WO2011015864A1 - Assay - Google Patents

Abstract

An improved in vitro assay is provided for assaying a sample comprising at least one amphiphilic assay component. The assay is characterized in that it comprises a step of occluding an air water interface at a surface of the assay, whereby amphiphilic aggregation at said assay surface is inhibited.

Description

Assay

This invention relates to an assay of a sample containing at least one amphiphilic species comprising a step of occluding an air water interface from a surface of the assay sample and apparatus for performing said assay.

BACKGROUND

Protein misfolding can be deleterious by leading to the formation of toxic amyloid species, which are the hallmark of many diseases. Despite differences in origin and function, these amyloid-related proteins or peptides possess similar structural, physical and chemical properties (17, 26-29). The physico-chemical properties that many amyloids have in common are amphiphilicity and surface activity (6, 8). In other words, they can accumulate at hydrophilic hydrophobic interfaces (HHIs) and use such interfaces to promote peptide chain alignment, folding and assembly into stacks of cross-β sheets, which would result in the formation of amyloid fibrils (9, 10). Typical examples of HHIs would be a simple air water interface (AWI), nonetheless systematically present in almost all in-vitro assays, and lipid membranes. Conventional amyloid inhibitors are failing in vivo after selection in vitro. In several cases this may be a generalized effect on air-water interfaces in the in vitro assays.

Until now the potential importance of the AWI on in vitro assays has been ignored. Accordingly, there remains a need to provide an in vitro assay system that overcomes the problems associated with amphiphilic sequestering at the assay AWI.

BRIEF SUMMARY OF THE DISCLOSURE

In accordance with a first aspect of the present invention there is provided an in vitro assay comprising at least one amphiphilic assay component in an assay sample, characterized in that said assay comprises a step of occluding an air water interface at a surface of the assay sample, whereby amphiphilic organisation at said assay sample surface is inhibited. Preferably said amphiphilic organisation is amphiphilic concentration, selective orientation, polymerisation or aggregation.

Preferably, said assay is a screening assay.

Preferably, said at least one amphiphilic assay component is a screening candidate. Alternatively, said at least one amphiphilic assay component is an assay substrate. Alternatively, said amphiphilic assay component is an enzymatic target of an assay. Preferably, said assay is performed in an assay vessel, for example a test tube, a dish, a flask or a multiwall plate.

Preferably, said step of occluding an air water interface is achieved by contacting the assay surface with a hydrophilic lid. Preferably, said hydrophilic lid comprises a projection dimensioned such that when the projection is contacted with the assay surface, the assay surface interfaces only with the projection.

Preferably said hydrophilic lid is formed from hydrophilic plastics material, for example polymethyl methacrylate. It will be appreciated that unlike conventional assays, the lid itself plays a role in the chemistry of the reaction under study insofar as it abolishes a catalytic surface with specific chemical properties capable of concentrating and orienting amphiphilic molecules.

In a further aspect the invention provides a closed assay system for conducting an in vitro assay comprising at least one amphiphilic assay component in an assay sample, wherein said system is provided with a means to occlude air from the assay system thereby preventing formation of an air water interface at the assay sample sample surface within said system during performance of an assay therein. Preferably, said closed assay system is a microfluidic assay system.

Preferably, said means to occlude air from the system is a hydrophilic body. Preferably, said hydrophilic body is formed from hydrophilic plastics material, for example polymethyl methacrylate.

Preferably, said at least one amphiphilic assay component is a screening candidate. Alternatively, said at least one amphiphilic assay component is an assay substrate. Alternatively, said amphiphilic assay component is an enzymatic target of an assay.

Preferably, said amphiphilic assay component is a candidate amyloid accumulation inhibitor.

In a further aspect the invention provides an assay apparatus comprising at least one well defined by at least one wall and a well lid characterized in that the well lid comprises at least one hydrophilic projection complementarily shaped to fit within the at least one well and dimensioned such that when the well contains a predetermined volume of assay sample and the projection is inserted into the well, the assay sample interfaces only with the at least one wall and the projection.

The well is defined by said at least one wall and has a closed bottom. The said at least one wall is preferably a continuous wall.

Preferably, the apparatus further comprises a base upon which said at least one wall defines a well.

Preferably, the apparatus comprises a plurality of walls each defining a well on said base and said lid comprises a plurality of hydrophilic projections.

Preferably, said plurality of projections on said lid is equal to said plurality of walls each defining a well. Preferably, said apparatus comprises 96 walls each defining a well.

Preferably, the apparatus further comprises an overflow cavity between each continuous wall defining a well. Preferably said hydrophilic projection(s) is/are formed from hydrophilic plastics material, for example said hydrophilic plastics material is polymethyl methacrylate.

Preferably, said base and walls are formed from glass, plastics material, or ceramic.

Preferably said assay sample comprises at least one amphiphilic assay component.

Preferably said at least one wall defines a well of circular configuration. Preferably, said lid and / or base are optically transparent.

In a further aspect the invention provides use of the closed assay system according to an aspect of the invention or use of the assay apparatus according to an aspect of the invention in an assay comprising an amphiphilic component.

In a further aspect the invention provides an apparatus configured to carry out the assay as hereinbefore described.

In a further aspect the invention provides a method of screening amphiphilic candidate components, the method comprising placing an assay sample in an assay apparatus according to an aspect of the invention and occluding an air water interface by contacting the assay sample surface with a hydrophilic lid so that the assay sample interfaces only with assay apparatus. Preferably, the amphiphilic candidate component is an anti-amyloidogenic agent.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention are further described hereinafter with reference to the accompanying drawings, in which:

Figure 1. AWI removal affects the kinetic properties of IAPP fibrilization. 1.3 (A), or 1.8 (β to E) or 4 μM (F to I) IAPP was incubated with 32 μM ThT with or without AWI removal at various time points (indicated by arrows). ThT fluorescence changes were monitored (A, B and F; insert in F showing a scale up) with the lag phases (C and G), elongation rates (D and H), and plateau heights (£ and I) depicted. In D, the elongation rates before (cyl. elong.) and after (cyl. elong.^*) AWI removal during the elongation are shown, a. u.: arbitrary units. The mean of at least three independent assays is shown. Error bars represent ± standard error of the mean.

Figure 2. Amyloid assembly kinetics considering nucleation pathways from both bulk micelles and AWI monomers. [A) Schematic showing two different nucleation pathways (arrows), from bulk micelles and from interfacial layer monomers. (S and C) Amyloid assembly kinetics resulting from modelling, taking into account nucleation from micelles and AWI monomers, peptide concentrations around (S) or above CMC (C) and simulation parameters similar to experimental values (see Table 1 ).

Figure 3. The enhancement of IAPP nucleation and propagation by anionic lipids is greater in absence of an AWI. 1.8 μM IAPP was incubated with 32 μM ThT with or without AWI removal at the fibrilization start and/or in presence or absence of 2.4 μM liposomes (DOPC or a 7 DOPC:3 DOPG molar ratio). ThT fluorescence changes were monitored (A, with S showing a scale-up to visualize the rapid fibrilization of some reactions). The lag phase (C), plateau height (D) and elongation rate (£) are depicted, a. u.: arbitrary units. The mean of at least three independent assays is shown. Error bars represent ± standard error of the mean.

Figure 4, The role of lipids in an in-vivo context has been previously underestimated. 1.8 μM IAPP was incubated with 32 μM ThT with or without AWI removal at the fibrilization start and/or in presence or absence of various liposome concentrations (DOPC or 7 DOPC:3 DOPG). The lag phases {A), plateau heights (S) and elongation rates (C) are represented as percentages of IAPP reactions with an AWI and without liposomes. The mean of at least three independent assays is shown. Error bars represent ± standard error of the mean. Figure 5 shows a diagram depicting the restrictions on the peptide required to form a dimer upon encountering. The restrictions are: i) θ < 10° and ii) the hydrogen bonding sides (depicted by red panels) of the peptide lie parallel to the x-axis. Figure 6 IAPP is able to form fibrils in the presence or absence of an AWI. Electron micrographs of negatively stained IAPP fibrils formed in the presence (A) or absence of an AWI (S). The scale-bar represents 100 nm. The arrows in (A) show an individual fibril.

Figure 7 illustrates a schematic mid section elevation of a closed assay system of the invention.

Figures 8a illustrates a schematic mid section elevation of an assay apparatus of the invention a) as individual components, b) as part of a multiwell plate and c) in use.

Figure 9 shows the presence of a hydrophobic-hydrophilic interface is as important for Aβ_1-40 as it is for IAPP. Changes in ThT fluorescence (165 μM in PBS) of a 15 μM Aβi. ₄₀ {A) reaction, in presence or absence of AWI at the beginning of fibril formation, were monitored over time. The lag phase of Aβ_1-40 fibrilization (A, inset), elongation rate (S) and plateau height (C) are depicted. Represented are the mean of triplicate experiments, with every experiment done in duplicate wells, a. u.: arbitrary units. The mean of at least three independent assays is shown. Error bars represent ± standard error of the mean.

Figure 10 (A) shows the effect of occlusion of the air-water interface on the assembly of Thioflavin positive fibrils of an amyloidogenic fragment of human neuronal form acetylcholinesterase, the peptide sequence comprising 14-residue peptide located between residues 586-599 at the C-terminus of human AChE. (A) Changes in ThT fluorescence (165 μM in PBS) of a 50 μM peptide solution in presence or absence of AWI at the beginning of fibril formation, were monitored over time. Represented are the mean of triplicate experiments, with every experiment done in duplicate wells, a. u.: arbitrary units. The mean of at least three independent assays is shown. Error bars represent ± standard error of the mean. (B) The elongation rate (slope of the thioflavin signal intensity against time in the fibril elongation phase) in the presence of an air-water interface (no cylinder) is highly significantly different from the elongation rate in the absence of an air-water interface (with cylinder). Figure 1 1 shows a graph demonstrating the effect of air-water interface occlusion on a screening assay for inhibitors of early amyloid oligomerisation.

DETAILED DESCRIPTION

The present inventors have identified the profound effect of a non-physiological air water interface on in vitro systems. In particular, the inventors have identified the significance of the air-water interface (AWI) upon in vitro systems involving

amphiphilic species.

The inventors have used an integrated biophysical modelling and experimental approach firstly to investigate the potential for amyloid surface activity and the AWI to influence the dynamics of amyloid aggregation, and secondly to study the role of lipid membranes in fibrilogenesis under conditions more closely mirroring the in-vivo context (in an AWI-free system). There is evidence that conventional amyloid inhibitors are failing in vivo after selection in vitro. The effect of the AWI may be a significant contributing factor to this failure, as within the body and particularly within cells, no such interfaces are present. The inventors have advantageously found that removal of the AWI dramatically affects the behavior of amphiphilic species in vitro and provides an in vitro system that more closely mimics in vivo conditions.

The amphiphilic sequestering interfacial affect is inherent to current in vitro screening methods, in which the presence of a non-physiological AWI is invariant. Although for some assays the AWI affect may be insignificant, for those assays involving amphiphilic species, the presence of such an interface may have a profound effect, as assay components may selectively concentrate and align at the AWI. The present invention overcomes the interfacial effect by removing the AWI thereby providing a more physiological environment for the assay.

The terms "amphiphile", "amphiphilic species" and "amphiphilic assay component" are used interchangeably herein to describe compounds possessing both hydrophilic and hydrophobic properties. The amphiphilic assay component may be anionic or cationic.

"Amphiphilic organisation", as used herein, relates to the organisation of amphiphilic assay components at an air water interface, and includes concentration, selective orientation, polymerisation or aggregation of amphiphilic species at an AWI. It includes self assembly of amphiphilic assay components into a variety of structural configurations, such as, by way of example only, micelles (spherical or cylindrical aggregates arranged so that the polar head groups are pointing out into the solvent and the non polar tails are sequestered into the aggregate and thereby protected form the solvent), bilayers, bicontinuous structures, inverted micelles, or spherical vesicles.

In one aspect, the invention provides an assay apparatus for conducting an assay in an AWI free environment. Referring now to figure 8a, an assay device of the invention comprises a well 22 defined be a continuous wall 23. The well 22 is provided with a lid 25 having a hydrophilic projection 26 depending therefrom. The hydrophilic projection 26 is shaped fit within the wells 22 and dimensioned such that when the well 22 contains a predetermined volume of assay sample and the projection 26 is inserted into the well 22, the assay sample cannot form an air water interface and instead interfaces only with projection 26 and an internal surface 30 of the continuous wall 23. In use, the fitting of the lid is such that the air-water interface is completely eliminated thus removing the source of catalysis. Referring now to figure 8b, a further assay apparatus of the invention is illustrated generally at 20. The apparatus comprises a base 21 having disposed on an upper surface 31 thereof a plurality wells 22', each defined by a continuous wall 23'. An overflow cavity 27 is provided between each well 22'. The apparatus of the invention can advantageously be freely re-shaped and scaled to a user's requirements.

A well lid 25' is provided having a plurality of hydrophilic projections 26' depending therefrom. The location of each of the projections 26' on the lid 25' corresponds to the location of each of the wells 22' on the base 21. Referring now to figure 8c, in use, each well 22' is loaded with a predetermined volume of assay sample 24. The lid 25' is positioned over the upper surface 31 of the base 21 so that each of the projections 26' is inserted into a well 22'. As each projection 26' enters a well 22, a proportion of the the assay sample 24 is displaced and will enter the overflow cavity 27. The volume of assay sample 24 is sufficient such that upon insertion of the projection 26' and displacement of a proportion of the assay sample into the overflow cavity 27, a desired volume of assay sample 24 is retained in the well 22'. Said volume of assay sample 24 retained in the well 22' cannot form an air water interface and instead interfaces with projection 26' and an internal surface 30' of the continuous wall 23' and an internal surface 32' of the base 21.

The projection 26,26' is dimensioned relative to the well 22,22' such that upon insertion of the projection into the well an outer surface 35 of the projection 26,26' is in contacting relation with the internal surface 30,30' of the wall 23,23'. The contacting relation must be sufficient such that when the well 22,22' contains the assay sample, insertion of the projection 26,26' occludes the AWI.

Alternatively a tray is provided having a plurality of wells downwardly extending therefrom for use with a lid as described above. The tray may be provided with a downwardly extending peripheral wall. Preferably, each well is defined by a continuous wall having a closed bottom.

In a preferred embodiment, the assay apparatus is a multiwell plate comprising a plurality of discrete wells, more preferably a plate comprising 96 wells, 384 wells or 1536 wells, the locations of which correspond to those of a conventional 96, 384 or 1536 well plate respectively, e.g. for a 96 well plate, twelve parallel columns and eight parallel rows. The apparatus of the invention can advantageously be freely re-shaped and scaled to a user's requirements. Preferably, the well(s) have a constant cross section. Preferably, the wells are cylindrical. Preferably, said well(s) and / or bases and / or trays of the assay apparatus are formed from glass, plastics material, or ceramic, more preferably from plastics material, using moulding or injection moulding techniques well known in the art. In use, assay samples are loaded into wells using dispensing methods well known in the art, for example by pipetting. The sample volume required volume required will depend on the assay sample properties, the volume of the well chamber and the dimensions of the projection. Preferably the sample volume is equal to or exceeds the volume of the chamber. The volume of assay sample in the well must be sufficient to interface with the projection, when inserted into the well, so as to occlude the AWI. The present invention employs a continuous solid interface thus ensuring that the AWI is eliminated while at the same time ensuring that there is no contribution to additional species in a bulk fluid. In a further aspect, the present invention provides a closed assay system for conducting an in vitro assay comprising at least one amphiphilic assay component. By closed assay system is meant a system in which an assay can be conducted in a closed environment, such as a chamber or channel. The closed assay system may be a microfluidic system, i.e. a system which incorporates a plurality of interconnected channels or chambers, through which assay samples, and in particular fluidic assay mixtures, can be transported to effect one analysis upon said assay mixtures.

Microfluidic devices comprising microfluidic channels and chambers are well known in the art. The closed system of the invention is provided with a means for occluding air from the system. The means may be a hydrophilic body formed from a hydrophilic material or having a hydrophilic surface. The means may be contacted with any surface of the assay sample in the microfluidic system where an air water interface is capable of forming.

Referring now to figure 7, a microfluidic channel 13 of a closed assay system of the invention is illustrated, in which hydrophilic bodies 10 and 10' are provided at a proximal end 1 1 and a distal end 12 of a microfluidic channel 13 to occlude air and thus occlude an air water interface forming at a first surface 14 and a second surface 15 respectively of an assay mixture 16.

Preferably the hydrophilic body is formed from a plastics material. Alternatively, a surface of the hydrophilic body which will contact the assay sample is hydrophilic. Suitable hydrophilic plastic materials for hydrophilic bodies are described herein below.

In a further aspect the present invention provides an in vitro assay comprising at least one amphiphilic assay component in an assay sample, which comprises a step of occluding an air water interface at a surface of the assay sample, thereby inhibiting amphiphilic aggregation at the assay sample surface.

Occlusion of the AWI is achieved by contacting any surface of an assay, where an AWI exists with a hydrophilic surface, for example a hydrophilic surface of an assay lid.

Preferably, the assay is conducted in an assay vessel such as, by way of example only, a multiwell plate, a test tube, a dish or a flask. Preferably the plate is a 96 well plate, a 384 well plate or a 1536 well plate. Preferably, the lid is dimensioned relative to the assay vessel such that upon contacting the lid with the assay sample surface the lid forms a contact relationship with a surface of the vessel sufficient to occlude the AWI.

A lid and / or projection in accordance with the each one of the aforementioned aspects of the invention are preferably formed form a material which exhibits high wettability, more preferably a material which is sufficiently wettable to maintain contact with the assay surface. Preferably the surface of the lid and / or projection is formed from a high energy solid. Preferably, the lid and / or projection is formed from a hydrophilic material. Preferably the lid and / or projection is formed from a plastics material, for example a hydrophilic plastics material selected from polyvinyl chloride, polyterephthalate, polymethyl methacrylate, or polyamide

Preferably the lid and/or projection is formed from a material which is dimensionally stable over the temperature range of the assay, such that it remains in contacting relation with the assay surface.

Preferably the material is a non-fluorescent material. Preferably the lid and / or projection is dimensioned to complement the curvature of the meniscus at the assay surface. In one embodiment the surface of the lid and /or projection which interfaces with the assay surface hemispherical.

In one embodiment, the lid and / or projection are optically transparent. Such a lid and or projection has the advantage of being suitable for use with conventional optical detection equipment, and is useful in assays operating via transmitted light (e.g.

absorption, nephelometric and chromogenic spectrophotometric assays etc) and would also overcome the effect of variable lensing due to overall surface tension changes altering meniscus curvature (Cottingham et al (2004) Lab Invest VoI 84 p523- 9).

The assay may be a screening assay, i.e. an assay or method for identifying candidate or test compounds or agents. More preferably the assay is an assay to identify amyloid inhibitors, for example a screening assay to identify amyloid oligomer inhibitors for the treatment of Alzheimer's disease.

An amphiphilic assay component in accordance with the invention, may be a screening candidate or test compound. Such amphiphilic assay candidates can be obtained using any of the numerous approaches in combinatorial library methods well known to a person of skill in the art, such as biological libraries or synthetic libraries. The amphiphilic screening candidate or test compound may be, for example, a peptide non-peptide oligomer or small molecule. The amphiphilic assay component may be an assay substrate or any other component of the assay mixture. The amphiphilic assay component may be biological, biochemical or chemical.

Preferably the amphiphilic component is a biological compound, for example a synthetic or naturally occurring peptide, a synthetic or naturally occurring

phospholipid, a synthetic or naturally occurring fatty acid, a synthetic or naturally occurring cholesterol, a synthetic or naturally occurring glycolipid, a synthetic or naturally occurring steroid, a heterocyclic drug candidate or a derivatised polymer (eg a carbohydrate such as dextran). Alternatively, the amphiphilic component is modified, e.g. post-translationally modified, to contain a hydrophobic acyl moiety (e.g. farnesylation of ras, without which it cannot signal).

More preferably the amphiphilc component is an amyloidogenic substrate, for example an Amyloid β peptide (implicated in Alzheimer's disease), an amyloidogenic fragment of an acetylcholinesterase, an Islet Amyloid Polypeptide (IAPP) or islet amyloid precursor implicated in type Il diabetes.

Examples

Materials and Methods

Synthetic peptides

IAPP (Bachem, Weil am Rhein, Germany) and Aβ_1-40 (EZBiolab, Carmel, IN, USA) were dissolved in DMSO, sonicated and centrifuged for 1 hour at 15,000 g at + 4°C prior to use (to remove any pre-aggregated species). Liposome preparation

Chloroform stocks of dioleoylphosphatidylcholine (DOPC) and

dioleoylphosphatidylglycerol (DOPG) (Avanti Polar Lipids, Alabaster, Alabama, USA) were mixed in a 7:3 molar ratio. After solvent evaporation and desiccation, the lipids were hydrated in 10 mM Na-HEPES pH 7.5, sonicated and extruded through 0.1 μm pore size filter (Avanti Polar Lipids). Potential free remaining lipids were separated from the liposomes using a Sephadex G-25 M (GE Healthcare) desalting column, from which the liposomes were collected in the void volume. The phospholipid

concentration was measured with a phosphate assay kit (BioAssay Systems,

Hayward, CA, USA).

Fibrilization experiments Thioflavin T (ThT) fluorescence in PBS (excitation 450 nm, emission 480 nm) was measured in a 96-well plate (black wall, clear bottom; Greiner Bio-One, Stonehouse, Gloucestershire, UK) without shaking at 37⁰C on a Polarstar plate reader (BMG labtech, Aylesbury, Buckinghamshire, UK). Any compound or physical device (e.g. Perspex cylinder) used was introduced at the same time in test (containing IAPP) and control wells (buffer and ThT without IAPP). Control values (with or without compounds or physical device) were subtracted from test values. At least three independent assays, with every assay done in duplicate wells, were performed and analyzed with the two-sample t-test.

Electron microscopy

4 μM IAPP in fibrilization buffer was incubated for 17 hours (corresponding to plateau in Fig. 1 F) in the presence or absence of Perspex cylinders. Then the samples were adsorbed onto carbon-coated 400 mesh copper grids, air dried, washed with distilled water, negatively stained with 2% aqueous uranyl acetate and viewed with an FEI Tecnai microscope operated at 80keV.

RESULTS

The AWI critically influences the probability and rate of β-sheet amyloid dimer formation (Model 1 )

Amyloidogenic species readily adsorb at HHIs and form β-sheet at such interfaces. Accordingly, a theoretical "toy" model was constructed, by combining a lattice model and reaction kinetic theory, to consider the potential for amyloid surface activity to influence the dynamics of aggregation (8-10). Specifically, the propensity for cross β- sheet dimerization of a short amyloid-forming peptide in the bulk non-surface associated solution and at the AWI was investigated. The peptide was assumed to be about six amino-acid in length (i.e., ~ 2 nm, in length), amyloid forming and having an alternating hydrophobic-hydrophilic side-chain pattern, which was demonstrated to promote amyloid formation (12-14). Due to its shortness, it was assumed that it was rod-like and thus the lattice model presented in Matsuyama and Kato was employed to analyze the adsorption amount, φ_s, at the AWI (15). Specifically, φ_s is the peptides' volume fraction at the surface monolayer. In the present lattice model, the lattice size, a, is set to be 0.5 nm and the aspect ratio, x, of the short peptide wis thus four. For a bulk concentration of 1 μM, the bulk volume fraction, φ, was 3.0 x 10^"7. Following (15), the volume ratio at the AWI, φ_s, is: χ(2φ - l)] , [1]

where δγ = a²(y_l -y_o)/k_BT w\th Y₁ being the surface tension for a pure solute, γ₂ the surface tension for a pure solvent, k_B the Boltzmann constant and 7 is the absolute temperature. The surface pressure induced by the presence of an islet amyloid polypeptide (IAPP) monolayer was shown to be Y₁ -γ₀ > 30 mN/m (6, 8).

In the present example considering a short generic model, it was assumed that Y₁ -Y₀ = 30 mN/m. It was further assumed that the peptide was amphiphilic in such a way that the solvent-solute interaction term, χ , was zero. With these parameters, the adsorption amount at the AWI, φ_s, was calculated to be 4.2 x 10^"4, which would be -1400 higher than the bulk concentration.

Besides an increase in the concentration, the fact that the adsorbed peptides diffuse in two dimensions at the AWI would also increase the collision frequency, and as a result, increase the nucleation rate. Specifically, the average waiting times before collision in 2D and 3D are (16): t_2D = -^- In^- 2D_s ax

U ' ^{[ J}

2Dax_s

where D (D₃) is the bulk (surface) diffusion constant, x is again the aspect ratio, and L_b (L_s ) denotes the average separation between the monomers in the bulk (at the AWI). With the parameters chosen as above, it was estimated that

L_b = α(x/φ)^1/3

~ 43 nm (or 49). Assuming D ~ D₈, the average waiting time seen before a binary collision occurs in the bulk is about 70 times longer than that at the AWI. Another promoting factor due to the presence of the AWI is that the peptides stretched out in parallel to the AWI with the hydrophilic side-chains orientated toward the water while the hydrophobic side-chains orientate toward the air. It is therefore expected that the peptide would expose its inter-peptide hydrogen bonding regions along the AWI and this would facilitate cross /3-sheet formation upon encountering. Assuming that a cross /3-sheet will form if the angle between them is less than 10°, the probability of dimerization in 2D would therefore 10°/180° = 0.056 upon encountering.

In the corresponding situation in the bulk, the probability for the encountering peptides to have the right configuration is estimated by the probability for θ to be less than 10°, multiplied by the probability for the hydrogen bonding side to be along the x-axis (Fig. 5). In other words, the probability on having the right configuration becomes:

So according to the proposed model, the probability of forming β-sheet amyloid dimer would be 14 (= 0.056/0.004) times higher at the AWI than in the bulk.

In summary, assuming that peptide dimerization is diffusion limited according to the scenario considered above, the relative rate of dimerization at the AWI would be about one thousand (= 70x14) times instead of 336 (24x0.056/0.004), i.e., it is more than three hundred times faster than that in the bulk. It is noted that a dimerization event as calculated here does not equate nucleation. But as the nucleation process is expected to be a series of dimerization events until a stable nucleus appears, it is expected that the nucleation will be many orders of magnitude faster with the presence of the AWI. It is thus proposed that below a critical monomer concentration (explained in more detail below in Model 2 and in Fig. 2) and without a HHI, amyloid fibrilization would not be supported, at least not in an experimentally relevant time frame. The AWI critically influences the kinetics of IAPP fibrilization

The above mentioned predictions of Model 1 were confirmed experimentally by determining the kinetic parameters of IAPP (which is involved in DMII) aggregation in presence and absence of an AWI, using staining with a typical amyloid dye that intercalates into stacked β-sheets (ThT) (17). The AWI was chosen because it represents a simple HHI model that is present in almost all current amyloidogenesis in-vitro assays. Moreover, in both IAPP and AD amyloid-β peptide (Aβ), another peptide of similar length also involved in an amyloid-related disease, were shown to form β-sheet at an AWI, with the β-sheets orientated parallel to the interface (6, 10). To remove the AWI, conical Perspex cylinders were introduced into both control and test wells. The introduction of Perspex cylinders into control wells gave the same low invariant ThT background signal as a control well without cylinder. Since IAPP is surface active, it lowers the surface tension of an aqueous solution and triggers meniscus curvature. The hemispherical tip of the cylinder compensates for this curvature and prevents the trapping of an air bubble beneath the cylinder upon insertion. Perspex was chosen because it is very hydrophilic, it interacts more strongly with water than surfactants, and surfactants cannot form a multilayer at Perspex- solution interfaces (18).

When the AWI was removed at the start of a 1.3 μM IAPP reaction, no fibrilization occurred (Fig. 1/A), confirming the Model 1 predictions. Above 1.3 μM IAPP (e.g 1.8 μM), the AWI removal at the start of the reaction allowed fibrilogenesis but

significantly affected nucleation with a lag phase increase (~ 4 fold), and decrease of elongation rate (~ 3 fold) and plateau height (~ 3.5 fold) (Fig. 1 β to £). The AWI requirement for stages other than nucleation was also investigated. The AWI removal just prior to elongation (at the end of the lag phase) of a 1.8 μM IAPP reaction significantly decreased the elongation rate (8.6 fold) and plateau height (5.5 fold) (Fig. 1 D and £). In contrast, the AWI removal during the elongation phase abolished further elongation, with immediate plateau (Fig. 1 B to E, cyl. elong.^*). The lag phases of the respective reactions (AWI removal prior and during elongation) were identical to those with the AWI present (Fig. 1 C). The elongation rate before cylinder introduction was indistinguishable from untreated samples (Fig. 1 D, cyl. elong. versus no cyl.). The AWI removal during plateau did not have any effect (data not shown), indicating that neither the technical process of cylinder insertion nor the properties of the Perspex cylinder itself had any effect on the reaction. At even higher IAPP concentrations (e.g. 4 μM), although the AWI absence did not affect nucleation as evident from the identical lag times for all reactions (Fig. 1 F and G), the elongation rate and plateau height were significantly reduced (Fig. 1 /-/ and /). When the AWI was removed from the start of the reaction, the elongation rate was decreased by 3.8 fold and the plateau height by 9.5 fold. When the AWI was removed just prior to elongation, the elongation rate was decreased by 3.5 fold and the plateau height by 5 fold. To further confirm that the reaction and plateau height observed were the result of IAPP fibrilization, and also to analyze the morphology of the amyloid material formed, we performed electron microscopy on a 4 μM IAPP reaction in the absence or presence of AWI (Fig. 6). After 17 hours incubation, which corresponds to plateau for 4 μM IAPP reaction (see Fig. 1 F), we observed fibrils in both conditions. In the presence of an AWI, some individual fibrils could be observed albeit rarely (Fig. 6A, arrows), but the majority of fibrils were laterally clumping to form a -50 nm width fibril bundle (Fig. 6A). In the absence of AWI, fibrils were readily formed, however they were not associating to form a fibril bundle as observed in the presence of AWI (Fig. 6S).

The initial high ThT fluorescence followed by a rapid drop in signal observed sometimes during the first few minutes of fibrilization experiments, as in Fig. 1>A and F, has been documented previously and suggested to be due to initial binding of ThT to the surface of the multiwell plate (19).

The interface requirement for efficient fibrilogenesis was also confirmed with the AD Aβ_1-40 peptide, for which the AWI removal from the start of a reaction decreased the elongation rate (3.9 fold) and the plateau height (2.2 fold) (p<0.012) (Fig. 9). Amyloid assembly can follow two distinct nucleation pathways (from bulk micelles and from AWI monomers) and is dependent on the critical micellar concentration of the amyloid peptide (Model 2)

It is currently thought that above a critical micellar concentration (CMC),

amyloidogenic monomers can associate to form micelles, and nuclei could originate predominantly from micelles rather than from monomers (the slow process), followed by elongation of the nuclei (the fast process) (20-22). This kinetic theory was generalized to incorporate the surface effect observed experimentally. Specifically, it assumed that besides the nucleation pathway from bulk micelles, nuclei can also form from AWI monomers (Fig. 2A). The current model used a set of ordinary differential equations and simulation parameters, which were either experimental IAPP values when available or of the same order of magnitude as values obtained experimentally for the AD Aβ peptide (Table 1 ). The monomer concentration was denoted by x, the concentration of fibrils of any length by y and the total concentration of monomers locked up in fibrils by z. In other words, the concentration of fibrils consisting of / monomers is denoted by y,, y^≡∑_iy_ι and z≡

■ With these notations, the model introduced by Lomakin et al. could be described by the following set of Ordinary Differential Equations (ODEs) (20, 21 ):

ForC₀ - z > c^* \

x(t) = c

C₀ - c - z(t)

w(t) =

m_n

dyit)

KM,)

ψ = ).

For C₀ - z≤c^*

dx(t) , . , .

dt

ψ dt = -«M<>.

where the constants appearing in the equations are described in Table 1.

It was noted that the process described by the above model 2 dictates that at long time (i.e, as t— >∞ ), all monomers will become fibrilized if the initial concentration, Co, is higher than the CMC, c^* . This prediction is a result of two assumptions: i) fibrils do not break apart, and ii) the fibrils grow indefinitely. These two assumptions are in contradiction to many experimental observations, e.g., see (23) on the Aβ peptides and (24) on IAPP. The experimental data presented herein also demonstrates that not all monomers will become fibrilized, as evident from the fact that the final fibril concentration depended critically on when the AWI was eliminated during the fibrilization process. Therefore to account for this observation, a fibrilization suppression term in a way that the final fibril mass will be constrained was introduced in the following manner:

z_f ≡ z(t→∞) = min( C₀ , Ly(t→∞)) ,

where L is the average number of monomers in a fibril at long time. Physically, it was assumed that the breakage rate of fibrils was negligible, i.e., the first assumption mentioned above which dictates that the fibrils do not break apart was kept; while the second assumption above was modified by constraining the growth of fibril to a maximum size consisting of L monomers. This has the effect that if the number of fibrils is small, then each one will grow to the maximum size with some remaining monomers in solution. On the other hand, if the number of fibrils is large enough, then all monomers will be fibrilized eventually.

A further modification to the original model was the incorporation of the surface effect observed in the experiments. The data clearly suggested that an AWI increases drastically the rate of fibrilization as well as the final fibril mass. To account for these observations, a nucleation pathway from the surface monomers was introduced. Specifically, we assumed that the amount of monomer at the AWI isγr , i.e., it is proportional to the bulk concentration with 0 < γ < 1 (c.f. Eq. [1] in the Model 1 ), and it was assumed that the surface nucleation process is described by first order kinetics, i.e., dy/dt = k_syx . This is a drastic simplification due primarily to the difficulty in modelling the microscopic dynamics of the nucleation process. With these two modifications, the model is described by the following set of ODEs:

ForCn - z > c

x(0

C₀ - c - z(t)

w(t)

m_n

dyit)

k_nw(t) + k_sjx(t)

dt

dz(t)

= nk_nw(t) + nk_syx(t) + ac^*y(t) - b(t)z(t)

dt

where

bit) = iayit) + nk_sy)

z Jt)

For C_n - z≤c^* ) + b(t)z(t)

- b(i)z(i).

The parameters employed in the simulations depicted in Fig. 2 are presented in Table 1.

The modelling results, with different peptide concentrations, are presented in Fig. 2B (-CMC) and 2C (>CMC). Around the CMC, the model predicted that removal of the AWI should result in an absence of fibrilization when done at the start of the reaction, in a slower elongation rate and lower plateau height when done at the end of the lag phase, and in the elongation being abolished when done during the elongation phase (Fig. 2S). Above the CMC, the model predicted that removal of the AWI should result in an unmodified nucleation phase but a slower elongation rate and lower plateau height when done at the start of the reaction or at the end of the lag phase, with the rate and plateau height being higher when the AWI was removed at the end of the lag phase than those when removal occurred at the start of the reaction (Fig. 2C).

Liposomes containing anionic lipids represent a better HHI for IAPP amyloidogenesis than the AWI

Membranes and lipids play a central role in controlling amyloidogenesis, mostly by inducing conformational changes and facilitating fibril growth (4-7). Although in vitro investigations offer insight into the mechanism of lipid-mediated fibrilogenesis, almost all of the assays were done in presence of an AWI. However, the results presented here demonstrate that the AWI plays a critical role in fibrilization. Since the lipid membrane represents the in vivo HHI the experiments herein have sought to determine its role in our AWI-free system. 100 nm diameter liposomes composed of either zwitterionic lipids (DOPC) or a combination of anionic lipids (DOPG) with DOPC were used. For the mixed liposomes, a 7 DOPC:3 DOPG molar ratio was chosen because it resembles the lipid ratio of the pancreatic islet cell membrane (where IAPP amyloidogenesis occurs) and also was shown to be the minimum membrane charge density allowing complete IAPP binding to lipids, under the conditions used by Knight and Miranker (4, 25). In the presence of an AWI, 2.4 μM DOPC/DOPG accelerated nucleation of 1.8 μM IAPP (3.4 fold, p<0.006) and increased the elongation rate (1.9 fold, p<0.004), whereas DOPC liposomes did not have any statistically significant effect (Fig. 3A, B, C and £). Both liposome types significantly reduced the plateau height (p<0.006) (Fig. 3D). Without AWI and lipids, every IAPP kinetic parameter was affected. Although the lag time increase (2.5 fold, p<0.032) and the elongation rate decrease (1.5 fold, p<0.037) were slightly different from those observed in Fig. 1 S to £, the overall outcome and trend were similar.

Importantly in presence of anionic lipids and without AWI, some assembly kinetics were not only enhanced when compared to no AWI-no lipids but also better than those of IAPP with an AWI and without lipids; 3.3 (p<0.006 to IAPP) and 8 fold

(p<0.017 to IAPP without AWI) acceleration in nucleation, and 1.3 (p<0.044 to IAPP) and 2 fold (p<0.0003 to IAPP without AWI) increase in elongation rate (Fig. 3C and £). The plateau height was not significantly different to that without AWI (Fig. 3D). In contrast, zwitterionic lipids did not compensate or rescue the absence of AWI with no statistical difference in lag phase and a 1.6 fold decrease in elongation rate (p<0.02) (Fig. 3C and £).

The role of anionic lipids during IAPP amyloidogenesis, in an in-vivo context (no AWI being present), has been previously underestimated

The enhancing effect of DOPC/DOPG liposomes on IAPP amyloidogenesis (1.8 μM) was at its greatest without AWI and with increased peptide:lipid molar ratios, with the lag time being shorter than those of IAPP with an AWI (p<0.023) and IAPP without an AWI (p<0.017) (Fig. 4A). Strikingly, at the two highest DOPC/DOPG concentrations the lag phase decrease was greater than that obtained with an AWI at the same DOPC/DOPG concentrations (p<0.014). In presence of AWI, increasing the

DOPC/DOPG concentration reduced or abolished the nucleation acceleration seen at lower DOPC/DOPG concentrations. Although increasing DOPC concentrations compensated to some extent the AWI absence (p<0.05), the lag time was never better than that of IAPP with an AWI (Fig. AA).

In presence of an AWI, the elongation rate followed the same trend as the lag phase, with the lowest DOPC/DOPG concentrations accelerating fibrilization (p<0.015) and the highest concentrations slowing it down (Fig. 4S). Without AWI, only 2.4 μM

DOPC/DOPG promoted elongation, with a 1.3 (p<0.008 to IAPP) and 2.5 fold (p<0.03 to IAPP without AWI) increase. None of the DOPC concentrations compensated the AWI absence (p<0.019). However with an AWI, DOPC liposomes accelerated IAPP elongation rate (p<0.035 for 1.7 and 4.2 μM).

All conditions tested led to a lower plateau height than IAPP alone (Fig. 4C).

Demonstration of the effect of air-water interface occlusion on a screening assay for inhibitors of early amyloid oligomerisation

Figure 1 1 shows selected data from a compound screening comparing the effects of candidate small molecules on elongation rate during amyloid fibril formation in the presence or absence of an air-water interface (AWI). The amyloidogenic substrate is 12mM Islet amyloid precursor peptide (IAPP) and the compounds are present at 4mM. Two of the compounds (SYC60 and SYC64) have similar inhibitory effects regardless of the AWI and it is postulated that they represent elongation inhibitors. Other members of this small R group substitution series (SYC61 and SYC63) show enhanced inhibition of nucleated assembly from the bulk (i.e. without AWI). The results demonstrate that one particular candidate, SYC63, is an excellent inhibitor of early stage oligomerisation in the bulk fluid phase, which is a much more physiological situation. It thus represents an excellent "hit" for drug development that otherwise could not have been identified without the AWI occlusion assay. The assay of the present invention advantageously reduces the false positives/negatives in in vitro screening ensuring that the most promising candidate therapeutics may proceed to in vivo testing. The assay also shows its utility as a potential high throughput screen for compounds inhibiting abnormal protein aggregation.

Discussion

By combining a lattice model and reaction kinetic theory (Model 1 in the examples), the inventors predict that at the AWI, compared to the bulk, the surface volume fraction would be -1400 higher, the binary collision would be -70 times faster, the probability of β-sheet amyloid dimer formation would be 14 times higher, and the dimerization rate would be more than 1000 times faster. Since amyloid nucleus formation may be seen as a series of dimerization processes, the nucleation rate would be expected to be much faster at the AWI than in the bulk. Consequently, the model could easily imply that in an experimentally relevant time frame it would be impossible to detect amyloid fibrilization in absence of a HHI and below a critical monomer concentration. The Model 1 predictions predictions have been

experimentally confirmed by demonstrating that the effect of the interface removal (in this case the AWI) was indeed dependent on the initial monomer concentration of amyloidogenic peptide (e.g. IAPP) and that, below a certain initial monomer concentration, fibrilogenesis was not detected (Fig. 1 ).

In our Model 2, current amyloid kinetic theories were modified, which involve the formation of micelles from monomers and the formation of nuclei from these micelles, to incorporate the experimental observations on the role of the AWI during fibril formation. Amyloid assembly kinetics deriving from two different nucleation pathways, one from micelles in the bulk solution and the other one from monomers at the AWI, were considered. Therefore, one important parameter playing a crucial role in the assembly kinetics would be the CMC of the amyloid peptide. The modelling results fit the experimental results with Fig. 2B (-CMC) equivalent to Fig. 1 S, and Fig. 2C (>CMC) to Fig. ~\ F. Indeed according to the model, above the CMC and without an AWI, nucleation could still occur in the bulk solution via micelles (equivalent to Fig. ^*\ F, 4 μM IAPP with no effect on the lag phase). In contrast, below the CMC and without an AWI, nucleation would not happen on a biologically relevant time scale (equivalent to Fig. 1 A 1.3 μM IAPP with no fibrilization). Moreover, the AWI absence from the start of a reaction should only increase nucleation time when IAPP concentration is approaching the CMC (equivalent to Fig. 1 F, 1.8 μM IAPP). Therefore, IAPP CMC lies between 1.3 and 1.8 μM.

Following the demonstration on the critical importance of the presence of an AWI on all kinetic parameters of IAPP fibrilogenesis, the effect of lipid membranes on these parameters in were assessed in an AWI-free system. Firstly, the experimental system was validated by confirming previous findings on the enhancing role of anionic lipids and the absence of effect of zwitterionic lipids on amyloid nucleation in presence of an AWI (Fig. 3) (4, 5, 9). Moreover, the same studies showed that above a critical amyloid:lipid ratio in the presence of an AWI, fibrilization was no longer accelerated but instead slowed, which is exactly what was observed at higher IAPP:DOPG ratios (Fig. 4) (4, 5). It is accordingly proposed that in presence of AWI and anionic lipids, IAPP accumulate at both interfaces. The increased lipid concentration also result in fewer adsorbed IAPP molecules per liposome. Indeed, investigations showed that at high lipid concentrations no IAPP high oligomeric species were bound to the membrane as observed at lower concentrations (4, 5).

The enhancing effect of DOPC/DOPG liposomes on IAPP amyloidogenesis was demonstrated to be at its greatest in a context more closely mirroring in-vivo conditions (namely in the absence of AWI). Indeed, at the two highest DOPC/DOPG concentrations, IAPP lag phase decrease was greater than that obtained with an AWI at the same DOPC/DOPG concentrations. Thus, the results clearly demonstrate that to understand the full extent of fibrilization enhancement by phospholipid bilayers, the AWI should be removed.

In the absence of AWI, it is believed that IAPP would adsorb and nucleate only at the lipid surface and would do so much faster than in the presence of a competing AWI. Although increasing DOPC concentrations was shown to compensate to some extent the AWI absence, the lag time was never better than that of IAPP with an AWI. With DOPC/DOPG liposomes, electrostatic interactions would lead to both IAPP adsorption to and IAPP cationic N-terminus insertion into the headgroup region, and penetration into the tail groups might also occur (5, 30). In contrast, with DOPC liposomes IAPP positive charges would be left unshielded and no insertion would occur, as was previously observed with the AD Aβ_1-40 peptide (9). Thus, IAPP would accumulate at the AWI rather than at the DOPC liposome surface. Without AWI, DOPC liposomes would provide a surface for IAPP monomers that, even if poor, would be better than remaining in the bulk solution. In presence of an AWI, some concentrations of DOPC liposomes accelerated IAPP elongation rate. It was postulated that β-sheet-rich IAPP oligomers could interact with zwitterionic membranes (5). The AWI would have facilitated oligomer formation.

Having shielded their charges, the oligomers could adsorb onto and be stabilized by zwitterionic lipids, which would facilitate further elongation.

All conditions tested led to a lower plateau height than IAPP alone (Fig. 4C), which may indicate that curved lipid-water interfaces (as found with liposomes) do not support long assemblies. Amyloid aggregation on membranes results in distortion or complete disruption of the lipid bilayer (1-3). Such lipid perturbation may not permit propagation of pre-aggregated species by not allowing charge shielding or by not retaining aggregated species. The present model predicted that without AWI nucleation would occur only through micelles (Fig. 2A), which could lead to fewer nuclei and therefore fewer fibrils formed. Alternatively, ThT binding could be weaker when IAPP is adsorbed on lipid surfaces.

The general applicability of the described assay modification is illustrated in figures 9 and 10, which shows the effect of AWI removal on two other polymer assembly systems. In figure 9A-C the substrate is a synthetic Aβ_1-40 peptide relevant to

Alzheimer's disease, while in figures 1 OA and B the substrate is an amyloidogenic fourteen residue peptide derived from the T40 tetramerisation domain of human neuronal form acetylcholinesterase. In both cases the absence of an AWI has a profound and highly significant effect on assembly behaviour.

Table 1. The various parameters employed in the simulations for Model 2 (third columns). The estimated values for the case of A/3 peptides are shown in the fourth column. Note that values with citations are taken from measurements in the literature. Table 1

REFERENCES

1. Kayed, R., Sokolov, Y., Edmonds, B., Mclntire, T. M., Milton, S. C, Hall, J. E., and Glabe, C. G. (2004) Permeabilization of lipid bilayers is a common conformation-dependent activity of soluble amyloid oligomers in protein misfolding diseases. J Biol Chem 279, 46363-46366

2. Quist, A., Doudevski, I., Lin, H., Azimova, R., Ng, D., Frangione, B., Kagan, B., Ghiso, J., and LaI, R. (2005) Amyloid ion channels: a common structural link for protein-misfolding disease. Proc Natl Acad Sci U S A 102, 10427-10432

3. Demuro, A., Mina, E., Kayed, R., Milton, S. C, Parker, I., and Glabe, C. G.

(2005) Calcium dysregulation and membrane disruption as a ubiquitous neurotoxic mechanism of soluble amyloid oligomers. J Biol Chem 280, 17294- 17300

4. Knight, J. D., and Miranker, A. D. (2004) Phospholipid catalysis of diabetic amyloid assembly. J MoI Biol 341 , 1 175-1 187 5. Knight, J. D., Hebda, J. A., and Miranker, A. D. (2006) Conserved and cooperative assembly of membrane-bound alpha-helical states of islet amyloid polypeptide. Biochemistry 45, 9496-9508

6. Lopes, D. H., Meister, A., Gohlke, A., Hauser, A., Blume, A., and Winter, R.

(2007) Mechanism of islet amyloid polypeptide fibrillation at lipid interfaces studied by infrared reflection absorption spectroscopy. Biophys J 93, 3132- 3141

7. Terzi, E., Holzemann, G., and Seelig, J. (1997) Interaction of Alzheimer beta- amyloid peptide(1-40) with lipid membranes. Biochemistry 36, 14845-14852 8. Soreghan, B., Kosmoski, J., and Glabe, C. (1994) Surfactant properties of

Alzheimer's A beta peptides and the mechanism of amyloid aggregation. J Biol Chem 269, 28551-28554

9. Chi, E. Y., Ege, C, Winans, A., Majewski, J., Wu, G., Kjaer, K., and Lee, K. Y.

(2008) Lipid membrane templates the ordering and induces the fibrillogenesis of Alzheimer's disease amyloid-beta peptide. Proteins 72, 1-24

10. Maltseva, E., Kerth, A., Blume, A., Mohwald, H., and Brezesinski, G. (2005) Adsorption of amyloid beta (1-40) peptide at phospholipid monolayers.

Chembiochem 6, 1817-1824

1 1. Ruschak, A. M., and Miranker, A. D. (2007) Fiber-dependent amyloid formation as catalysis of an existing reaction pathway. Proc Natl Acad Sci U S A 104,

12341-12346

12. Yoon, S., and Welsh, W. J. (2004) Detecting hidden sequence propensity for amyloid fibril formation. Protein Sci 13, 2149-2160

13. Fernandez-Escamilla, A. M., Rousseau, F., Schymkowitz, J., and Serrano, L.

(2004) Prediction of sequence-dependent and mutational effects on the aggregation of peptides and proteins. Nat Biotechnol 22, 1302-1306

14. DuBay, K. F., Pawar, A. P., Chiti, F., Zurdo, J., Dobson, C. M., and

Vendruscolo, M. (2004) Prediction of the absolute aggregation rates of amyloidogenic polypeptide chains. J MoI Biol 341 , 1317-1326

15. Matsumaya, A., and Kato, T. (1995) Theory of Surface Ordering on Solutions of Rigid-Rod-Like Molecules. Macromol. 28, 131-135

16. Jackson, M. B. (2006) Molecular and Cellular Biophysics, University of

Cambridge Press, Cambridge 17. LeVine, H., 3rd (1993) Thioflavine T interaction with synthetic Alzheimer's disease beta-amyloid peptides: detection of amyloid aggregation in solution. Protein Sci 2, 404-410

18. Szymczyk, K., Zdziennicka, A., Janczuk, B., and Wojcik, W. (2006) The

wettability of polytetrafluoroethylene and polymethyl methacrylate by aqueous solution of two cationic surfactants mixture. J Colloid Interface Sci 293, 172- 180

19. Engel, M. F., Khemtemourian, L., Kleijer, C. C, Meeldijk, H. J., Jacobs, J., Verkleij, A. J., de Kruijff, B., Killian, J. A., and Hoppener, J. W. (2008)

Membrane damage by human islet amyloid polypeptide through fibril growth at the membrane. Proc Natl Acad Sci U S A 105, 6033-6038

20. Lomakin, A., Chung, D. S., Benedek, G. B., Kirschner, D. A., and Teplow, D.

B. (1996) On the nucleation and growth of amyloid beta-protein fibrils:

detection of nuclei and quantitation of rate constants. Proc Natl Acad Sci U S A 93, 1 125-1 129

21. Lomakin, A., Teplow, D. B., Kirschner, D. A., and Benedek, G. B. (1997)

Kinetic theory of fibrillogenesis of amyloid beta-protein. Proc Natl Acad Sci U S A 94, 7942-7947

22. Sabate, R., and Estelrich, J. (2005) Evidence of the existence of micelles in the fibrillogenesis of beta-amyloid peptide. J Phys Chem B 109, 1 1027-1 1032

23. O'Nuallain, B., Shivaprasad, S., Kheterpal, I., and Wetzel, R. (2005)

Thermodynamics of A beta(1-40) amyloid fibril elongation. Biochemistry 44, 12709-12718

24. Larson, J. L., Ko, E., and Miranker, A. D. (2000) Direct measurement of islet amyloid polypeptide fibrillogenesis by mass spectrometry. Protein Sci 9, 427-

431

25. Rustenbeck, I., Matthies, A., and Lenzen, S. (1994) Lipid composition of

glucose-stimulated pancreatic islets and insulin-secreting tumor cells. Lipids 29, 685-692

26. Rochet, J. C, and Lansbury, P. T., Jr. (2000) Amyloid fibrillogenesis: themes and variations. Curr Opin Struct Biol 10, 60-68

27. Harper, J. D., and Lansbury, P. T., Jr. (1997) Models of amyloid seeding in Alzheimer's disease and scrapie: mechanistic truths and physiological consequences of the time-dependent solubility of amyloid proteins. Annu Rev Biochem 66, 385-407

28. Sipe, J. D., and Cohen, A. S. (2000) Review: history of the amyloid fibril. J

Struct Biol 130, 88-98

29. Klunk, W. E., Pettegrew, J. W., and Abraham, D. J. (1989) Quantitative

evaluation of congo red binding to amyloid-like proteins with a beta-pleated sheet conformation. J Histochem Cytochem 37, 1273-1281

30. Engel, M. F., Yigittop, H., Elgersma, R. C, Rijkers, D. T., Liskamp, R. M., de

Kruijff, B., Hoppener, J. W., and Antoinette Killian, J. (2006) Islet amyloid polypeptide inserts into phospholipid monolayers as monomer. J MoI Biol 356,

783-789

Throughout the description and claims of this specification, the words "comprise" and "contain" and variations of them mean "including but not limited to", and they are not intended to (and do not) exclude other moieties, additives, components, integers or steps. Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.

Features, integers, characteristics, compounds, chemical moieties or groups described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith. All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and/or steps are mutually exclusive. The invention is not restricted to the details of any foregoing embodiments. The invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including any

accompanying claims, abstract and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.

The reader's attention is directed to all papers and documents which are filed concurrently with or previous to this specification in connection with this application and which are open to public inspection with this specification, and the contents of all such papers and documents are incorporated herein by reference.

Claims

1. An in vitro assay comprising at least one amphiphilic assay component in an assay sample, characterized in that said assay comprises a step of occluding an air water interface at a surface of the assay sample, whereby amphiphilic organisation at said assay sample surface is inhibited.

2. An in vitro assay according to claim 1 , wherein said amphiphilic organisation is amphiphilic concentration, selective orientation, polymerisation or aggregation.

3. An in vitro assay according to claim 1 or claim 2, wherein said assay is a screening assay.

4. An in vitro assay according to claim 3, wherein said at least one amphiphilic assay component is a screening candidate.

5. An in vitro assay according to claim 4 wherein the amphiphilic assay component is a candidate amyloid accumulation inhibitor or anti-amyloidogenic agent.

6. An in vitro assay according to claim any one of claims 1 to 3, wherein said at least one amphiphilic assay component is an assay substrate.

7. An in vitro assay according to any one of claims 1 to 3, wherein said at least one amphiphilic assay component is an enzymatic target of an assay.

8. An in vitro assay according to any preceding claim, wherein said assay is performed in an assay vessel.

9. An in vitro assay according to claim 8, wherein said vessel is a test tube, a dish, a flask or a multiwall plate.

10. An in vitro assay according to claim 8 or 9, wherein said step of occluding an air water interface is achieved by contacting the assay surface with a hydrophilic lid.

1 1. An in vitro assay according to claim 10, wherein said hydrophilic lid comprises a projection dimensioned such that when the projection is contacted with the assay surface, the surface interfaces only with the projection.

12. An in vitro assay according to claim 1 1 , wherein said hydrophilic lid is formed from hydrophilic plastics material.

13. An in vitro assay according to claim 1 1 , wherein said hydrophilic plastics material is polymethyl methacrylate.

14. A closed assay system for conducting an in vitro assay comprising at least one amphiphilic assay component, wherein said system is provided with a means to occlude air from the assay system thereby preventing formation of an air water interface within said system during performance of an assay therein.

15. A closed assay system according to claim 14, wherein said closed assay system is a microfluidic assay system.

16. A closed assay system according to claim 15, wherein said means to occlude air from the system is a hydrophilic body.

17. A closed assay system according to claim 16, wherein said hydrophilic body is formed from hydrophilic plastics material.

18. A closed assay system according to claim 17, wherein said hydrophilic plastics material is polymethyl methacrylate.

19. A closed assay system assay according to any one of claims 14 to 18, wherein said at least one amphiphilic assay component is a screening candidate.

20. A closed assay system assay according to claim 19 wherein the amphiphilic assay component is a candidate amyloid accumulation inhibitor or anti-amyloidogenic agent

21. A closed assay system assay according to any one of claims 14 to18, wherein said at least one amphiphilic assay component is an assay substrate.

22. A closed assay system assay according to any one of claims 14 to18, wherein said at least one amphiphilic assay component is an enzymatic target of an assay.

23. An assay apparatus comprising at least one well defined by at least one wall and a well lid characterized in that the well lid comprises at least one hydrophilic projection

complementarily shaped to fit within the at least one well and dimensioned such that when the well contains a predetermined volume of assay sample and the projection is inserted into the well, the assay sample interfaces only with the at least one wall and the projection.

24. An assay apparatus according to claim 23, wherein the apparatus further comprises a base upon which said at least one wall defines a well.

25. An assay apparatus according to claim 24, wherein the apparatus comprises a plurality of walls each defining a well on said base and said lid comprises a plurality of hydrophilic projections.

26. An assay apparatus according to claim 25, wherein said plurality of projections on said lid is equal to said plurality of walls each defining a well.

27. An assay apparatus according to claim 25 or 26, wherein said apparatus comprises 96 walls each defining a well.

28. The assay apparatus according to any one of claims 25 to 27, further comprising an overflow cavity between each continuous wall defining a well.

29. The assay apparatus according to any one of claims 23 to 28, wherein said hydrophilic projection(s) is/are formed from hydrophilic plastics material.

30. The assay apparatus according to claim 29, wherein said hydrophilic plastics material is polymethyl methacrylate.

31. The assay apparatus according to any one of claims 23 to 30, wherein said base and walls are formed from metal, glass, plastics material, or ceramic.

32. The assay apparatus according to any one of claims 23 to 31 , wherein said assay sample comprises at least one amphiphilic assay component.

33. The assay apparatus according to any one of claims 23 to 32, wherein said at least one continuous wall defines a well of circular configuration. 34. The assay apparatus according to any one of claims 23 to 33, wherein said lid and / or base are optically transparent.

35. Use of the closed assay system according to any one of claims 14 to 22 or the assay apparatus according to any one of claims 23 to 34 in an assay comprising an amphiphilic component.

34. An apparatus configured to carry out the assay according to any one of claims 1 to 13.

35. A method of screening amphiphilic candidate components, the method comprising placing an assay sample in an assay apparatus according any one of claims 23 to 34 and occluding an air water interface by contacting the assay sample surface with a hydrophilic lid so that the assay sample interfaces only with the assay apparatus.

36. An assay substantially as described herein with reference to the accompanying drawings.

37. An assay apparatus substantially as described herein with reference to the

accompanying drawings.