amyloid fibrils bind dyes like Congo red and Thioflavin T.^85–88 This allows the kinetics of amyloid fibril formation to be efficiently followed using fluorescence spectroscopy.^89,90 In these experiments, when a dissolved dye is excited, it typically relaxes by internal transfer to the ground state. If the dye however intercalates into the fibril,⁹¹ bond rotation and thereby energy relaxation is hindered. This leads to an increased quantum yield of the photoemission, resulting in an enhanced fluorescence signal intensity and extended fluorescence lifetime.⁹² Fluorescence signals may also increase in the presence of (on or off-pathway) oligomers or due to an increasing light scattering background.⁹³ It is thus necessary to check the formation of fibrils using complementary methods, such as spectroscopic or microscopic techniques.^93–96

Originally, the mature insoluble amyloid fibrils were considered to be associated with amyloid diseases. However, recent studies suggest that transient soluble oligomers may be the toxic species.^64,97–101 Either way, the toxic species either involves an oligomeric state in the process of amyloid aggregation into mature fibrils or an off-pathway product, separate from the fibril formation.^{46,61,102,103} Insights into the molecular and pathological connections between protein aggregation and amyloid diseases were recently summarised by Ke et al.⁴⁶ Multiple strategies exist in the development of drugs against amyloid diseases, e.g. drugs that decrease or inhibit peptide aggregation or that stimulate the aggregate clearance.¹⁰⁴

It is highly important to understand the aggregation mechanism and to both identify and characterise the transient oligomers.^105–108 Moreover, functional amyloid fibrillar structures have recently received attention as novel materials for nanotechnology.¹⁰⁹ A detailed understanding of the aggregation mechanism will support the tailored design of such materials.

2.2 Mechanisms of peptide aggregation

The overall kinetics of amyloid peptide aggregation follows, at least, two phases and starts with monomeric peptide or protein units.⁴⁶ The primary ‘lag’ phase involves the rare formation of nucleation seeds until a critical nucleus¹¹⁰ size is reached (nucleation, phase 1) to initiate a phase of exponential fibril growth (elongation, phase 2) (see Fig. 2).^{56,105,111,112} A sigmoidal shape is characteristic for the aggregation kinetics of amyloidogenic peptides into insoluble fibrils (see Fig. 2).¹¹¹

This shape is typical for nucleated aggregation processes and can in principle be described using the classical nucleation theory.^113–115

Several models have been described to identify the bottleneck or energy barrier for the formation of a critical nucleus.^{111,116–118} Classical nucleation theory describes the process in which a critical sized nucleus is formed as a new bulk phase out of the homogenous solution. An attractive interaction between the dissolved particles and the nucleus leads to an energy gain during the aggregation process. Accordingly, the formation of a bulk nucleus phase is energetically favourable. However, the occurrence of a phase boundary is non-ideal and results in surface tension of the nucleus. The competition between the favourable, attractive interaction within the bulk and the unfavourable surface tension determines whether a nucleus grows or dissolves. The attractive interaction in the bulk depends on the volume of the (spherical) nucleus and is therefore proportional to R³, where R is the radius of the nucleus. The surface energy is proportional to the surface area, and therefore grows with R². Initially, when the nucleus is still small, the surface-to-bulk ratio is large. Accordingly, the nucleus is unstable and the probability that it grows further is low. If, however, the nucleus reaches a size in which the attractive interaction in the bulk counterbalances the surface tension, a so called critical nucleus is formed. These geometric dimension arguments suggest that, in case of amyloid fibrils, the critical nucleus is not a one-dimensional sheet but likely involves growth of the fibril in multiple dimensions observed for crystals in X-ray experiments.^79,119 Growth beyond the critical nucleus is energetically favourable and therefore occurs more rapidly.^113–115

Finke and Watzky developed a model that mathematically describes a two-step process consisting of a slow nucleation (k₁) followed by rapid growth (k₂) (see Fig. 3A, nucleated polymerization (NP)).^117,120,121 Classical nucleation theory only considers the average size and not the inner composition of the nucleus. There are, however, two possibilities for the processes of nucleation and fibril growth. Firstly, peptide monomers could directly assemble into ordered β-sheet rich nuclei. In this case, classical nucleation theory could be applied, as only the size of the nucleus would determine its growth probability (see Fig. 3A, nucleated polymerization (NP)). Secondly, weak and unspecific interactions between the peptides could first lead to unstructured oligomers with low β-sheet content. In this scenario, the rate-determining step would be the subsequent internal structuring process, which eventually leads to ordered β-sheet rich nuclei (see Fig. 3B, nucleated conformational conversion (NCC)). In either scenario, the β-sheet rich nuclei initiate amyloid fibril growth.¹²²

Both models, nucleated polymerization (NP) and nucleated conformational conversion (NCC), have been discussed in the context of amyloid aggregation in the literature.^111,122,123 Lansbury first described a nucleated polymerization (NP) mechanism as analogous to the classical nucleation theory for amyloid beta and the scrapie-associated prion protein (see Fig. 3A).^123–125 He concluded that this simple model correctly describes the observed fibril formation lag time, the exponential growth during the elongation phase, the critical concentration which must be exceeded to observe fibril formation and the effects of preformed seeds on the aggregation kinetics. Nielsen et al. applied the model to the kinetics of insulin aggregation¹²⁶ and Lee et al. developed a more detailed kinetic description to explain the growth of insulin monomers into mature fibrils.¹²⁷

The nucleation phase during amyloid peptide aggregation can be catalysed or even avoided if the peptide solution is seeded with critical nuclei or mature fibrils that act as templates (i.e. templated assembly, TA, or a secondary nucleation pathway).^106,128,129 For some proteins, it was found that only one peptide monomer is necessary to act as a nucleus. This implies that the peptide monomer can change its conformation such that it then has a greater propensity to aggregate into the fibril state in a rate-limiting step.¹³⁰ This conformationally changed monomer can then convert other monomers which then bind to the fibril ends (monomer-directed conversion, MDC).^131,132

The nucleated conformational conversion (NCC) mechanism was first proposed by Serio et al. (see Fig. 3B).^116,122 This model consists of three steps: (i) loosely assembled, unstructured oligomers are formed in a concentration dependent manner (k_1a), (ii) that then restructure into β-sheet rich, well ordered oligomers (nuclei) (k_1b), (iii) that rapidly grow into mature fibrils (k₂). The unstructured oligomers associate into structured clusters while in equilibrium with both monomers and larger assemblies.¹²² The restructuring process of these oligomers (step ii, k_1b) is rate-limiting and therefore responsible for the observed lag phase. As this model requires the formation of unstructured oligomers, it is in agreement with the observation that a critical protein concentration must be exceeded in order to initiate fibril growth.¹²³ Once the critical concentration is reached, the kinetics changes little over a broad range of peptide concentrations. This observation is in agreement with a rate limiting rearrangement step within the assembled unstructured oligomers. The NCC model is supported by recent aggregation studies using the Amyloid beta (Aβ) peptide,¹³³ the yeast prion protein SUP35,^122,134 the β₂-microglobulin protein¹³⁵ and the barstar protein.¹³⁶ Thus, the NCC model might be shared among a large group of amyloidogenic proteins and peptides or perhaps be universal among all amyloidogenic systems.

Classical nucleation theory (nucleated polymerization, NP) assumes a rate-limiting assembly of monomers into ordered oligomers of a critical size, whereas the NCC mechanism describes the rearrangement of unstructured oligomers as rate-limiting step.^122,123 Thus, in order to discern which model best describes the fibril formation process, time-resolved structural investigations of the peptide oligomerisation are necessary. However, because oligomers are by no means stable intermediates but transient species with a broad size distribution, such time-dependent structural elucidation is challenging.

Gladytz et al. studied the time-resolved structure of oligomers during the aggregation of insulin.¹³⁷ Laser induced liquid beam ion desorption mass spectrometry (LILBID-MS) was applied to detect the time and concentration dependent loss of peptide monomers and the appearance of small oligomers. LILBID is an emerging soft ionization technique which is optimized for ionization from aqueous or alcoholic solutions and shows an excellent linearity between the analyte concentration in solution and the measured signal.^138,139 It was found that the lag time for insulin fibril formation is about 50 min at a given concentration and pH (10⁻⁴ mol L⁻¹ at pH 2). The measured lag time was only slightly concentration dependent. This strongly suggests that a unimolecular process should be rate-limiting during the aggregation process. This process could be the restructuring of a preformed oligomer, whereas the aggregation of the monomers to the unstructured oligomers should, in contrast, be strongly concentration dependent.

Dynamic light scattering (DLS) was used in order to determine the hydrodynamic radius of the resulting aggregates. DLS allows detection of lag times as any larger aggregates will scatter the light more than monomers. The first significant change of the hydrodynamic radius was observed after 45 min in good agreement with the lag time monitored using LILBID-MS. At times beyond 45 min, the DLS signals broadened and additional signals at higher hydrodynamic radii appeared. These changes were correlated with structural information from circular dichroism (CD) spectroscopy. While the oligomers at 45 min still showed a predominantly α-helical structure, a secondary transition between 50 and 60 min was consistent with the conversion into β-sheet rich oligomers. These oligomers continued to rapidly grow into mature fibrils.¹³⁷

These findings fit to the model of nucleated conformational conversion (NCC) characterised by two nucleation steps (see Fig. 3B). Destabilized monomers (pH 2) first assembled loosely to largely unstructured oligomers driven by hydrophobic effects (k_1a). Once the unstructured oligomers reached a critical size, the oligomers were stable enough to enable a slow, but irreversible isomerization step to β-sheet rich oligomers (k_1b) that readily grew to form elongated fibrils (k₂). Atomic force microscopy (AFM) as well as dynamic light scattering (DLS) experiments revealed a relatively narrow size distribution of the oligomeric species of around 50 nm in diameter. The NCC model explains why only oligomers around this narrow size region were observed: larger oligomers are sufficiently long-lived to undergo the irreversible conformational change and therefore quickly grow to mature fibrils, which were detected in atomic force and X-ray microscopy images. Smaller oligomers, however, are unstable and therefore have a low abundance; hence are undetectable. The combination of experimental techniques enabled the direct observation of the isomerization step of initially unstructured to β-sheet rich oligomers as rate-determining step of amyloid peptide aggregation.

3. Effects of nanoparticles on amyloid peptide aggregation

3.1 Nanoparticles in nature

With the understanding of peptide aggregation in free solution, the following section introduces the effects of surfaces, or more precisely nanoparticles, on the aggregation behaviour of amyloidogenic peptides. Nanoparticles, in the diameter range between 1 nm and 10 nm, are known for their size-related properties (quantum size effects)¹⁴⁰ and are intensely discussed in the literature from many perspectives.^{7,23,31,35,140–142}

Nanoparticles are ubiquitous in our environment and can be used for a variety of applications due to the varied composition with extensive properties, sizes and shapes.^65,140 There is a fundamental interest in the understanding of the interaction of amyloidogenic peptides and proteins with nanoparticles as humans are exposed to nanoparticles in everyday life.^7,143,144 Nanoparticles are for instance part of exhaust fumes,^22,23 paints²⁴ and cosmetic products.^25–27 Furthermore, nanomedicine^28–31 and nanodiagnostics^34,35 are emerging research fields in which nanoparticles are intentionally introduced into the human body.

The enormously large surface-to-mass-ratio makes nanoparticles attractive for studying surface effects on amyloidogenic peptides in general.¹⁴⁰ Surface effects are proposed to play a major role for the assembly of amyloidogenic peptides in vivo as they may explain why these peptides and proteins misfold in vivo at concentrations which are insufficient for peptide fibril formation in solution in vitro.^69,123 Surfaces provide an external constraint for amyloidogenic peptide aggregation and might thus act as seeds to catalyse the aggregation process.^11,37,145

3.2 Structural role of nanoparticles on amyloid peptide aggregation

Colloidal nanoparticles are unstable in solution and thus are usually stabilised with carboxylic acids, alcohols or polymers (interfacial layer). The stabilised nanoparticle surface is then covered with a biomolecule layer, the ‘corona’, upon contact with a biological medium.^6,7 This ‘corona’ determines the nanoparticle's surface properties.^8,9 The competition between the peptide's intrinsic propensity for aggregation and the nanoparticle–peptide attraction governs the behaviour of the amyloidogenic peptides in solution (see Fig. 4).⁶⁹

In recent years, numerous research groups have investigated the effects of various nanoparticles on amyloidogenic peptides and proteins.^{11,39,40,144,146} Depending on the specific nanoparticle material,^{37,70,147–151} their size,^9,40,65,152 the peptide or protein sequence^3,37,153 and the chemical and physical conditions^68,69,154 used, the observed effects included a strong acceleration^{11,37,43,68,73} to a complete inhibition^{38,41,58,66,70,71} of fibril formation. The various possibilities for amyloidogenic peptide and protein transformations and assemblies at nanoparticle surfaces are illustrated in Fig. 4. An overview of known effects of gold nanoparticles (AuNPs, gold colloids) on amyloid peptide aggregation is presented in Table 1. Studies with other nanoparticles (polymers, quantum dots) are summarised elsewhere (ESI† in Gladytz et al.¹¹)§.

---

A model which explains the observed, differentiated effects of nanoparticles on amyloidogenic peptide systems is the so called condensation-ordering mechanism (COM) (see Fig. 5).⁷³ The COM model assumes that the relatively weak and non-specific hydrophobic attraction between nanoparticle surfaces and peptides accelerates the formation of unstructured oligomers. The hydrophobic attraction leads to a local up-concentration, i.e. condensation, of peptides on the nanoparticle surface while still allowing mobility and reorganization of the peptide monomers. The condensation of peptides on the surface increases the probability for the formation of small, disordered oligomers and reduces the dissociation rate. The disordered oligomers can subsequently reorder into β-sheet rich oligomers while they grow in size. Clearly, the faster formation of critical nuclei will, in turn, facilitate the secondary ‘ordering’ step (nucleated conformational conversion model, NCC) leading to an overall reduction of the lag time of fibril formation.⁷³ The nanoparticles are ‘artificially’ creating environments of increased peptide concentration that can catalyse the formation of unstructured oligomers which subsequently reorder into β-sheet rich oligomers (critical nuclei).

Experiments by Campioni et al., however, showed that the observed acceleration of amyloid fibril formation is not only caused by this increase in local peptide concentration.¹⁵⁸ They studied the concentration-dependent fibril formation of α-synuclein in the presence and in the absence of an air–water-interface. In accord with the condensation-ordering mechanism (COM), an increase of the peptide concentration due to the presence of the interface showed accelerated fibril formation. However, in the absence of the interface, no fibril formation was observed, independent of the chosen peptide concentration. Accordingly, the local concentration increase on the surface may not be the only effect of the interface on fibril formation. Further, as a peptide corona is observed in most, if not all, experiments (see Table 1), a local increase in concentration does not necessarily imply acceleration of fibril growth. Campioni et al. repeated their experiments with small amounts of preformed nucleation seeds added to the peptide mixtures. The peptide aggregation was generally faster and presented more mature peptide fibrils in the presence of the interface than in the absence. The authors concluded that in the presence of the air–water-interface, elongation of the added nucleation seeds and de novo nucleation processes took place in parallel; whereas in the absence of the interface, only elongation of the added nucleation seeds was observed.

From these experiments, one can conclude that the observed accelerating and β-sheet-forming effects are not solely explained by the local up-concentration of the peptides on the surface. The surface additionally influenced the secondary ordering step which leads to an isomerization of the unstructured oligomers into β-sheet rich oligomers.¹⁵⁸ This can be partially explained by heterogeneous nucleation.¹⁵⁹ By approximating an oligomer as a sphere, its solvent exposed surface is always smaller once attached to a surface or an interface than in free solution (a simple geometric argument).¹⁵⁹ Hence, the surface energy is decreased upon attractive interaction with a surface or an interface. Accordingly, the critical size at which the surface energy is overcompensated by the bulk energy is smaller. Therefore, nucleation at the surface results in a more rapid population of stable oligomers which do not dissociate but have time to ‘order’⁷³ or to undergo ‘conformational conversion’.¹²² Nevertheless, open questions remain. First, these purely kinetic arguments do not give any explanation for the observed structural change to β-sheet rich oligomers. Even more importantly, they also do not explain why some particles accelerate fibril formation while others do not have an effect or can even delay or inhibit fibril formation (see Table 1), although peptide adsorption on their surface is observed.

Further properties such as the influence of the nanoparticle material or coating,^{37,70,147–150} attractive forces between peptide and nanoparticle^{15,69,73,160–163} or nanoparticle diameter^9,40,65 were investigated to develop general patterns for the role of nanoparticles on amyloid peptide aggregation. Vácha et al. studied the effects of nanoparticle surfaces with different peptide binding strengths (mainly determined by charge and hydrophobicity) on the kinetics of peptide aggregation.⁶⁹ They found that weakly attractive surfaces delay the aggregation of amyloidogenic peptides while strongly attractive surfaces lead to acceleration. In another study, hydrophobic interfaces were identified to promote the primary heterogeneous nucleation rate and thus triggered peptide aggregation.¹⁶⁴ In addition, the nanoparticle surface effects were identified to be determined by the intrinsic aggregation propensity of the peptide in bulk solution, in relation to the surface attraction. Amyloid aggregation of a peptide with a high intrinsic aggregation propensity was delayed by an attractive surface, whereas the fibril formation of a peptide with a low intrinsic aggregation propensity was accelerated.⁶⁹ Radic et al. used coarse-grained molecular dynamics simulations and found that moderate attraction leads to accelerated fibril growth.¹⁶⁰ This can be rationalized using an analogy to the above mentioned condensation-ordering mechanism.⁷³ If the attractive forces were further increased, fast immobilization of the peptides and delay or even inhibition of fibril formation was observed.¹⁶⁰ According to Shezad et al.,¹⁴⁹ the surface roughness and mobility of the peptides bound on the surface also influence peptide aggregation. Rough surfaces decelerate the surface-diffusion of peptides compared to a flat surface and thus can delay or even inhibit the amyloid aggregation process.¹⁴⁹

In recent years, nanoparticle surfaces were modified in such a way to inhibit amyloid fibril formation, hence aim to reduce amyloid toxicity.^{70,147,148,156} Dubey et al. modified gold and silver nanoparticles with tyrosine and tryptophan amino acids. These coated nanoparticles inhibited the spontaneous aggregation of insulin, suppressed nucleation seed-induced aggregation and even promoted the disassembly of insulin aggregates.⁷⁰ In a similar approach, Xiong et al. studied the inhibitory effect of different peptides immobilised on gold nanoparticles. The free peptides alone presented only weak inhibitory effects of amyloid aggregation, whereas the nanoparticle immobilised peptides strongly inhibited aggregation.¹⁴⁸ Sen et al. further found that D-glutamic acid immobilised on gold nanoparticles is more effective in inhibiting amyloid fibril formation than L-glutamic acid.¹⁵⁷ However, it has been shown that accelerated protein aggregation is beneficial in reducing the population of toxic oligomeric species.^165–167 Thus, the inhibition of amyloid peptide aggregation with tailored nanoparticles, as described above, might actually achieve the opposite effect, an increase in the amount of the toxic species.

Several studies use the term of ‘nanoparticles’ when they study surface effects on amyloid peptide aggregation. However, their definition for the size of nanoparticles ranges from a few nanometres¹⁵⁰ up to hundred nanometres.⁹ Hence, the particle size dependence has recently been subject to investigation.^{40,65,150,168} Gao et al. correlated the size of nanoparticles with their effect on Aβ aggregation. Large gold nanoparticles (36 nm, 18 nm) accelerated peptide aggregation whereas small gold nanoparticles (6 nm) or nanoclusters (1.9 nm) significantly delayed or completely inhibited the aggregation process.⁴⁰ Chan et al. also observed the effect that small quantum dots (2.8 nm NAC-QD, 2.2 nm MPA-QD) inhibit and large quantum dots (3.8 nm NAC-QD, 3.5 nm MPA-QD) promote peptide aggregation, whereas medium sized particles (3.1 nm NAC-QD and MPA-QD) showed no effect.¹⁵⁰ It was proposed that the larger quantum dots have a higher density of bound charged ligands that promote the aggregation process.¹⁵⁰ Kim et al. observed large amorphous peptide aggregates for very large nanoparticles (80 nm) and protofibrils for large nanoparticles (20 nm).⁶⁵

From a geometric perspective, largely sized nanoparticles are more similar to a planar substrate, while a medium-sized nanoparticle collects and orients peptides in the form of a nanoparticle–protein corona that might accelerate the nucleation step of peptide aggregation. Small sized nanoparticles, in contrast, may interfere with the self-assembly of peptides into amyloid fibrils^40,46 by binding to specific segments of the peptide and thus inhibiting amyloid peptide aggregation. Interestingly, similar nanoparticle size effects may also explain the role of other molecular entities, such as vesicles or micelles in organisms on amyloid peptide fibrillation.⁴⁰

The concentration ratio of nanoparticles to peptides^67,160 is also essential for the nucleation and growth mechanism of amyloidogenic peptides. It is reported that low concentrations of positively charged polystyrene beads can accelerate fibrillation of Aβ₄₀.⁶⁷ If the bead concentration is increased, acceleration changes to peptide aggregation delay and inhibition. The authors rationalize that at very high nanoparticle concentrations, all peptides are bound to the surface so that the concentration of dissolved monomers is too low for the elongation of the peptide nuclei.⁶⁷ Cabaleiro-Lago et al.⁶⁷ and Radic et al.¹⁶⁰ additionally pointed out that a small nanoparticle-to-peptide ratio leads to a reduction of all these effects, as the environment increasingly resembles the conditions present in the absence of nanoparticles.

Physicochemical parameters, such as the pH value, temperature or the presence of metal ions or chemicals are known to influence the peptide aggregation process, even in the absence of nanoparticles.^{154,169–171} Kim et al. employed gold nanoparticles for an accelerated monitoring of amyloid peptide aggregation and tested its method at varying temperatures, pH values and with 2,2,2-trifluoroethanol (TFE). High temperatures, acidic environments or 50% TFE solution strongly accelerated the peptide aggregation in the presence of gold nanoparticles; similar to the effects without nanoparticles.¹⁵⁴ Ghavami et al. studied peptide aggregation in the absence and presence of silica (hydrophilic) and polystyrene (hydrophobic) nanoparticles.¹⁷² At 37 °C, both nanoparticles accelerated the aggregation process, whereas, at 42 °C, the silica nanoparticles led to acceleration and the polystyrene nanoparticles to inhibition of the amyloid peptide aggregation process. These results indicate that the temperature influences the peptide–nanoparticle interactions.¹⁷² When studying the effects of nanoparticles on amyloid peptide aggregation, one also needs to consider the inherent colloidal stability of the nanoparticles of interest.^173,174 Nanoparticles are unstable at certain pH values or ionic strengths of electrolytes depending on their nature.¹⁷³

The models discussed above show the importance of surface properties and the resulting attractive forces between peptide and surface for the nanoparticle effect on amyloid aggregation. This also implies that a secondary ordering mechanism on the nanoparticle surface determines the overall velocity of fibril formation. In order to test this hypothesis, Gladytz et al. designed an experiment which separates adsorption, ordering and elongation.^11,37 To achieve this, nanoparticles were exposed to a peptide solution with a peptide concentration that is high enough to cover the nanoparticles with peptides and low enough to suppress the elongation phase (within the measured time frame). Immediately upon contact of the amyloidogenic peptides with the nanoparticles in a low-concentration of peptide solution, adsorption of the peptides became detectable by a redshift of the plasmon-resonance of the nanoparticles and a 4 to 7 nm thick corona halo around the nanoparticles was present in the scanning electron microscopy (SEM) images.³⁷

To determine the time for the conversion of the unstructured peptide assemblies into active nuclei, the peptide concentration in solution was increased, with varying incubation times in freshly prepared monomer solution (t₁, see Fig. 6). Fibrils that formed were detected using fluorescence and electron microscopy measurements (t₂, time between concentration increase and fibril detection). The results were compared with the normal lag time of highly concentrated peptide solutions and it was found that t₁ + t₂ always equalled the normal lag time. If the pre-incubation time (t₁) was equal to, or larger than, the normal lag time, fibrils formed immediately upon concentration increase (t₂ = 0).³⁷ This means that bound peptides had enough time to potentially convert into β-rich oligomers that could act as nuclei. Thus, one can conclude that rather than the oligomer assembly (coverage of the surface with peptides), the ordering of the peptides at the surface might be the actual rate limiting step.^11,37 The detailed mechanism of the restructuring process at the nanoparticle surface is dependent on nanoparticle size and surface properties.

3.3 Amyloid sensing by surface plasmon resonance

To further understand the peptide corona formation at the nanoparticle surface, our group studied the localised surface plasmon resonance (LSPR) of gold nanodisks (diameter ≈ 100 nm, height 10 nm) on peptide adsorption. To achieve this, we used an instrument that simultaneously measures LSPR and QCM-D (quartz crystal microbalance with dissipation monitoring) signals (see Fig. 7). LSPR spectroscopy is highly sensitive to detect changes in the dielectric constant (refractive index) of materials at the interface of the nanostructures.^175,176 QCM-D, in addition, senses nanogram mass binding to the sensor surface.^177–181 Both methods enable the detection of peptide mass loading onto nanostructured surfaces. Wet mass can be derived from QCM-D and dry mass from LSPR, which can help to understand the constitution and film thickness of the peptide corona.

The amyloid peptides GNNQQNY (sequence from Sup35 protein) and NNFGAIL (sequence from IAPP/amylin protein) were studied at concentrations that are high enough for the peptide corona formation but too low for fibril formation. The peptide adsorption resulted in a small but immediate redshift, indicating that the gold surface was immediately covered with peptide (GNNQQNY: 685.8 nm → 686.3 nm, shift: +0.5 nm; NNFGAIL: 681.5 nm → 682.4 nm, shift: +0.9 nm) (see phase II in Fig. 7a and b). This was confirmed by concomitant mass binding in the QCM-D signal (frequency decrease in Fig. 7d).

To study the corona restructuring after peptide adsorption, the sensors were equilibrated for 45 min. However, no significant change was observed in the SPR signal indicating that the dielectric properties of the bound peptide corona were not or only slightly changed by peptide restructuring into β-sheets, or that no peptide restructuring took place (see phase III in Fig. 7a and b). The peptide molecules formed a rigid layer on the gold surface as can be seen from the small change in dissipation in the QCM-D data (Fig. 7d). The final water rinse did not significantly change the dielectric properties (small additional binding, i.e. redshift, observed) suggesting a firm binding and structure formation at the gold surface (see phase IV in Fig. 7a and b). These results support an immediate amyloid peptide binding to the gold sensor surface.

The gold nanoparticle surface served as a nucleation seed and induced a local increase in the concentration of peptide monomers that could form unstructured oligomers (condensation-ordering mechanism, COM). The further structural rearrangement into β-sheet rich structures was not measured in our experiments. The LSPR redshift is a valuable tool to study the peptide loading on nanostructured surfaces. Both QCM-D and LSPR can help to understand the constitution and film thickness of the peptide corona. Further, kinetic information of the adsorption is accessible but it remains open whether structural changes within the peptide corona can be quantified. The method might be a useful tool to sense the nanoscale thickness of adsorbed peptide layers.¹⁵⁴

3.4 Molecular mechanism of oligomer restructuring at nanoparticle surfaces

Having a general model of the nanoparticle effects in mind, it is now possible to investigate the peptide adsorption motifs at a molecular level. In order to identify a mechanism for the restructuring processes at the nanoparticle surface, molecular dynamics (MD) simulations have been used in a number of studies.^{11,39,71,73,160,182,183} The development of a reliable force field that correctly describes the interaction between peptides and inorganic surfaces, such as gold, was a crucial prerequisite for these studies.^141,184,185 Corni et al. developed the polarizable GolP-CHARMM force field for gold surfaces that is based on the GoIP force field and compatible with the bio-organic force field CHARMM.^14,184,185 It was parametrized based on quantum calculations (DFT, MP2) and experimental data.¹⁴

In a first approach, Hoefling et al. studied the interaction of single amino acids with the Au(111) surface.¹⁸⁶ It was found that these gold surfaces induce β-sheet-like conformations due to favoured backbone interactions of the amino acids with the gold surface.¹⁸⁶ Amino acids with high intrinsic β-sheet propensity easily interacted with the gold surface while other conformations, such as helices, needed to be distorted.¹⁸⁶ Further, uncharged peptide domains presented a favoured binding to gold surfaces after a cationic arginine residue initiated the peptide-surface binding.¹⁸⁷ Peptide-surface binding occurred on a short time scale and surfaces were covered with a peptide layer that rearranged and determined the gold surface properties.¹⁸⁷ Bellucci and Corni investigated the free energy landscape of the alanine dipeptide interaction on a gold surface.¹⁸⁸ They found that elongated conformations of the peptide, such as fibrils, were favoured. The peptide conformations were modified by changes in the order and shape of local energy minima as well as changes in the paths and barriers between these minima.¹⁸⁸

Skepö studied the binding of a negatively charged protein, statherin, to an uncharged, a negatively and a positively charged surface by Monte Carlo simulations using a one bead model for each residue.¹⁸⁹ The affinity of the protein to adsorb to the surface and the charge distribution along the protein chain determined the protein conformation and its adlayer thickness on the surface.¹⁸⁹ Peptide–nanoparticle interactions of varying strengths were studied using coarse-grained MD simulations by Radic et al.¹⁶⁰ Weak peptide–nanoparticle attraction led to a locally increased peptide concentration and increased amyloid aggregation, whereas strong peptide–nanoparticle attraction led to unstable fibrils due to reduced lateral diffusion and thus an inhibition of conformational changes necessary for amyloid aggregation.¹⁶⁰ Bellucci et al. further described a peptide concentration dependent effect.⁷¹ Nanoparticle surfaces reduced the Aβ_16–22 fibril formation propensity while the peptides bound to the surface leading to an increased peptide concentration on the nanoparticle surface that can then catalyse peptide nuclei formation.⁷¹ In another study with Aβ₄₂, Bellucci et al. found that fibril-prone peptide conformations were favoured and thus amyloid aggregation promoted upon gold nanoparticle contact, independent of peptide concentration.¹⁸³

Lopez et al. developed a coarse-grained model to study the adsorption energies of globular proteins on nanoparticles as a function of nanoparticle size and hydrophobicity.¹⁹⁰ It was shown that the nanoparticle charge has only little effect on the protein adsorption energy or the orientation of protein adsorption on the surface. Rather, the nanoparticle charge affected the van der Waals interactions and the nanoparticle size presented significant influence on the adsorption energy as the nanoparticle curvature determines the available surfaces for peptide binding.¹⁹⁰

Brancolini et al. suggested that nanoparticle surfaces may stabilise or destabilise amyloidogenic peptides and proteins as a result of balancing electrostatic and hydrophobic interactions with the amino-terminal peptide region being the most perturbed.^15,39 Citrate-coated gold nanoparticles interfered with the formation of dimers and even disassociated previously formed dimers, thus inhibiting fibril formation.¹⁹¹ Gladytz et al. studied the mechanisms during the rate-limiting structural reorganisation of the peptide corona that was initially formed when in contact with the biomolecules.¹¹ Preformed small, loose oligomers (dimers, trimers) were unstable and dissociated upon gold surface binding, whereas larger oligomers, such as hexamers, were stable. Partial physisorption of peptides on the gold surface led to the formation of aligned peptide monolayers and (parallel) oligomers. Unstructured oligomers did not grow into fibrils but a structural rearrangement needed to take place. Entropic energy barriers for structural ordering into parallel β-sheets were overcome by an increase in peptide concentration (translational) and N-terminal peptide physisorption^11,191 to the nanoparticle surface (rotational). The lag time needed until a critical amyloid nucleus is formed was attributed to the rearrangements of unstructured oligomers into ordered β-sheets.

Gladytz et al. concluded that the peptide–surface interaction strength determines the effect on amyloid peptide aggregation. Too strong peptide–surface interaction would inhibit conformational restructuring on the nanoparticle surface and thus inhibit amyloid peptide aggregation. If the interaction is too weak, peptide–peptide interactions dominate and the nanoparticles have no significant effect on aggregation. In case of medium attraction, N- or C-terminal physisorption to the nanoparticle surface and subsequent structural rearrangements within the peptide oligomers occur that lead to the formation of critical peptide nuclei that can catalyse fibril formation.¹¹

Our group continued to investigate the influence of gold nanoparticles on peptide aggregation by studying the amyloid model peptides GNNQQNY (from Sup35 protein) NNFGAIL (from IAPP/amylin protein) and VQIVYK (from tau protein). Our MD simulations confirmed the structure forming properties of the citrate-gold surface (see Fig. 8). The GNNQQNY (Fig. 8a) and NNFGAIL (Fig. 8b) peptide monomers presented a favoured N-terminal adsorption to the citrate-gold surface. This is due to the electrostatic attraction of citrate-oxygen atoms and the positively charged N-terminus. In the case of VQIVYK (Fig. 8c), the peptide monomers presented no favoured peptide terminus for binding. Both the N-terminus and the C-terminal lysine side chain were adsorbed to the citrate-gold layer. For all peptides, the initial contact via a charged group to the gold surface led to a local up-concentration and alignment of peptide monomers at the surface. This alignment is the potential cause for the formation of a first layer of β-sheet rich oligomers. The peptides formed both parallel and antiparallel dimers within the simulation time (Fig. 8d and e) which lead to mature oligomers and fibrils.¹⁹²

The N-terminal peptide adsorption to the gold surfaces was confirmed by analysing the relative RMSF (root mean square fluctuation) values of the amino acid residues within each peptide (see Fig. 9). All peptides showed higher fluctuations at their termini compared to their centre which was expected. Further, the GNNQQNY (black) and NNFGAIL (red) peptide presented very low fluctuations for the N-terminus if bound to gold compared to its C-terminus. The VQIVYK peptide (blue) showed higher fluctuations for the N-terminus because it binds both via its N-terminus and C-terminus (lysine side chain) to the gold surface. Wang et al. recently demonstrated that the IAPP peptide initiates its gold surface binding via the N-terminal domain.¹⁹³ Following conformational changes, that are facet-dependent, the gold nanoparticles accelerated peptide aggregation; especially for the larger nanoparticles.

The peptide monomer surface binding did not affect the (internal) conformational accessibility of the peptides. All the allowed regions for structural arrangements within the peptides were sampled as apparent in the Ramachandran plots in Fig. 10. The gold surface had no effect on the structure of single peptides compared to a free monomer in solution but rather catalysed the alignment into oligomeric structures. However, the gold-surface bound peptide monomers displayed a highly occupied dihedral angle space at very low psi angles around 0° (see Ramachandran plots in Fig. 10a and b). Interestingly, a similar conformational flexibility has been observed at the loose ends of growing oligomers and fibrils.¹¹ Both of these observations suggest that the surface may not accelerate fibril formation by enhancing a transition within the monomer to another structural intermediate such as, for example, an alpha-helix, as has been proposed in other studies,¹⁹⁴ but rather by the alignment of monomers on the surface.

To confirm the observed aggregation mechanism on gold surfaces, our group studied the interaction of GNNQQNY with a mica surface, which is a sheet silicate (phyllosilicate) with a negative surface charge. The MD simulations did not present any up-concentration of peptide monomers at the mica surface (see Fig. 10d) and displayed only few conformations within the dihedral angle space that was identified to be characteristic for the growing areas of fibrils (Fig. 10c). We found no significant influence of surface contributions; thus confirming the structure forming properties of gold nanoparticles.

Using our MD simulations, we studied the peptide adsorption and conformational behaviour of several amyloidogenic peptide sequences. Such atomistic or united-atom simulations are limed in their time-scale and thus, our work can only describe the peptide behaviour within the first nanoseconds (200 ns). Amyloid peptide fibril formation experimentally presents lag-times of minutes to hours.³⁷ As a consequence, such MD simulations illustrate the initial behaviour between monomers and gold surfaces. However, we cannot make final conclusions about the rate-limiting rearrangement into critical β-sheet rich nuclei that initiate the fibril formation. The simulations confirm a local up-concentration (condensation-ordering mechanism, COM) of peptide monomers which are subsequently arranged on the gold surface to form dimers. This rearrangement might be the cause for the accelerating effect of the nanoparticles.

4. Conclusions

The research presented in this article illustrates the various effects that nanoparticles have on the amyloid peptide aggregation process. To our current knowledge, nanoparticles’ impact on the development of amyloid-related neurodegenerative diseases cannot be generalized and thus nanoparticles are neither a ‘blessing’ nor a ‘curse’. Depending on a number of parameters, nanoparticles accelerate, retard or inhibit amyloid peptide aggregation. In some cases, the peptides or proteins are even unperturbed by the presence of nanoparticles. In addition to environmental factors such as pH value, temperature and intrinsic peptide aggregation propensity, the nanoparticle size matters. This raises the issue of a clear size definition when studying the nanoparticle influence on peptide aggregation. Further, the charge of the particles and their surface coating largely influences the surface properties. If one wants to understand the physiological relevance of nanoparticles in organisms, bare gold nanoparticles are virtually absent in organisms as soon as they enter a living system. Upon contact with biological medium, surfaces are covered with a biofilm that consists of the main components of the body fluids. This biofilm, also called ‘corona’ for nanoparticles, then determines the nanoparticles’ physiological impact. Thus, to understand the in vivo effects and mechanisms of nanoparticles, it is essential to study nanoparticle systems that are covered with biologically abundant molecules.

This article summarises our present knowledge on the impact of nanoparticle model systems on amyloid peptide aggregation, with a special focus on noble metal (gold) nanoparticles. General patterns for the influence of a number of parameters were shown and discussed in the context of the strength of attractive forces between peptides and nanoparticle surfaces. In a nutshell, too strong peptide–nanoparticle interactions inhibit amyloid aggregation due to inhibited structural rearrangements, whereas too weak peptide–nanoparticle interactions have no effect because the peptides do not sense large enough external constraints. Intermediate attraction leads to a binding, structural rearrangements and subsequent amyloid aggregation acceleration.

5. Methods

5.1 In situ localised surface plasmon resonance (LSPR) and quartz crystal microbalance with dissipation monitoring (QCM-D) measurements

The localised surface plasmon resonance (LSPR) is a collective oscillation of electrons in nanoparticles upon incident light excitation.¹⁷⁵ LSPR is highly confined to the nanoparticle surface. The sensitivity to changes of the dielectric constant in the immediate adjacent environment allows to sense physical and chemical properties in nanoscopic volumes by spectral changes of the reflected light.^176,195 A quartz crystal microbalance with dissipation monitoring (QCM-D) applies the piezoelectric effect of single quartz crystals.¹⁷⁷ Binding of peptides onto the sensor surface causes mass changes and thus frequency changes of the oscillating quartz crystal with nanogram sensitivity.^179,196 Viscoelastic properties of the bound molecules can be obtained from measured losses in energy, recorded as dissipation values.¹⁹⁷

In this work, LSPR (Acoulyte, Insplorion AB, Göteborg, Sweden) was used in conjunction with QCM-D (E1, Q-Sense, Biolin Scientific, Göteborg, Sweden) to measure the adsorption of peptides onto a sensor surface. Sensors consisted of plasmonic gold nanodisks (approximate diameter 100 nm, height 10 nm, 10% coverage) deposited on the SiO₂ top layer of a 5 MHz quartz crystal. The experiments are similar to previously described QCM-D measurements,¹⁷⁸ but were performed in a window chamber to allow optical measurements for LSPR. Cleaning of the sensors was adapted to care for the nanoplasmonic surface: Harsh chemical cleaning was omitted and sensors were instead cleaned twice with UV/ozone (UV/Ozone ProCleaner, BioForce Nanosciences Inc., Ames (IA), USA) for 10 min, rinsed with UV-treated ultrapure water of 18.2 MΩ cm (arium mini, Sartorius AG, Göttingen, Germany) and propan-2-ol (LiChrosolv, Merck KGaA, Darmstadt, Germany) and dried under a gentle nitrogen stream. LSPR data were recorded at 1 Hz from 370 to 1000 nm, averaged from 100 readings with an integration time of 0.3 ms and analysed within the software Insplorer (Insplorion AB, Göteborg, Sweden). The analysis range was set to 470–850 nm for peak finding and a 60 nm span width to give the centroid value (centre of mass) for the plasmon extinction peak λ_LSPR. QCM-D data were processed by QTools (Q-Sense, Biolin Scientific, Göteborg, Sweden) and analysed in OriginPro 8G (OriginLab Corp., Northampton (MA), USA). Changes in frequency and dissipation are shown for the 7^th harmonic.

The peptides, H-GNNQQNY-OH and H-NNFGAIL-OH, were purchased from Eurogentec S.A. (Seraing, Belgium) with a purity of >95% confirmed by HPLC. Both peptides (0.3 mg mL⁻¹ GNNQQNY, 0.1 mg mL⁻¹ NNFGAIL) were freshly prepared and dissolved in ultrapure water with 0.5% DMSO (Merck KGaA, Darmstadt, Germany) and filtered through hydrophobic polypropylene membrane filters (GH Polypro, 0.2 μm, PALL Life Sciences, USA) prior to each injection into the system. The measuring cells were rinsed with ultrapure water with 0.5% DMSO before each peptide introduction step to obtain a baseline. Peptide solutions were introduced at 5 μL min⁻¹ for 15 min. The system was equilibrated without flow for 45 min and finally rinsed with ultrapure water with 0.5% DMSO.

5.2 Molecular dynamics (MD) simulations

The MD simulations were performed using GROMACS 4.5.7 at 300 K.^198–202 Structural information for the studied peptides GNNQQNY (PDB: 2OMM),⁸⁴ NNFGAIL (PDB: 3DGJ)²⁰³ and VQIVYK (PDB: 2ON9)⁸⁴ were obtained from the Protein Data Bank (PDB). The peptides were simulated with charged termini.

Simulations were visualised using VMD²⁰⁴ and analysed with GROMACS tools. Data were plotted with OriginPro 8G (OriginLab Corp., Northampton (MA), USA).

Conflicts of interest

The authors declare no conflict of interest.

Acknowledgements

This research was financially supported by the German Science Foundation (DFG, SFB-TRR 102, project B1). TJ thanks the Friedrich-Ebert-Stiftung for a PhD fellowship, and the Australian Government Department of Education and Training and Scope Global for the support through a 2018 Endeavour Research Fellowship. The authors acknowledge Jenny Andersson from Insplorion AB (Göteborg, Sweden) for providing instrumentation and support for LSPR and QCM-D measurements. BA thanks the Miller Institute of the UC Berkeley for a Somorjai-Miller-Guest-Professorship Award 2018.

References

1. W. Norde, Colloids Surf., B, 2008, 61, 1–9 CrossRef PubMed .
2. J. J. Gray, Curr. Opin. Struct. Biol., 2004, 14, 110–115 CrossRef PubMed .
3. C. Cabaleiro-Lago, O. Szczepankiewicz and S. Linse, Langmuir, 2012, 28, 1852–1857 CrossRef PubMed .
4. K. Nakanishi, T. Sakiyama and K. Imamura, J. Biosci. Bioeng., 2001, 91, 233–244 CrossRef PubMed .
5. J. M. Anderson, A. Rodriguez and D. T. Chang, Semin. Immunol., 2008, 20, 86–100 CrossRef PubMed .
6. C. Czeslik, Z. Phys. Chem., 2004, 218, 771–801 CrossRef .
7. I. Lynch and K. A. Dawson, Nano Today, 2008, 3, 40–47 CrossRef .
8. T. Cedervall, I. Lynch, S. Lindman, T. Berggård, E. Thulin, H. Nilsson, K. A. Dawson and S. Linse, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 2050–2055 CrossRef PubMed .
9. M. Lundqvist, J. Stigler, G. Elia, I. Lynch, T. Cedervall and K. A. Dawson, Proc. Natl. Acad. Sci. U. S. A., 2008, 105, 14265–14270 CrossRef PubMed .
10. J. A. Yang, B. J. Johnson, S. Wu, W. S. Woods, J. M. George and C. J. Murphy, Langmuir, 2013, 29, 4603–4615 CrossRef PubMed .
11. A. Gladytz, B. Abel and H. J. Risselada, Angew. Chem., Int. Ed., 2016, 55, 11242–11246 CrossRef PubMed .
12. P. Jonkheijm, D. Weinrich, H. Schröder, C. M. Niemeyer and H. Waldmann, Angew. Chem., Int. Ed., 2008, 47, 9618–9647 CrossRef PubMed .
13. V. Hlady and J. Buijs, Curr. Opin. Biotechnol., 1996, 7, 72–77 CrossRef PubMed .
14. F. Iori, R. Di Felice, E. Molinari and S. Corni, J. Comput. Chem., 2009, 30, 1465–1476 CrossRef PubMed .
15. A. Keller, M. Fritzsche, Y. P. Yu, Q. Liu, Y. M. Li, M. Dong and F. Besenbacher, ACS Nano, 2011, 5, 2770–2778 CrossRef PubMed .
16. R. L. Willett, K. W. Baldwin, K. W. West and L. N. Pfeiffer, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 7817–7822 CrossRef PubMed .
17. R. Hassert, M. Pagel, Z. Ming, T. Häupl, B. Abel, K. Braun, M. Wiessler and A. G. Beck-Sickinger, Bioconjugate Chem., 2012, 23, 2129–2137 CrossRef PubMed .
18. A. Gladytz, T. John, T. Gladytz, R. Hassert, M. Pagel, S. Naumov, H. J. Risselada, A. G. Beck-Sickinger and B. Abel, Phys. Chem. Chem. Phys., 2016, 18, 23516–23527 RSC .
19. M. Pagel, R. Hassert, T. John, K. Braun, M. Wießler, B. Abel and A. G. Beck-Sickinger, Angew. Chem., Int. Ed., 2016, 55, 4826–4830 CrossRef PubMed .
20. D. Khatayevich, M. Gungormus, H. Yazici, C. So, S. Cetinel, H. Ma, A. Jen, C. Tamerler and M. Sarikaya, Acta Biomater., 2010, 6, 4634–4641 CrossRef PubMed .
21. S. R. Meyers and M. W. Grinstaff, Chem. Rev., 2012, 112, 1615–1632 CrossRef PubMed .
22. D. B. Kittelson, J. Aerosol Sci., 1998, 29, 575–588 CrossRef .
23. Q. Zhiqiang, K. Siegmann, A. Keller, U. Matter, L. Scherrer and H. C. Siegmann, Atmos. Environ., 2000, 34, 443–451 CrossRef .
24. J. R. Cole and N. J. Halas, Appl. Phys. Lett., 2006, 89, 153120 CrossRef .
25. R. H. Müller, M. Radtke and S. A. Wissing, Adv. Drug Delivery Rev., 2002, 54, 131–155 CrossRef .
26. P. Somasundaran, S. Chakraborty, Q. Qiang, P. Deo, J. Wang and R. Zhang, J. Cosmet. Sci., 2004, 55, S1–S18 Search PubMed .
27. J. Pardeike, A. Hommoss and R. H. Müller, Int. J. Pharm., 2009, 366, 170–184 CrossRef PubMed .
28. M. Rai, A. Yadav and A. Gade, Biotechnol. Adv., 2009, 27, 76–83 CrossRef PubMed .
29. I. Sondi and B. Salopek-Sondi, J. Colloid Interface Sci., 2004, 275, 177–182 CrossRef PubMed .
30. V. Wagner, A. Dullaart, A.-K. Bock and A. Zweck, Nat. Biotechnol., 2006, 24, 1211–1218 CrossRef PubMed .
31. K. Riehemann, S. W. Schneider, T. A. Luger, B. Godin, M. Ferrari and H. Fuchs, Angew. Chem., Int. Ed., 2009, 48, 872–897 CrossRef PubMed .
32. I. Javed, Y. Sun, J. Adamcik, B. Wang, A. Kakinen, E. H. Pilkington, F. Ding, R. Mezzenga, T. P. Davis and P. C. Ke, Biomacromolecules, 2017, 18, 4316–4322 CrossRef PubMed .
33. International Bureau, WO2017/223085Al, 2017 Search PubMed .
34. V. P. Zharov, K. E. Mercer, E. N. Galitovskaya and M. S. Smeltzer, Biophys. J., 2006, 90, 619–627 CrossRef PubMed .
35. H. M. E. Azzazy, M. M. H. Mansour and S. C. Kazmierczak, Clin. Chem., 2006, 52, 1238–1246 CrossRef PubMed .
36. T. R. Walsh and M. R. Knecht, Chem. Rev., 2017, 117, 12641–12704 CrossRef PubMed .
37. A. Gladytz, M. Wagner, T. Häupl, C. Elsner and B. Abel, Part. Part. Syst. Charact., 2015, 32, 573–582 CrossRef .
38. Y. Liao, Y. Chang, Y. Yoshiike, Y. Chang and Y. Chen, Small, 2012, 8, 3631–3639 CrossRef PubMed .
39. G. Brancolini, A. Corazza, M. Vuano, F. Fogolari, M. C. Mimmi, V. Bellotti, M. Stoppini, S. Corni and G. Esposito, ACS Nano, 2015, 9, 2600–2613 CrossRef PubMed .
40. G. Gao, M. Zhang, D. Gong, R. Chen, X. Hu and T. Sun, Nanoscale, 2017, 9, 4107–4113 RSC .
41. C. Cabaleiro-Lago, F. Quinlan-Pluck, I. Lynch, S. Lindman, A. M. Minogue, E. Thulin, D. M. Walsh, K. A. Dawson and S. Linse, J. Am. Chem. Soc., 2008, 130, 15437–15443 CrossRef PubMed .
42. Q. Ma, G. Wei and X. Yang, Nanoscale, 2013, 5, 10397 RSC .
43. Y. D. Álvarez, J. A. Fauerbach, J. V. Pellegrotti, T. M. Jovin, E. A. Jares-Erijman and F. D. Stefani, Nano Lett., 2013, 13, 6156–6163 CrossRef PubMed .
44. V. L. Colvin and K. M. Kulinowski, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 8679–8680 CrossRef PubMed .
45. L. Fei and S. Perrett, Int. J. Mol. Sci., 2009, 10, 646–655 CrossRef PubMed .
46. P. C. Ke, M. A. Sani, F. Ding, A. Kakinen, I. Javed, F. Separovic, T. P. Davis and R. Mezzenga, Chem. Soc. Rev., 2017, 46, 6492–6531 RSC .
47. A. Lotfabadi, M. J. Hajipour, H. Derakhshankhah, A. Peirovi, S. Saffar, E. Shams, E. Fatemi, E. Barzegari, S. Sarvari, F. Moakedi, M. Ferdousi, F. Atyabi, A. A. Saboury and R. Dinarvand, ACS Chem. Neurosci., 2018, 9, 1725–1734 CrossRef PubMed .
48. M. W. Risør, D. W. Juhl, M. Bjerring, J. Mathiesen, J. J. Enghild, N. C. Nielsen and D. E. Otzen, Biophys. J., 2017, 113, 580–596 CrossRef PubMed .
49. M. Goedert and M. G. Spillantini, Science, 2006, 314, 777–781 CrossRef PubMed .
50. I. W. Hamley, Chem. Rev. Am. Chem. Soc., 2012, 112, 5147–5192 CrossRef PubMed .
51. Y. Raz, J. Adler, A. Vogel, H. A. Scheidt, T. Häupl, B. Abel, D. Huster and Y. Miller, Phys. Chem. Chem. Phys., 2014, 16, 7710–7717 RSC .
52. J. Adler, M. Baumann, B. Voigt, H. A. Scheidt, D. Bhowmik, T. Häupl, B. Abel, P. K. Madhu, J. Balbach, S. Maiti and D. Huster, ChemPhysChem, 2016, 17, 2744–2753 CrossRef PubMed .
53. H. A. Scheidt, I. Morgado, S. Rothemund and D. Huster, J. Biol. Chem., 2012, 287, 2017–2021 CrossRef PubMed .
54. S. B. Prusiner, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 13363–13383 CrossRef .
55. C.-Y. King, P. Tittmann, H. Gross, R. Gebert, M. Aebi and K. Wuthrich, Proc. Natl. Acad. Sci. U. S. A., 1997, 94, 6618–6622 CrossRef .
56. S. B. Padrick and A. D. Miranker, Biochemistry, 2002, 41, 4694–4703 CrossRef PubMed .
57. J. D. Green, C. Goldsbury, J. Kistler, G. J. S. Cooper and U. Aebi, J. Biol. Chem., 2004, 279, 12206–12212 CrossRef PubMed .
58. M. Wang, A. Kakinen, E. H. Pilkington, T. P. Davis and P. C. Ke, Biomater. Sci., 2017, 5, 485–493 RSC .
59. H. A. Lashuel, B. M. Petre, J. Wall, M. Simon, R. J. Nowak, T. Walz and P. T. Lansbury, J. Mol. Biol., 2002, 322, 1089–1102 CrossRef PubMed .
60. S. Gallegos, C. Pacheco, C. Peters, C. Opazo and L. G. Aguayo, Front. Neurosci., 2015, 9, 1–11 Search PubMed .
61. M. S. Goldberg and P. T. Lansbury, Nat. Cell Biol., 2000, 2, E115–E119 CrossRef PubMed .
62. F. Chiti and C. M. Dobson, Annu. Rev. Biochem., 2006, 75, 333–366 CrossRef PubMed .
63. F. Chiti and C. M. Dobson, Annu. Rev. Biochem., 2017, 86, 27–68 CrossRef PubMed .
64. M. Bucciantini, E. Giannoni, F. Chiti, F. Baroni, L. Formigli, J. Zurdo, N. Taddei, G. Ramponi, C. M. Dobson and M. Stefani, Nature, 2002, 416, 507–511 CrossRef PubMed .
65. Y. Kim, J.-H. Park, H. Lee and J.-M. Nam, Sci. Rep., 2016, 6, 19548 CrossRef PubMed .
66. C. Cabaleiro-Lago, I. Lynch, K. A. Dawson and S. Linse, Langmuir, 2010, 26, 3453–3461 CrossRef PubMed .
67. C. Cabaleiro-Lago, F. Quinlan-Pluck, I. Lynch, K. A. Dawson and S. Linse, ACS Chem. Neurosci., 2010, 1, 279–287 CrossRef PubMed .
68. D. Zhang, O. Neumann, H. Wang, V. M. Yuwono, A. Barhoumi, M. Perham, J. D. Hartgerink, P. Wittung-Stafshede and N. J. Halas, Nano Lett., 2009, 9, 666–671 CrossRef PubMed .
69. R. Vácha, S. Linse and M. Lund, J. Am. Chem. Soc., 2014, 136, 11776–11782 CrossRef PubMed .
70. K. Dubey, B. G. Anand, R. Badhwar, G. Bagler, P. N. Navya, H. K. Daima and K. Kar, Amino Acids, 2015, 47, 2551–2560 CrossRef PubMed .
71. L. Bellucci, A. Ardèvol, M. Parrinello, H. Lutz, H. Lu, T. Weidner and S. Corni, Nanoscale, 2016, 8737–8748 RSC .
72. P. Nedumpully-Govindan, E. N. Gurzov, P. Chen, E. H. Pilkington, W. J. Stanley, S. A. Litwak, T. P. Davis, P. C. Ke and F. Ding, Phys. Chem. Chem. Phys., 2016, 18, 94–100 RSC .
73. S. Auer, A. Trovato and M. Vendruscolo, PLoS Comput. Biol., 2009, 5, e1000458 CrossRef PubMed .
74. C. Elsner, D. Hintzen, A. Prager, K. R. Siefermann and B. Abel, Z. Phys. Chem., 2015, 229, 427–442 Search PubMed .
75. S. Krishnan, C. J. Weinman and C. K. Ober, J. Mater. Chem., 2008, 18, 3405–3413 RSC .
76. A. Wolter, R. Niessner and M. Seidel, Anal. Chem., 2007, 79, 4529–4537 CrossRef PubMed .
77. M. Malmsten, K. Emoto and J. M. van Alstine, J. Colloid Interface Sci., 1998, 202, 507–517 CrossRef .
78. M. B. Pepys, Annu. Rev. Med., 2006, 57, 223–241 CrossRef PubMed .
79. D. Eisenberg and M. Jucker, Cell, 2012, 148, 1188–1203 CrossRef PubMed .
80. E. D. Eanes and G. G. Glenner, J. Histochem. Cytochem., 1968, 16, 673–677 CrossRef PubMed .
81. M. Sunde, L. C. Serpell, M. Bartlam, P. E. Fraser, M. B. Pepys and C. C. F. Blake, J. Mol. Biol., 1997, 273, 729–739 CrossRef PubMed .
82. O. S. Makin and L. C. Serpell, FEBS J., 2005, 272, 5950–5961 CrossRef PubMed .
83. W. T. Astbury, S. Dickinson and K. Bailey, Biochem. J., 1935, 29, 2351–2360 CrossRef PubMed .
84. M. R. Sawaya, S. Sambashivan, R. Nelson, M. I. Ivanova, S. A. Sievers, M. I. Apostol, M. J. Thompson, M. Balbirnie, J. J. W. Wiltzius, H. T. McFarlane, A. Ø. Madsen, C. Riekel and D. Eisenberg, Nature, 2007, 447, 453–457 CrossRef PubMed .
85. H. Puchtler and F. Sweat, J. Histochem. Cytochem., 1965, 13, 693–694 CrossRef PubMed .
86. H. Puchtler, F. Sweat and M. Levine, J. Histochem. Cytochem., 1962, 10, 355–364 CrossRef .
87. M. Biancalana and S. Koide, Biochim. Biophys. Acta, Proteins Proteomics, 2010, 1804, 1405–1412 CrossRef PubMed .
88. J. Duboisset, P. Ferrand, W. He, X. Wang, H. Rigneault and S. Brasselet, J. Phys. Chem. B, 2013, 117, 784–788 CrossRef PubMed .
89. H. Levine, Protein Sci., 1993, 2, 404–410 CrossRef PubMed .
90. C. Xue, T. Y. Lin, D. Chang and Z. Guo, R. Soc. Open Sci., 2017, 4, 160696 CrossRef PubMed .
91. M. R. H. Krebs, E. H. C. Bromley and A. M. Donald, J. Struct. Biol., 2005, 149, 30–37 CrossRef PubMed .
92. L. S. Wolfe, M. F. Calabrese, A. Nath, D. V. Blaho, A. D. Miranker and Y. Xiong, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 16863–16868 CrossRef PubMed .
93. M. R. Nilsson, Methods, 2004, 34, 151–160 CrossRef PubMed .
94. D. Pinotsi, A. K. Büll, C. Galvagnion, M. Dobson, G. S. K. Schierle and C. F. Kaminski, Nano Lett., 2014, 14, 339–345 CrossRef PubMed .
95. J. Seo, W. Hoffmann, S. Warnke, X. Huang, S. Gewinner, W. Schöllkopf, M. T. Bowers, G. von Helden and K. Pagel, Nat. Chem., 2016, 9, 39–44 CrossRef PubMed .
96. J. L. Larson, E. Ko and A. D. Miranker, Protein Sci., 2000, 9, 427–431 CrossRef PubMed .
97. R. Kayed, E. Head, J. L. Thompson, T. M. McIntire, S. C. Milton, C. W. Cotman and C. G. Glabe, Science, 2003, 300, 486–489 CrossRef PubMed .
98. M. Bucciantini, G. Calloni, F. Chiti, L. Formigli, D. Nosi, C. M. Dobson and M. Stefani, J. Biol. Chem., 2004, 279, 31374–31382 CrossRef PubMed .
99. C. G. Glabe, J. Biol. Chem., 2008, 283, 29639–29643 CrossRef PubMed .
100. R. Kayed and C. A. Lasagna-Reeves, J. Alzheimer's Dis., 2013, 33, S67–S78 Search PubMed .
101. M. Stefani and C. M. Dobson, J. Mol. Med., 2003, 81, 678–699 CrossRef PubMed .
102. C. Goldsbury, P. Frey, V. Olivieri, U. Aebi and S. A. Müller, J. Mol. Biol., 2005, 352, 282–298 CrossRef PubMed .
103. S. Matsumura, K. Shinoda, M. Yamada, S. Yokojima, M. Inoue, T. Ohnishi, T. Shimada, K. Kikuchi, D. Masui, S. Hashimoto, M. Sato, A. Ito, M. Akioka, S. Takagi, Y. Nakamura, K. Nemoto, Y. Hasegawa, H. Takamoto, H. Inoue, S. Nakamura, Y. I. Nabeshima, D. B. Teplow, M. Kinjo and M. Hoshia, J. Biol. Chem., 2011, 286, 11555–11562 CrossRef PubMed .
104. D. Lee, G. Lee and D. S. Yoon, Analyst, 2018, 143, 2204–2212 RSC .
105. N. L. Fawzi, Y. Okabe, E. H. Yap and T. Head-Gordon, J. Mol. Biol., 2007, 365, 535–550 CrossRef PubMed .
106. S. I. A. Cohen, M. Vendruscolo, C. M. Dobson and T. P. J. Knowles, J. Mol. Biol., 2012, 421, 160–171 CrossRef PubMed .
107. P. Arosio, T. P. J. Knowles and S. Linse, Phys. Chem. Chem. Phys., 2015, 17, 7606–7618 RSC .
108. A. Sidhu, I. Segers-Nolten, V. Raussens, M. M. A. E. Claessens and V. Subramaniam, ACS Chem. Neurosci., 2017, 8, 538–547 CrossRef PubMed .
109. G. Wei, Z. Su, N. P. Reynolds, P. Arosio, I. W. Hamley, E. Gazit and R. Mezzenga, Chem. Soc. Rev., 2017, 46, 4661–4708 RSC .
110. A. Lomakin, D. B. Teplow, D. A. Kirschner and G. B. Benedek, Proc. Natl. Acad. Sci. U. S. A., 1997, 94, 7942–7947 CrossRef .
111. J. E. Gillam and C. E. MacPhee, J. Phys.: Condens. Matter, 2013, 25, 373101 CrossRef PubMed .
112. J. Nasica-Labouze and N. Mousseau, PLoS Comput. Biol., 2012, 8, e1002782 CrossRef PubMed .
113. M. Volmer and A. Weber, Z. Phys. Chem., 1926, 119, 277–301 Search PubMed .
114. R. Becker and W. Döring, Ann. Phys., 1935, 416, 719–752 CrossRef .
115. J. Frenkel, J. Chem. Phys., 1939, 7, 538–547 CrossRef .
116. J. W. Kelly, Nat. Struct. Biol., 2000, 7, 824–826 CrossRef PubMed .
117. M. A. Watzky, A. M. Morris, E. D. Ross and R. G. Finke, Biochemistry, 2008, 47, 10790–10800 CrossRef PubMed .
118. A. M. Morris, M. A. Watzky and R. G. Finke, Biochim. Biophys. Acta, Proteins Proteomics, 2009, 1794, 375–397 CrossRef PubMed .
119. D. Eisenberg, R. Nelson, M. R. Sawaya, M. Balbirnie, S. Sambashivan, M. I. Ivanova, A. Ø. Madsen and C. Riekel, Acc. Chem. Res., 2006, 39, 568–575 CrossRef PubMed .
120. A. M. Morris, M. A. Watzky, J. N. Agar and R. G. Finke, Biochemistry, 2008, 47, 2413–2427 CrossRef PubMed .
121. M. A. Watzky and R. G. Finke, J. Am. Chem. Soc., 1997, 119, 10382–10400 CrossRef .
122. T. R. Serio, A. G. Cashikar, A. S. Kowal, G. J. Sawicki, J. J. Moslehi, L. Serpell, M. F. Arnsdorf and S. L. Lindquist, Science, 2000, 289, 1317–1321 CrossRef PubMed .
123. J. D. Harper and P. T. Lansbury Jr., Annu. Rev. Biochem., 1997, 66, 385–407 CrossRef PubMed .
124. J. H. Come, P. E. Fraser and P. T. Lansbury, Proc. Natl. Acad. Sci. U. S. A., 1993, 90, 5959–5963 CrossRef .
125. J. T. Jarrett and P. T. Lansbury, Cell, 1993, 73, 1055–1058 CrossRef PubMed .
126. L. Nielsen, R. Khurana, A. Coats, S. Frokjaer, J. Brange, S. Vyas, V. N. Uversky and A. L. Fink, Biochemistry, 2001, 40, 6036–6046 CrossRef PubMed .
127. C.-C. Lee, A. Nayak, A. Sethuraman, G. Belfort and G. J. McRae, Biophys. J., 2007, 92, 3448–3458 CrossRef PubMed .
128. S. Griffith, Nature, 1967, 215, 1043–1044 CrossRef PubMed .
129. S. I. A. Cohen, S. Linse, L. M. Luheshi, E. Hellstrand, D. A. White, L. Rajah, D. E. Otzen, M. Vendruscolo, C. M. Dobson and T. P. J. Knowles, Proc. Natl. Acad. Sci. U. S. A., 2013, 110, 9758–9763 CrossRef PubMed .
130. F. A. Ferrone, J. Mol. Biol., 2015, 427, 287–290 CrossRef PubMed .
131. S. Prusiner, Science, 1991, 252, 1515–1522 CrossRef PubMed .
132. F. E. Cohen, K. M. Pan, Z. Huang, M. Baldwin, R. J. Fletterick and S. B. Prusiner, Science, 1994, 264, 530–531 CrossRef PubMed .
133. J. Lee, E. K. Culyba, E. T. Powers and J. W. Kelly, Nat. Chem. Biol., 2011, 7, 602–609 CrossRef PubMed .
134. R. Krishnan and S. L. Lindquist, Nature, 2005, 435, 765–772 CrossRef PubMed .
135. T. Eichner, A. P. Kalverda, G. S. Thompson, S. W. Homans and S. E. Radford, Mol. Cell, 2011, 41, 161–172 CrossRef PubMed .
136. S. Kumar, S. K. Mohanty and J. B. Udgaonkar, J. Mol. Biol., 2007, 367, 1186–1204 CrossRef PubMed .
137. A. Gladytz, E. Lugovoy, A. Charvat, T. Häupl, K. R. Siefermann and B. Abel, Phys. Chem. Chem. Phys., 2015, 17, 918–927 RSC .
138. B. Abel, A. Charvat, U. Diederichsen, M. Faubel, B. Girmann, J. Niemeyer and A. Zeeck, Int. J. Mass Spectrom., 2005, 243, 177–188 CrossRef .
139. A. Charvat and B. Abel, Phys. Chem. Chem. Phys., 2007, 9, 3335–3360 RSC .
140. M.-C. Daniel and D. Astruc, Chem. Rev., 2004, 104, 293–346 CrossRef PubMed .
141. G. Brancolini, L. Z. Polzi and S. Corni, J. Self-Assem. Mol. Electron., 2015, 2, 1–26 CrossRef .
142. C. A. E. Hauser, S. Maurer-Stroh and I. C. Martins, Chem. Soc. Rev., 2014, 43, 5326 RSC .
143. M. Mahmoudi, H. R. Kalhor, S. Laurent and I. Lynch, Nanoscale, 2013, 5, 2570–2588 RSC .
144. M. Mahmoudi, I. Lynch, M. R. Ejtehadi, M. P. Monopoli, F. B. Bombelli and S. Laurent, Chem. Rev., 2011, 111, 5610–5637 CrossRef PubMed .
145. S. Linse, C. Cabaleiro-Lago, W.-F. Xue, I. Lynch, S. Lindman, E. Thulin, S. E. Radford and K. A. Dawson, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 8691–8696 CrossRef PubMed .
146. M. Assfalg, L. Ragona, K. Pagano, M. D'Onofrio, S. Zanzoni, S. Tomaselli and H. Molinari, Biochim. Biophys. Acta, Proteins Proteomics, 2016, 1864, 102–114 CrossRef PubMed .
147. M. Zhang, X. Mao, Y. Yu, C.-X. Wang, Y.-L. Yang and C. Wang, Adv. Mater., 2013, 25, 3780–3801 CrossRef PubMed .
148. N. Xiong, Y. Zhao, X. Dong, J. Zheng and Y. Sun, Small, 2017, 13, 1–14 Search PubMed .
149. K. Shezad, K. Zhang, M. Hussain, H. Dong, C. He, X. Gong, X. Xie, J. Zhu and L. Shen, Langmuir, 2016, 32, 8238–8244 CrossRef PubMed .
150. H. M. Chan, L. Xiao, K. M. Yeung, S. L. Ho, D. Zhao, W. H. Chan and H. W. Li, Biomaterials, 2012, 33, 4443–4450 CrossRef PubMed .
151. Z. Bednarikova, P. D. Q. Huy, M.-M. Mocanu, D. Fedunova, M. S. Li and Z. Gazova, Phys. Chem. Chem. Phys., 2016, 18, 18855–18867 RSC .
152. H. Arkin and W. Janke, Phys. Rev. E, 2017, 96, 1–14 CrossRef PubMed .
153. A. Morriss-Andrews, G. Bellesia and J.-E. Shea, J. Chem. Phys., 2012, 137, 145104 CrossRef PubMed .
154. H. Y. Kim and I. Choi, Nanoscale, 2016, 8, 1952–1959 RSC .
155. H. R. Barros, M. Kokkinopoulou, I. C. Riegel-Vidotti, K. Landfester and H. Thérien-Aubin, RSC Adv., 2018, 8, 980–986 RSC .
156. K. A. Moore, K. M. Pate, D. D. Soto-Ortega, S. Lohse, N. van der Munnik, M. Lim, K. S. Jackson, V. D. Lyles, L. Jones, N. Glassgow, V. M. Napumecheno, S. Mobley, M. J. Uline, R. Mahtab, C. J. Murphy and M. A. Moss, J. Biol. Eng., 2017, 11, 1–11 CrossRef PubMed .
157. S. Sen, S. Dasgupta and S. DasGupta, J. Phys. Chem. C, 2017, 121, 18935–18946 CrossRef .
158. S. Campioni, G. Carret, S. Jordens, L. Nicoud, R. Mezzenga and R. Riek, J. Am. Chem. Soc., 2014, 136, 2866–2875 CrossRef PubMed .
159. R. P. Sear, J. Phys.: Condens. Matter, 2007, 19, 33101 CrossRef .
160. S. Radic, T. P. Davis, P. C. Ke and F. Ding, RSC Adv., 2015, 5, 105489–105498 RSC .
161. S. Sen, S. Konar, A. Pathak, S. Dasgupta and S. DasGupta, J. Phys. Chem. B, 2014, 118, 11667–11676 CrossRef PubMed .
162. A. Morriss-Andrews, G. Bellesia and J. E. Shea, J. Chem. Phys., 2011, 135, 085102 CrossRef PubMed .
163. C. E. Giacomelli and W. Norde, Macromol. Biosci., 2005, 5, 401–407 CrossRef PubMed .
164. F. Grigolato, C. Colombo, R. Ferrari, L. Rezabkova and P. Arosio, ACS Nano, 2017, 11, 11358–11367 CrossRef PubMed .
165. J. Chen, A. H. Armstrong, A. N. Koehler and M. H. Hecht, J. Am. Chem. Soc., 2010, 132, 17015–17022 CrossRef PubMed .
166. J. Bieschke, M. Herbst, T. Wiglenda, R. P. Friedrich, A. Boeddrich, F. Schiele, D. Kleckers, J. M. Lopez Del Amo, B. A. Grüning, Q. Wang, M. R. Schmidt, R. Lurz, R. Anwyl, S. Schnoegl, M. Fändrich, R. F. Frank, B. Reif, S. Günther, D. M. Walsh and E. E. Wanker, Nat. Chem. Biol., 2012, 8, 93–101 CrossRef PubMed .
167. E. H. Pilkington, M. Lai, X. Ge, W. J. Stanley, B. Wang, M. Wang, A. Kakinen, M. A. Sani, M. R. Whittaker, E. N. Gurzov, F. Ding, J. F. Quinn, T. P. Davis and P. C. Ke, Biomacromolecules, 2017, 18, 4249–4260 CrossRef PubMed .
168. A. K. Jana, M. K. Tiwari, K. Vanka and N. Sengupta, Phys. Chem. Chem. Phys., 2016, 18, 5910–5924 RSC .
169. M.-S. Lin, L.-Y. Chen, H.-T. Tsai, S. S.-S. Wang, Y. Chang, A. Higuchi and W.-Y. Chen, Langmuir, 2008, 24, 5802–5808 CrossRef PubMed .
170. S. J. Wood, B. Maleeff, T. Hart and R. Wetzel, J. Mol. Biol., 1996, 256, 870–877 CrossRef PubMed .
171. A. M. Isaacs, D. B. Senn, M. Yuan, J. P. Shine and B. A. Yankner, J. Biol. Chem., 2006, 281, 27916–27923 CrossRef PubMed .
172. M. Ghavami, M. Rezaei, R. Ejtehadi, M. Lotfi, M. A. Shokrgozar, B. Abd Emamy, J. Raush and M. Mahmoudi, ACS Chem. Neurosci., 2013, 4, 375–378 CrossRef PubMed .
173. J. Zhou, J. Ralston, R. Sedev and D. A. Beattie, J. Colloid Interface Sci., 2009, 331, 251–262 CrossRef PubMed .
174. J. Kimling, M. Maier, B. Okenve, V. Kotaidis, H. Ballot and A. Plech, J. Phys. Chem. B, 2006, 110, 15700–15707 CrossRef PubMed .
175. K. A. Willets and R. P. Van Duyne, Annu. Rev. Phys. Chem., 2007, 58, 267–297 CrossRef PubMed .
176. J. N. Anker, W. P. Hall, O. Lyandres, N. C. Shah, J. Zhao and R. P. Van Duyne, Nat. Mater., 2008, 7, 442–453 CrossRef PubMed .
177. A. Janshoff, H.-J. Galla and C. Steinem, Angew. Chem., Int. Ed., 2000, 39, 4004–4032 CrossRef PubMed .
178. T. John, Z. X. Voo, C. Kubeil, B. Abel, B. Graham, L. Spiccia and L. L. Martin, MedChemComm, 2017, 8, 1112–1120 RSC .
179. G. Sauerbrey, Z. Phys. Chem., 1959, 155, 206–222 Search PubMed .
180. L. L. Martin, C. Kubeil, S. Piantavigna, T. Tikkoo, N. P. Gray, T. John, A. N. Calabrese, Y. Liu, Y. Hong, M. A. Hossain, N. Patil, B. Abel, R. Hoffmann, J. H. Bowie and J. A. Carver, Pept. Sci., 2018, 110, e24052 CrossRef .
181. T. John, B. Abel and L. L. Martin, Aust. J. Chem., 2018, 71, 543 CrossRef .
182. F. Ramezani and H. Rafii-Tabar, Mol. BioSyst., 2015, 11, 454–462 RSC .
183. L. Bellucci, G. Bussi, R. Di Felice and S. Corni, Nanoscale, 2017, 9, 2279–2290 RSC .
184. L. B. Wright, P. M. Rodger, T. R. Walsh and S. Corni, J. Phys. Chem. C, 2013, 117, 24292–24306 CrossRef .
185. L. B. Wright, P. M. Rodger, S. Corni and T. R. Walsh, J. Chem. Theory Comput., 2013, 9, 1616–1630 CrossRef PubMed .
186. M. Hoefling, F. Iori, S. Corni and K.-E. Gottschalk, Langmuir, 2010, 26, 8347–8351 CrossRef PubMed .
187. M. Hoefling, S. Monti, S. Corni and K. E. Gottschalk, PLoS One, 2011, 6, e20925 CrossRef PubMed .
188. L. Bellucci and S. Corni, J. Phys. Chem. C, 2014, 118, 11357–11364 CrossRef .
189. M. Skepö, J. Chem. Phys., 2008, 129, 185101 CrossRef PubMed .
190. H. Lopez and V. Lobaskin, J. Chem. Phys., 2015, 143, 243138 CrossRef PubMed .
191. G. Brancolini, M. C. Maschio, C. Cantarutti, A. Corazza, F. Fogolari, V. Bellotti, S. Corni and G. Esposito, Nanoscale, 2018, 10, 4793–4806 RSC .
192. G. Wei, A. I. Jewett and J.-E. Shea, Phys. Chem. Chem. Phys., 2010, 12, 3622 RSC .
193. S. T. Wang, Y. Lin, N. Todorova, Y. Xu, M. Mazo, S. Rana, V. Leonardo, N. Amdursky, C. D. Spicer, B. D. Alexander, A. A. Edwards, S. J. Matthews, I. Yarovsky and M. M. Stevens, Chem. Mater., 2017, 29, 1550–1560 CrossRef PubMed .
194. M. G. Zagorski and C. J. Barrow, Biochemistry, 1992, 31, 5621–5631 CrossRef PubMed .
195. E. A. Elbassal, C. Morris, T. W. Kent, R. Lantz, B. Ojha, E. P. Wojcikiewicz and D. Du, J. Phys. Chem. C, 2017, 121, 20007–20015 CrossRef PubMed .
196. K. A. Marx, Biomacromolecules, 2003, 4, 1099–1120 CrossRef PubMed .
197. M. V. Voinova, M. Rodahl, M. Jonson and B. Kasemo, Phys. Scr., 1999, 59, 391–396 CrossRef .
198. B. Hess, C. Kutzner, D. van der Spoel and E. Lindahl, J. Chem. Theory Comput., 2008, 4, 435–447 CrossRef PubMed .
199. D. van der Spoel, E. Lindahl, B. Hess, G. Groenhof, A. E. Mark and H. J. C. Berendsen, J. Comput. Chem., 2005, 26, 1701–1718 CrossRef PubMed .
200. E. Lindahl, B. Hess and D. van der Spoel, J. Mol. Model., 2001, 7, 306–317 CrossRef .
201. H. J. C. Berendsen, D. van der Spoel and R. van Drunen, Comput. Phys. Commun., 1995, 91, 43–56 CrossRef .
202. S. Pronk, S. Páll, R. Schulz, P. Larsson, P. Bjelkmar, R. Apostolov, M. R. Shirts, J. C. Smith, P. M. Kasson, D. Van Der Spoel, B. Hess and E. Lindahl, Bioinformatics, 2013, 29, 845–854 CrossRef PubMed .
203. J. J. W. Wiltzius, S. A. Sievers, M. R. Sawaya, D. Cascio, D. Popov, C. Riekel and D. Eisenberg, Protein Sci., 2008, 17, 1467–1474 CrossRef PubMed .
204. W. Humphrey, A. Dalke and K. Schulten, J. Mol. Graphics, 1996, 14, 33–38 CrossRef PubMed .
205. W. L. Jorgensen, D. S. Maxwell and J. Tirado-Rives, J. Am. Chem. Soc., 1996, 118, 11225–11236 CrossRef .
206. W. L. Jorgensen and J. Tirado-Rives, J. Am. Chem. Soc., 1988, 110, 1657–1666 CrossRef PubMed .
207. U. Essmann, L. Perera, M. L. Berkowitz, T. Darden, H. Lee and L. G. Pedersen, J. Chem. Phys., 1995, 103, 8577–8593 CrossRef .
208. T. Darden, D. York and L. Pedersen, J. Chem. Phys., 1993, 98, 10089–10092 CrossRef .
209. B. Hess, H. Bekker, H. J. C. Berendsen and J. G. E. M. Fraaije, J. Comput. Chem., 1997, 18, 1463–1472 CrossRef .
210. S. Miyamoto and P. A. Kollman, J. Comput. Chem., 1992, 13, 952–962 CrossRef .
211. H. J. C. Berendsen, J. P. M. Postma, W. F. van Gunsteren and J. Hermans, in The Jerusalem Symposia on Quantum Chemistry and Biochemistry, ed. B. Pullman, Reidel, Dordrecht, 1981, pp. 331–342 Search PubMed .
212. I. Yeh and M. L. Berkowitz, J. Chem. Phys., 1999, 111, 3155–3162 CrossRef .
213. G. Bussi, D. Donadio and M. Parrinello, J. Chem. Phys., 2007, 126, 14101 CrossRef PubMed .
214. R. T. Cygan, J.-J. Liang and A. G. Kalinichev, J. Phys. Chem. B, 2004, 108, 1255–1266 CrossRef .
215. M. Catti, G. Ferraris and G. Ivaldi, Eur. J. Mineral., 1989, 1, 625–632 CrossRef .
216. L. D. Schuler, X. Daura and W. F. van Gunsteren, J. Comput. Chem., 2001, 22, 1205–1218 CrossRef .

---

Footnotes

† Both authors contributed equally to this work.

‡ Current address: Institute of Fluid Dynamics, Helmholtz-Zentrum Dresden-Rossendorf, Bautzner Landstraße 400, 01328 Dresden, Germany.

§ The contents of the literature review in this article are partially based on the PhD thesis ‘Peptides near interfaces: from affinities to interaction mechanisms’ completed by A. Gladytz in 2016.

This journal is © The Royal Society of Chemistry 2018