Hydropathy: the controlling factor behind the inhibition of Aβ fibrillation by graphene oxide

Abstract

Protein and peptide aggregation/fibrillation is reported to be responsible for several neurological disorders. Fibrillation of the amyloid β-peptide fragment (25–35) which is a biologically active region of the full length peptide, has been observed to be significantly inhibited in presence of the two dimensional nanomaterial graphene oxide (GO). Fibrillation and inhibition of the Aβ_25–35 peptide by GO has been performed at 37 °C at physiological pH (pH 7.4). The inhibition process is monitored by ThioflavinT fluorescence (ThT), circular dichroism spectroscopy, matrix assisted laser desorption/ionization mass spectrometry, dynamic light scattering experiments etc. The soluble fraction of the protein is quantified by the BCA assay. Microscopic techniques are used to study the morphology of the fibrils formed. GO is observed to inhibit the fibrillation even at very low concentrations and is amplified with increase in concentration of GO. ThT kinetic data fitted well with a sigmoidal curve and shows that GO is able to lengthen the lag phase of the fibrillation process. It appears that surface adsorption of protein on the nanomaterial prevents the monomers to come together. It is speculated that the presence of both polar and non-polar moieties in GO interact strongly with the hydrophobic and hydrophilic residues of the Aβ_25–35 peptide monomer units, thus preventing further aggregation.

---

Introduction

Insoluble filamentous aggregates, a hallmark of neurodegenerative pathogenesis, rich in predominantly 39–42 residue long amyloid beta (Aβ) peptides, are found to play a major role in neurotoxicity which leads to oxidative stress and finally cell death.¹ Excess accumulation of Aβ peptides in the brain causes aggregation of peptides producing amyloid fibrils possessing specific crossed β-sheet structures.^2–4 Sequential proteolysis of the membrane bound amyloid precursor protein at its β- and γ-secretase sites generates the ∼4.2 kDa Aβ peptides along with smaller, often chemically modified amyloidogenic fragments Aβ_25–35, Aβ_3–40, Aβ_17–42 which not only retain the cytotoxic character of the full length peptide but also have the ability to assemble rapidly.^5–9 Scientists have investigated the pathogenesis of the full length Aβ peptides in vivo and in vitro but the behaviour of smaller fragments has not been explored in a detailed manner.^10,11

The undecapeptide Aβ fragment (Aβ_25–35) of the full length Aβ peptide, with the amino acid sequence, GSNKGAIIGLM is reported to originate in vivo by cerebral proteases^12–14 and represents the biologically active region of the full length Aβ peptide.¹⁵ Aβ_25–35 is linked to the accumulation of reactive oxygen species in hippocampal neuronal cultures which in turn produce toxic effects.¹⁶ Varadarajan et al. has reported that Met-35 plays a significant role in fibril formation and its toxicity.¹⁷ An investigation on the fibrillation potency of the Aβ_25–35 peptide in presence of additives is thus of importance. Vitiello et al. proposed that omega-3 fatty acids restrict the fibrillation of the Aβ_25–35 peptide in presence of model lipid membranes.¹⁸ Acetylation of Lys-28 is found to not only change the morphology of the Aβ_25–35 fibrils but also regulate the unzipping force of fibrils.¹⁹ Effects of the lipid membrane formed by 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho-L-serine (POPS) on the in vitro fibrillation of Aβ_25–35 have been studied using spectroscopic approaches.²⁰ Recently, we have shown that interactions of preformed Aβ_25–35 fibrils with surfactant micelles of different charges are predominated by hydrophobic interactions over electrostatic interactions.²¹ Our research group has also reported that Aβ_25–35 fibrils form different patterns upon drying on substrates of different wettability and the physics of the forces behind such patterning has been investigated.²²

Nanomaterials in various forms have emerged into a broad area of research in the field of biomedicine including cancer therapy, cell imaging, drug delivery etc. because of their efficiency to cross the blood–brain barrier.^23,24 Nanoparticles have also exhibited their potency both in the inhibition and stimulation of amyloid fibril formation.^25,26 Recent reports mention that carbon nanomaterials like fullerene C₆₀ and carbon dots retard the fibrillation process in vitro.^27,28 Adsorption of the Aβ peptide on a graphite surface has also been studied previously.^29–31 Graphene is a planar, two dimensional single atom thick carbon nanomaterial, formed solely of sp² carbon atoms in a honeycomb lattice array arrangement with excellent electronic, magnetic, mechanical and optical properties.^32–35 However, the application of graphene in biological systems becomes limited because of its hydrophobic nature and poor solubility in water which is normally overcome by functionalising the graphene surface with hydrophilic groups.^36–38 Graphene oxide (GO) contains a hydrophobic basal plane and is additionally decorated with hydrophilic groups which can ideally provide a platform for studying the amyloid fibrillation process.³⁹ Recent progress in studies relating to the inhibition of amyloid fibrillation for the^33–42 fragment showed that adsorption of the peptide monomers on the GO surface leads to inhibition of the fibrillation process.⁴⁰ A combination of experimental and theoretical studies has reported that the GO based protein corona, graphene quantum dots, pristine graphene can inhibit amyloid fibrillation.^41–44 In some cases, the Aβ peptide is shown to aggregate rapidly on the graphene surface.^45,46 A study by Qing et al. showed that even surface chirality influences amyloid fibrillation of the Aβ_1–40 peptide using stereoselectively modified GO surface as the potential platform.⁴⁷

To the best of our knowledge, the effect of GO on the Aβ_25–35 peptide and its aggregation has not been examined so far. In the present article, we show that GO inhibits fibrillation of the Aβ_25–35 peptide using several biophysical and spectroscopic measurements. Factors modulating the fibrillation of this peptide include hydrophobic factors with a role of the functional groups of GO. Graphene oxide was synthesised and characterised by several spectroscopic and microscopic techniques. The interaction of the Aβ_25–35 peptide fragment with GO was studied by the Thioflavin T (ThT) fluorescence assay and matrix assisted laser desorption/ionisation mass spectrometry (MALDI), dynamic light scattering (DLS) experiments. The amounts of soluble protein in each set was quantified by the UV-visible spectroscopy based bicinchoninic acid (BCA) assay. Circular dichroism (CD) and Fourier transform infrared (FTIR) spectroscopy are used to monitor changes in the secondary structural content of proteins in the inhibition process. Several microscopic techniques were also used for detailed morphological analyses of the GO treated and untreated samples. Interactions of GO with Aβ peptide fragments were further analysed by hydrophobicity calculations. Our findings show that GO is able to effectively prevent the fibrillation of the Aβ_25–35 peptide under physiological conditions in vitro which can provide a promising application of carbon nanomaterials in AD.

Results and discussion

The amyloid β-peptide (25–35) fragment depicted as Aβ_25–35 is found to be present in neuronal cells as well as in inclusion-body myositis muscle.⁴⁸ This 11-residue long peptide coalesces in a rapid manner to produce fibres and imparts cytotoxic effects on antioxidant enzymes.⁴⁸ Targeting the aggregation behaviour of the Aβ_25–35 peptide and its possible clearance provides a promising way to understand the fibrillation process and methods or materials that might inhibit the process. Graphene oxide, a biocompatible carbon nanomaterial exhibits its distinctive properties in many biological applications.^49–52 This study aims to investigate the effects of GO on the aggregation propensity of the Aβ_25–35 peptide in vitro using several spectroscopic and microscopic techniques.

Characterisation of prepared graphene oxide

Primary characterisation of the prepared graphene oxide was achieved with the aid of UV-vis spectroscopy and FTIR spectroscopy. Results of the characterisation studies have been provided in the ESI.† Fig. S1† shows the absorption spectrum of GO in water. The strong absorption maxima at 232 nm and the broad peak at 301 nm appear due to (π–π*) transitions of (C=C) and (n–π*) transitions of (C=O) respectively which confirms the presence of such moieties.⁵³ Structural aspects of GO are further confirmed from the FTIR spectrum of GO as represented in Fig. S2.† The representative peaks at 3409 cm⁻¹, 1720 cm⁻¹, 1226 cm⁻¹ and 1060 cm⁻¹ are assigned to the following vibrations: (O–H) symmetric stretching, (C=O) stretching of (–COOH) group and stretching vibration of epoxy and alkoxy group respectively. The peak situated at 1618 cm⁻¹ is attributed to the unoxidised (C=C) group of the graphitic domain.⁵⁴ The XRD pattern of GO reveals that the strong peak of the (002) plane at 2θ = 11.0° appears due to an interlayer separation of 0.8 nm. It confirms the oxidation of graphite to GO [Fig. S3(a)†]. The morphology of GO in solution was determined from an AFM topographic image [Fig. S3(b)†]. GO sheets are found to be present on the mica surface and a line scan provides the height profile of the GO sheet. The height of the GO sheet shows a thickness of ∼0.9 nm confirming a GO sheet monolayer.

Inhibition of fibrillation: ThT fluorescence

The ThT dye is a universally used potential β-sheet marker which specifically recognises crossed β-sheet fibrillar architecture because of its proper fitting into the channels of sheets orienting the shorter axis of the molecule perpendicular to the fibrillar axis.^55,56 The structure of Thioflavin T is shown in Scheme 1. ThT not only detects and stains amyloid fibrils but also provides a quantitative estimation of fibril formation.⁵⁷ Binding of ThT with these structures shifts its absorption maxima from 412 nm to 450 nm with substantial increase in fluorescence intensity at ∼485 nm.⁵⁷

Scheme 1 Structure of Thioflavin T.

Excitation spectra given in Fig. 1(a) of ThT and ThT bound Aβ_25–35 (both in presence and absence of GO) shows that ThT exhibits an excitation peak at ∼412 nm that is red shifted to ∼440–445 nm for the ThT bound to the Aβ_25–35 peptide incubated in 50 mM phosphate buffer (pH 7.4) at 37 °C for 6 days. This observation confirms the formation of amyloid fibrils [Fig. 1(a)]. Similar results have also been obtained for amyloid fibrils of other proteins and peptides.^57,58 The Aβ_25–35 peptide incubated with 100 μg ml⁻¹ GO under the same conditions as mentioned above did not bind with ThT as is evident from its excitation maxima [Fig. 1(a)]. It was observed that ThT incubated with Aβ_25–35 peptide in presence of GO shows an excitation maxima at ∼412 nm, indicating the presence of only free dye and absence of amyloid fibrils.

Fig. 1 (a) Excitation spectra (λ_em = 485 nm) (b) fluorescence emission spectra (λ_ex = 450 nm) of ThT, ThT with Aβ_25–35 peptide (incubated for 6 days) for and ThT with Aβ_25–35 peptide containing GO (incubated for 6 days).

In the present study, the formation of amyloid fibrils and their inhibition by GO has been monitored by ThT fluorescence measurements. The fluorescence intensity of ThT at ∼485 nm upon excitation at 450 nm is observed to increase sharply after binding with incubated Aβ_25–35 peptide, thus indicating formation of amyloid fibrillar species. The fibrillation process continued for 6 days till complete aggregation was achieved. Aβ_25–35 peptide when incubated with 100 μg ml⁻¹ GO in 50 mM phosphate buffer (pH 7.4) at 37 °C for 6 days shows a marked drop in ThT fluorescence showing almost complete inhibition [Fig. 1(b)]. The emission intensity of ThT in presence of only graphene oxide in buffer (without peptide) upon excitation under the same conditions was acquired to check the interaction of GO with ThT. It was found that the spectral characteristics of ThT did not change indicating that there was no specific interaction of the dye with GO. A dose dependent study (GO concentrations: 25, 50, 100 μg ml⁻¹) on the fibrillation process by GO shows that it is able to prevent peptide aggregation in a concentration dependent manner. The lowest chosen concentration of GO (25 μg ml⁻¹) also exhibits efficient inhibitory propensity [Fig. 2] as it reduces the ThT fluorescence intensity by 50% after incubation for 6 days.

Fig. 2 Histogram of ThT fluorescence intensity of Aβ_25–35 peptide with GO (concentrations: 0, 25, 50, 100 μg ml⁻¹) at 485 nm.

Further increase in the inhibitory efficacy is obtained with increased concentrations of GO. The bare peptide (without GO) coalesces very rapidly and reaches the steady state within 8 h. No further increment of ThT fluorescence intensity was detected with time. Monitoring the ThT fluorescence intensity can thus selectively detect the extent of fibrils present. In general, the fibrillation of amyloid β-peptides is initiated by oligomerisation of the monomer units of peptides to form nuclei referred to as the nucleation phase. This is followed by the elongation phase where the small clusters or protofibrils undergo rapid aggregation to form fibrils. Further fibrillation is stopped when almost all peptide units are aggregated. This step gives rise to the plateau region of the sigmoidal curve and comprises the stationary phase. The lag time that is the time needed to form the critical nuclei for further fibrillation/aggregation to occur depends on several factors. These include most importantly the fibrillation conditions and the aggregation propensity of the peptide, a property primarily guided by the sequence of amino acids. The kinetic data obtained were fitted to a sigmoidal equation (eqn (1)) which describes the fibrillation pathway including the initial nucleation or lag phase, sequentially followed by a growth phase and then the plateau region indicating equilibration.^59–61

Here, y represents the ThT fluorescence at time t, y₀ and y_max are the initial and maximum ThT fluorescence intensities, t_1/2 represents the time required to reach half the maximum fluorescence intensity and k is the apparent first order aggregation constant. The lag time of the fibrillation process was calculated as:

The parameters from the fitting curve were extracted and analysed. The peptides containing GO (of different concentrations) also reach the maxima at ∼8 h but interestingly the initial nucleation period of fibrillation is found to increase with increase in concentration of GO [Fig. 3]. In the absence of GO, there is almost no lag phase demonstrating the high propensity of the peptide towards aggregation. The lag phases and t_1/2 values observed during aggregation in presence of the different concentrations of GO are tabulated in Table 1. Interactions of the peptide monomers with GO prevents the peptides to come in close proximity of each other and thus is able to prevent further clustering resulting in a lower population of fibrils.

Fig. 3 Kinetics of fibrillation and inhibition of Aβ_25–35 peptide by GO monitored by typical ThT fluorescence: ThT fluorescence kinetics in presence of 0, 25, 50 and 100 μg ml⁻¹ GO fitted in the sigmoidal equation. Inset: Initial lag phases of the kinetic measurements obtained from parameters of fitted curves.

Table 1 The lag phases and t_1/2 values, obtained from the sigmoidal curve fitting of the kinetic data of the fibrillation process in presence of different concentrations of GO

Set composition	t1/2 (h)	Lag phase (h)
Aβ25–35	3.04	∼0
Aβ25–35 + 25 μg ml^−1 GO	3.96	2.35
Aβ25–35 + 50 μg ml^−1 GO	4.29	3.97
Aβ25–35 + 100 μg ml^−1 GO	5.49	4.24

---

Circular dichroism spectroscopy and Fourier transform infrared spectroscopy

Secondary structural changes of the Aβ_25–35 peptide alone and in presence of the highest concentration of GO (100 μg ml⁻¹) upon incubation at 37 °C for 6 days were examined from CD spectra. Fig. 4 shows the CD spectra of the peptide and the peptide treated with GO. The incubated peptide shows a broad minima at ∼220 nm, characteristic of β-sheet rich species. Presence of GO decreases the CD signal to a less negative value and the minimum near 220 nm is less apparent. Secondary structural contributions have also been calculated by the DICHROWEB server⁶² and are tabulated in Table 2. The amount of β-strands decrease from 53% to 46% due to incubation of peptide in presence of GO. The results suggest that the presence of GO restricts the peptide from forming ordered β-strands but instead induces formation of unstructured elements, turns etc.

Fig. 4 Circular dichroism spectra of Aβ_25–35 peptide incubated in absence and presence of 100 μg ml⁻¹ GO.

Table 2 Secondary structural contents of Aβ_25–35 peptide incubated at 37 °C for 6 days in absence and presence of GO. Parameters have been obtained from the structural analysis component of the online DICHROWEB server⁶²

Set composition	% β-sheet	% turns	% unordered
Aβ25–35	53.3	0.0	17.5
Aβ25–35 + 100 μg ml^−1 GO	45.9	14.2	22.1

---

FTIR spectroscopy is extensively used for determination of secondary structural elements of proteins and peptides including the Aβ_25–35 peptide.^63–65 The amide I regions of the spectra have been extracted and analysed. Fig. 5(a) and (b) show the FTIR spectra of Aβ_25-35 peptide in buffer and in presence of 100 μg ml⁻¹ GO respectively. The Aβ_25–35 peptide after 6 days of incubation at 37 °C displays multiple peaks in the amide I region [Fig. 5(a)]. The region exhibiting a small hump at 1630 cm⁻¹ and a sharp peak at 1639 cm⁻¹ originates due to the presence of parallel β-sheets whereas the hump at 1687 cm⁻¹ shows the presence of antiparallel β-sheets. The appearance of these peaks confirms the presence of β-sheet rich species, thus supporting fibril formation.^64,65 The weak peaks at 1650 cm⁻¹ and 1663 cm⁻¹ arise due to the contribution of α-helices and turns.^64,65 Fig. 5(b) shows the amide I region of Aβ_25–35 peptide in presence of GO. The main peaks are found to be shifted to 1646 cm⁻¹ and 1668 cm⁻¹, an overlapping region of random coils and turns. The absence of a peak at ∼1687 cm⁻¹ indicates the loss of β-sheet content. These results are in agreement with the findings from CD spectroscopy.

Fig. 5 FTIR spectra of (a) Aβ_25–35 peptide incubated in buffer only (b) Aβ_25–35 peptide incubated in presence of GO.

Quantification of proteins in soluble fraction by BCA assay

The BCA assay for quantification of the soluble protein fraction is based on the principle that peptide bonds present in proteins reduce the Cu²⁺ present in the assay solution and the BCA forms a purple-coloured water soluble complex (λ_max = 562 nm) with Cu⁺ in an alkaline medium.^66,67 The extent of formation of the complex is thus linearly proportional to the amount of peptide present in the system. Fibrils present in the incubated sets were pelleted down by centrifugation and the supernatant collected to quantify the amount of proteins present in the soluble part of the incubated samples. The concentration of peptides in the soluble fraction of each set was measured from a calibration curve and is shown in Fig. 6. The concentration of peptide in the set containing peptide only is 83.7 ± 13.9 μg ml⁻¹ whereas with 25, 50 and 100 μg ml⁻¹ GO, there is an increase in the concentration of soluble protein content to 142.1 ± 27.9, 191.5 ± 47.8 and 270.7 ± 70.4 μg ml⁻¹ respectively. These results also indicate that GO inhibits the formation of fibrils in a concentration dependent manner. The findings are in good agreement with the results from the ThT fluorescence and secondary structure analyses.

Fig. 6 Quantification of soluble fraction of proteins by BCA assay. The histogram shows increasing concentrations of protein in soluble fraction of sets with increase in concentration of GO.

Dynamic light scattering study

The size of the Aβ_25–35 peptide monomer was found to be ∼68 nm. Earlier studies report that HFIP treated Aβ_1–42 peptide shows a size range of (3–100) nm (ref. 68) [Fig. 7(a)]. Incubation of Aβ_25–35 peptide at physiological temperature for several days leads to unfolding and aggregation to form amyloid fibrils of larger sizes. DLS experiments show that the Aβ_25–35 peptide (incubated in 50 mM phosphate buffer (pH 7.4) at 37 °C for 6 days) contains particles of size ∼680 nm and ∼5560 nm [Fig. 7(b)]. Addition of 25, 50 and 100 μg ml⁻¹ GO to the peptide during incubation produced smaller particles of ∼500 nm (∼26% reduction), ∼440 nm (∼35% reduction) and ∼400 nm (∼40% reduction) respectively after 6 days incubation under the same external conditions (Fig. 7(c)–(e) respectively). Interestingly, larger aggregates (∼5560 nm) were found to be present in fibrils only (without GO) and are absent in the sets containing GO. Since DLS cannot distinguish between fibrillar and non-fibrillar or globular like aggregates, it is likely that the particles found in presence of GO are non-fibrillar species and mixed aggregates of GO and peptides formed after 6 days incubation. ThT studies confirm that fibrillar species are not present in sets with higher concentrations of GO [Fig. 1(b)]. Thus the observed species [Fig. 7(e)] are most likely non fibrillar in nature. Such aggregates with globular characteristics are also found in the microscopic images [Fig. 10(d) and 11(d)]. Thus the formation of peptide aggregates is hindered by attachment of the two dimensional nanosheets with the peptide by various operating forces.^40,69

Fig. 7 Average hydrodynamic radii of protein monomer and protein aggregates in solution: (a) Aβ_25–35 peptide monomer and (b–d) Aβ_25–35 peptide incubated with (GO at 37 °C for 6 days: (b) 0 (c) 25 (d) 50 (e) 100 μg ml⁻¹ Inset: Aβ_25–35 peptide incubated without GO showing the large protein aggregates at size ∼5560 nm.

Matrix assisted laser desorption/ionization mass spectrometry

The inhibition study of amyloid fibril formation by GO was further investigated by MALDI TOF analysis. From Fig. 8(a), it is evident that the Aβ_25–35 peptide upon incubation forms aggregates of ∼8.6 kDa. This molecular weight corresponds to aggregates that may occur due to the self-assembly of 8 monomeric peptide units which possess ∼1.06 kDa molecular weight each.²¹ A simulation based study has shown that the Aβ_25–35 peptide requires at least 8 monomeric peptide units to aggregate and form fibrils. Thus the findings from the MALDI TOF experiment are in agreement with the simulation study.⁷⁰ Graphene oxide (100 μg ml⁻¹) is observed to completely break down these aggregates as is evident from Fig. 8(b).

Fig. 8 MALDI spectra of (a) aggregated form of bare Aβ_25–35 peptide obtained by incubation in PB for 6 days (b) peptide incubated with GO.

Morphology of the species: ThT staining and confocal microscopy

Long hair-like fibrils with a dense fibrillar network are obtained from the set containing peptide only as observed in Fig. 9(a). Treatment of the peptide with 25 μg ml⁻¹ of GO results in the formation of mixed aggregates with smaller hair like fibrils and oligomeric species [Fig. 9(b)]. Apart from the lower extent of fibril formation there is no branching observed. Fig. 9(c) and (d) show the confocal microscopic images of the ThT stained aggregates of Aβ_25–35 peptide containing 50 and 100 μg ml⁻¹ GO respectively. It is clear from the images that in the presence of 50 and 100 μg ml⁻¹ GO there is almost complete loss of the fibrillar network.

Fig. 9 Detection of amyloid fibrils by ThT staining. Confocal microscopic images of ThT stained amyloid aggregates containing (a) 0 (b) 25 (c) 50 and (d) 100 μg ml⁻¹ GO (scale bar = 100 μm).

Field emission scanning electron microscopy

FESEM images [Fig. 10(a–d)] of the peptide containing 0, 25, 50, 100 μg ml⁻¹ GO indicate the formation of long, twisted fibrils that are formed by the peptide due to incubation at 37 °C for 6 days [Fig. 10(a)]. Graphene oxide (25 μg ml⁻¹) induces the formation of smaller fibrils with some oligomeric species [Fig. 10(b)]. Again, the higher concentrations of GO (50, 100 μg ml⁻¹) are found to form smaller amorphous aggregates, thus pointing towards the interaction of smaller peptide molecules with the two dimensional material [Fig. 10(c) and (d) respectively]. The results obtained from FESEM experiments are in accordance with the spectroscopic and microscopic findings.

Fig. 10 Field emission scanning electron microscopic images (a–d) of Aβ_25–35 peptide aggregates in presence of different concentrations of GO (scale bar = 1 μm).

Atomic force microscopy: surface topography of aggregated species

Fibrillar growth and progression of the inhibition process were also monitored by observing topography images of GO treated and untreated peptide deposited on freshly cleaved mica surface. Thread like fibrils, tangled with each other are obtained from incubation of Aβ_25–35 peptide at 37 °C [Fig. 11(a)]. Aβ_25–35 peptide with 25 μg ml⁻¹ GO is noticed to produce scattered aggregates [Fig. 11(b)]. Further increase in concentration of GO (50 μg ml⁻¹) is observed to form more scattered smaller particles [Fig. 11(c)]. The chosen highest concentration of GO (100 μg ml⁻¹) is shown to form larger aggregates probably because of the agglomeration of peptide oligomers and GO sheets together [Fig. 11(d)].

Fig. 11 Topography (a–d) of amyloid fibrillar species on freshly cleaved mica surface: (a) Aβ_25–35 fibrillar species (b) Aβ_25–35 peptide + 25 μg ml⁻¹ GO (c) Aβ_25–35 peptide + 50 μg ml⁻¹ GO (d) Aβ_25–35 peptide + 100 μg ml⁻¹ GO.

Transmission electron microscopy

Electron microscopy images of Aβ_25–35 peptide (incubated in buffer at 37 °C) and peptide treated with 100 μg ml⁻¹ of GO are compared [Fig. 12(a) and (b) respectively]. A highly dense fibrillar network is obtained from the peptide whereas GO decreases the amount of fibril content.

Fig. 12 Transmission electron microscopic images of (a) Aβ_25–35 amyloid fibrils (b) Aβ_25–35 peptide co-incubated with 100 μg ml⁻¹ GO. The scale bar represents 200 nm.

Nanomaterials have been proven to tune protein aggregation depending on the surface area and nature (hydrophobicity or hydrophilicity) of the nanomaterial surface.^71,72 The results obtained from the experiments described above have directed us towards the inhibition property of nanomaterials. The ThT fluorescence assay pointed out the effective inhibitory potency of the two dimensional nanosheets in the fibrillation of the Aβ_25–35 peptide. The amount of fibril formation was found to decrease with increase in concentration of GO. This indicates that a higher availability of nanoparticle surface area leads to more efficient inhibition. The lowest concentration (GO: 25 μg ml⁻¹) studied also shows inhibition suggesting that there exists a strong interaction between the Aβ_25–35 peptide and GO. Other spectroscopic measurements, performed also corroborate the results obtained from the ThT assay. Morphological evolution of the aggregation and inhibition process was monitored by microscopic analyses and they were found to be in agreement with the spectroscopic studies.

Graphene oxide possesses a combination of polar and non-polar moieties where the hydrophobic basal plane is decorated with oxygen containing hydrophilic groups. Such structural features make it viable for biological applications as well as a compact binding propensity with several types of biomacromolecules. The nature of the interactions has been investigated from different perspectives. Since the Aβ_25–35 peptide does not contain any aromatic functionality, π-stacking interactions with GO are not possible. However, the possibility of hydrophobic and H-bonding interactions cannot be ruled out. GO contains a hydrophobic basal plane in addition to hydrophilic groups which permit both hydrophobic and hydrophilic interactions. We observe that GO is able to inhibit the fibrillation process of the Aβ_25–35 peptide even at very low concentrations unlike the Aβ_33–42 peptide which is found to require higher concentrations to inhibit the process.⁴⁰ Hydropathy of an amino acid is described as a measure of relative hydrophobicity or hydrophilicity. A larger hydropathy value is indicative of a higher hydrophobic character of the amino acid. Several hydropathy scales are known in the literature.^73,74 The average hydropathy score for a specific sequence length (window size) of amino acids of the protein/peptide is obtained for the length of the sequence (from N-terminus to C-terminus) and assigned to the central residue of the window. A hydropathy plot of a protein/peptide is obtained by plotting the average hydropathy score on the y-axis against the central amino acid residue index on the x-axis. The grand average of hydropathy score (GRAVY) of a protein or peptide is defined as the sum total of hydropathy scores of amino acids divided by the total chain length of the protein or peptide. Hydropathy plots of the Aβ_25–35 and Aβ_33–42 peptides and the corresponding amino acid sequences are given in Fig. 13. A comparison of the hydrophobic character of both peptides, obtained from a hydropathy calculation (GRAVY Scores), revealed that the Aβ_25–35 peptide is less hydrophobic than Aβ_33–42 [Fig. 13]. This implies that the extent of hydrophobic interactions in the Aβ_25–35 peptide is lower compared to the Aβ_33–42 peptide. Other non-bonding interactions possible with GO could be H-bonding and/or electrostatic interactions. Apart from hydrogen bonding possibilities of the peptide backbone (sequence of Aβ_25–35: GSNKGAIIGLM), the side chains of Ser, Asn and Lys at the N-terminal end of the peptide can be involved in hydrogen bonds with GO. A significant difference between the two peptides is the presence of Lys-28 in Aβ_25–35 which is protonated under the experimental conditions and can participate in electrostatic interactions with the carboxylate moieties of GO. Similar electrostatic interactions of GO with Lys residues of lysozyme have been reported.⁷⁵ In addition, peptides and amino acids are also found to show H-bonding, hydrophobic interactions with the basal plane of GO.^76–78 A schematic representation of possible electrostatic interactions between the Aβ_25–35 peptide and GO is shown in Scheme 2.

Fig. 13 Amino acid sequences, GRAVY scores and hydropathy plots of Aβ peptide fragments.

Scheme 2 Schematic representation of possible electrostatic interactions between the Aβ_25–35 peptide and GO.

Material and methods

Materials

Amyloid β peptide fragment (25–35), thioflavin T (ThT), hexafluoroisopropanol (HFIP), QuantiPro BCA assay kit (linear in the range 0.5–30 μg ml⁻¹ protein), and graphite powder (size <20 μm) were purchased from Sigma Aldrich (St. Louis, USA) and used as received. Potassium permanganate (KMnO₄; extra pure), sodium nitrate (NaNO₃) and all other materials for buffer preparation etc. were purchased from SRL, India. Hydrogen peroxide (30% H₂O₂), concentrated sulphuric acid (H₂SO₄) were purchased from Merck (India) Ltd. All the aqueous stock solutions, buffers were prepared in autoclaved Milli-Q and filtered again with 0.22 μm Millex Syringe Driven Filter units. Glasswares and microtips were carefully cleaned and autoclaved before use to stay away from any contamination.

Methods

Preparation of the monomeric peptide solution. The Aβ_25–35 peptide, obtained in lyophilized form was stored at −20 °C. Monomerisation of the peptide was performed following the HFIP method. For this step, 1 mg peptide was kept at room temperature to attain equilibration. Then 1 ml HFIP was added to it and bath sonicated for 15 min. The solution was incubated at room temperature overnight. The pre-existing oligomers break down to monomeric peptides via this treatment.⁷⁹ The peptide stock solution in HFIP was divided into 10 aliquots and stored at −80 °C. Before use, HFIP was evaporated under a N₂ atmosphere at ice-cold temperature to produce a peptide film on the wall of the tubes.

Preparation of peptide solutions for inhibition study. A stock solution of GO was prepared by dissolving 1 mg GO in 1 ml autoclaved ultrapure water. The GO solution was sonicated for several hours to obtain a monolayer of GO. For the dose-dependent inhibition study of GO on Aβ_25–35 peptide, the peptide films were dissolved in 50 mM phosphate buffer (pH 7.4) keeping the peptide concentration at 400 μM and the GO concentrations were maintained at 0, 25, 50 and 100 μg ml⁻¹ in different sets.

Thioflavin T fluorescence spectroscopy. ThT fluorescence spectra of fibrillar samples were obtained using a Horiba Jobin Yvon Fluoromax 4 or Fluorolog-3 spectrofluorimeter. For dose dependent and kinetic studies, aliquots of Aβ_25–35 fibrillar samples (in presence and absence of GO) were withdrawn at different time intervals, diluted with glycine-NaOH buffer (pH 8.5) containing ThT, incubated and scanned in a quartz cell of path length 1 cm. Samples were excited at 450 nm and emission range was set at 470–600 nm. The final peptide and ThT concentrations were kept as 5 μM and 10 μM respectively. Both excitation and emission slit widths were fixed to 5 nm and the integration time was kept as 0.3 s. Excitation spectra of the samples incubated with ThT were recorded by monitoring the emission at 485 nm under the same incubation conditions as the emission spectra. Each spectrum was corrected with respect to its corresponding blank.

Circular dichroism spectroscopy. Circular dichroism experiments were performed in a Jasco J-810 spectrophotometer using a quartz cuvette of pathlength 0.1 cm at 25 °C. CD spectra of Aβ_25–35 fibrils and fibrils containing 100 μg ml⁻¹ GO after incubation at 37 °C for 6 days were acquired at a scan rate of 10 nm min⁻¹. The scan range was fixed to 190–240 nm and 5 accumulations were averaged. Both spectra were corrected with respect to the corresponding blank spectra. Secondary structural content were analysed using the online DICHROWEB server.⁶²

Fourier transform infrared spectroscopy. Secondary structural changes of the peptide upon treatment with GO was also checked by FTIR spectroscopy. Aβ_25–35 peptide and the peptide (each 20 μL) with GO were dropped on a KBr pellet and air dried. The dried pellets were subjected to spectra acquisition in a Nexus 870 FTIR spectrometer (Thermo Nicolet Corporation) at room temperature in the spectral range (4000–400) cm⁻¹.

Bicinchoninic acid assay for protein quantification. Aβ_25–35 peptide samples containing GO of different concentrations (0, 25, 50 and 100 μg ml⁻¹) were subjected to centrifugation at 8000 rpm for 50 min to pellet down the insoluble fibrillar material. The supernatants were collected and mixed with the BCA working reagent followed by incubation at 65 °C for 1 h. Different concentrations of the standard, bovine serum albumin (BSA) were also incubated for calibration in the working range (0.5–30) μg ml⁻¹. Absorbance of the incubated samples were estimated at 562 nm in the Shimadzu UV-1800 UV-vis spectrophotometer. Amounts of soluble protein estimated in all sets are averaged over three separate experiments and the error bars are presented in the data.

Dynamic light scattering study. Light scattering measurements were carried out in Zetasizer nano ZS machine (Malvern instruments, Worcestershire, U.K) at 25 °C. Aβ_25–35 fibrils in presence and absence of GO were diluted in phosphate buffer (pH 7.4) for size analysis. A He–Ne laser of 4 mW was used as the source.

Matrix assisted laser desorption/ionization mass spectrometry. Average molecular weights of the Aβ_25–35 fibrils and fibrils containing 100 μg ml⁻¹ GO were estimated by matrix assisted laser desorption/ionisation mass spectrometry (MALDI) in a Bruker Daltonics Ultraflex MALDI TOF/TOF Mass Spectrometer (Germany) under an accelerating voltage of 20 kV. Sinapinic acid (3,5-dimethoxy-4-hydroxycinnamic acid) was used as the matrix. The saturated solution of the matrix was prepared by dissolving in 1 : 1 water–acetonitrile mixture (HPLC grade) containing 0.1% trifluoroacetic acid and mixed with the peptide samples. A linear acquisition mode was used for the measurements and 500 laser shots were summed over the mass range (5–50) kDa.

Thioflavin T staining and confocal microscopy. Fluorescence emission of ThT upon binding with fibrils was observed by confocal microscopy. Fibrillar samples were incubated with 1 mM ThT for at least 30 min to achieve staining and drop casted on precleaned glass slides. Spatially resolved images of the ThT fluorescence of amyloid fibrils were obtained with a confocal microscope (Leica TCS SP8, Leica Microsystems). An I3 filter cube was used and samples were viewed under a 10-fold, 0.4 NA dry objective. An emission bandwidth of 40 nm was used for the spectral measurements. It was found that photobleaching was insignificant during image and spectrum acquisition.

Field emission scanning electron microscopy. Morphology of the fibrillar samples were obtained from FESEM analyses. Bare amyloid fibrils and fibrils containing GO after 6 days incubation at 37 °C were diluted and drop casted on clean glass slides for FESEM analyses. The samples were dried, gold coated and viewed under NOVA NANOSEM 450 operating at 5 kV.

Atomic force microscopy. AFM was used to obtain the topography of the fibrils with different concentrations of GO. The samples were drop casted on freshly cleaved mica foils and dried overnight. The dried samples were scanned under an atomic force microscope, Model No. 5500, Agilent Technologies. A silicon probe cantilever (length 215–235 μm), resonance frequency of 146–236 kHz and force constant 21–98 N m⁻¹ was used for capturing the topographic images of the samples.

Transmission electron microscopy. Images of Aβ_25–35 fibrils and fibrils with the maximum concentration of GO were also captured using electron microscopic techniques. Samples were placed on TEM grids, negatively stained with 1% (w/v) uranyl acetate. The dried samples were scanned under TECNAI G2 20S-TWIN TEM, operating at an accelerating voltage of 120 kV.

Investigating nature of interactions: hydropathy plot. Hydrophobicity of peptide fragments were analysed using hydrophobicity values of Kyte–Doolittle scale.⁷³ The grand average hydropathy (GRAVY) values were calculated and compared.

Conclusions

Tuning the fibrillation of amyloid β-peptides has been investigated using several types of small molecules and nanomaterials. In the present case, the self-association of the Aβ_25–35 peptide was modulated by prepared monolayer graphene oxide. The water solubility and biocompatibility of GO has drawn its attention towards biological applications. In this work, we demonstrate that GO inhibits fibrillation of the Aβ_25–35 peptide in a dose dependent manner. ThT fluorescence kinetics reveal that fibril formation is attenuated even at low concentrations of GO. Secondary structural estimation by CD, FTIR analysis also supports these findings. The amount of unordered structures increases with a reduction of β-sheet content upon treatment with GO. Quantification of the soluble part of the protein by the BCA assay shows that GO is able to retard peptide aggregation. AFM, confocal microscopy and electron microscopy validate the observations from the spectroscopic measurements. An investigation regarding the inhibition of fibrillation suggest that the adhesion of a relatively large accessible surface area of Aβ_25–35 peptide with GO by hydrophobic, electrostatic and H-bonding interactions reduces the fibrillar assembly even at low concentrations of GO. Such strong binding characteristics of nanomaterials with the Aβ_25–35 peptide can further provide an insight into the mechanism of fibrillation and facilitate the design of effective inhibitors of the fibrillation process.

Acknowledgements

The authors are grateful to the Ministry of Human Resource Development, Government of India for research grants (IIT/SRIC/ATDC/CEM/2013-2014/118). SB and AS wish to thank CSIR, New Delhi and MHRD, New Delhi respectively for financial assistance. We also gratefully acknowledge the Central research facility (CRF), IIT Kharagpur for the support with instrumental facilities.

References

1. D. J. Selkoe, J. Biol. Chem., 1996, 271, 18295.
2. M. Sunde and C. Blake, Adv. Protein Chem., 1997, 50, 123.
3. M. Sunde, L. C. Serpell, M. Bartlam, P. E. Fraser, M. B. Pepys and C. C. Blake, J. Mol. Biol., 1997, 273, 729.
4. T. L. Benzinger, D. M. Gregory, T. S. Burkoth, H. Miller-Auer, D. G. Lynn, R. E. Botto and S. C. Meredith, Biochemistry, 2000, 39, 3491.
5. G. G. Glenner and C. W. Wong, Biochem. Biophys. Res. Commun., 1984, 120, 885.
6. T. Kubo, S. Nishimura, Y. Kumagae and I. Kaneko, J. Neurosci. Res., 2002, 70, 474.
7. E. Gowing, A. E. Roher, A. S. Woods, R. J. Cotter, M. Chaney, S. P. Little and M. J. Ball, J. Biol. Chem., 1994, 269, 10987.
8. Y. Harigaya, T. C. Saido, C. B. Eckman, C. M. Prada, M. Shoji and S. G. Younkin, Biochem. Biophys. Res. Commun., 2000, 276, 422.
9. J. Wiltfang, H. Esselmann, M. Bibl, A. Smirnov, M. Otto, S. Paul, B. Schmidt, H. W. Klafki, M. Maler, T. Dyrks, M. Bienert, M. Beyermann, E. Rüther and J. Kornhuber, J. Neurochem., 2002, 81, 481.
10. H. C. Kim, K. Yamada, A. Nitta, A. Olariu, M. H. Tran, M. Mizuno, A. Nakajima, T. Nagai, H. Kamei, W. K. Jhoo, D. H. Im, E. J. Shin, O. P. Hjelle, O. P. Ottersen, S. C. Park, K. Kato, M. E. Mirault and T. Nabeshima, Neuroscience, 2003, 119, 399.
11. K. Ono, K. Hasegawa, H. Naiki and M. Yamada, J. Neurosci. Res., 2004, 75, 742.
12. B. A. Yankner, L. K. Duffy and D. A. Kirschner, Science, 1990, 250, 279.
13. S. M. Yatin, M. Aksenov and D. A. Butterfield, Neurochem. Res., 1999, 24, 427.
14. M. E. Clementi, S. Marini, M. Coletta, F. Orsini, B. Giardina and F. Misiti, FEBS Lett., 2005, 579, 2913.
15. K. Sato, A. Wakamiya, T. Maeda, K. Noguchi, A. Takashima and K. Imahori, J. Biochem., 1995, 118, 1108.
16. Q. Liu and B. Zhao, Br. J. Pharmacol., 2004, 141, 746.
17. S. Varadarajan, S. Yatin, J. Kanski, F. Jahanshahi and D. A. Butterfield, Brain Res. Bull., 1999, 50, 133.
18. G. Vitiello, S. Di Marino, A. M. D'Ursi and G. D'Errico, Langmuir, 2013, 29, 14239.
19. A. Karsai, A. Nagy, A. Kengyel, Z. Mártonfalvi, L. Grama, B. Penke and M. S. Kellermayer, J. Chem. Inf. Model., 2005, 45, 1641.
20. A. Accardo, V. Shalabaeva, M. Cotte, M. Burghammer, R. Krahne, C. Riekel and S. Dante, Langmuir, 2014, 30, 391.
21. S. Bag, S. Chaudhury, D. Pramanik, S. DasGupta and S. Dasgupta, Proteins, 2016, 84, 1213.
22. A. Sett, S. Bag, S. Dasgupta and S. DasGupta, Int. J. Biol. Macromol., 2015, 79, 344.
23. N. Erathodiyil and J. Y. Ying, Acc. Chem. Res., 2011, 44, 925.
24. I. I. Slowing, B. G. Trewyn, S. Giri and V. S.-Y. Lin, Adv. Funct. Mater., 2007, 17, 1225.
25. W. H. Wu, X. Sun, Y. P. Yu, J. Hu, L. Zhao, Q. Liu, Y. F. Zhao and Y. M. Li, Biochem. Biophys. Res. Commun., 2008, 373, 315.
26. L. Xiao, D. Zhao, W. H. Chan, M. M. Choi and H. W. Li, Biomaterials, 2010, 31, 91.
27. S. Li, L. Wang, C. C. Chusuei, V. M. Suarez, P. L. Blackwelder, M. Micic, J. Orbulescu and R. M. Leblanc, Chem. Mater., 2015, 27, 1764.
28. L. Xie, Y. Luo, D. Lin, W. Xi, X. Yang and G. Wei, Nanoscale, 2014, 6, 9752.
29. X. Yu, Q. Wang, Y. Lin, J. Zhao, C. Zhao and J. Zheng, Langmuir, 2012, 28, 6595.
30. L. Dusan, L. M. Lisandra, A. Marie-Isabel and H. S. David, Pept. Sci., 2006, 84, 519.
31. T. Kowalewski and D. M. Holtzman, Proc. Natl. Acad. Sci. U. S. A., 1999, 96, 3688.
32. A. K. Geim, Science, 2009, 324, 1530.
33. C. N. Rao, A. K. Sood, K. S. Subrahmanyam and A. Govindaraj, Angew. Chem., Int. Ed., 2009, 48, 7752.
34. J. Allen, V. C. Tung and R. B. Kaner, Chem. Rev., 2010, 110, 132.
35. A. K. Geim, Angew. Chem., Int. Ed., 2011, 50, 6966.
36. D. Li, M. B. Muller, S. Gilje, R. B. Kaner and G. G. Wallace, Nat. Nanotechnol., 2008, 3, 101.
37. S. Stankovich, D. A. Dikin, R. D. Piner, K. A. Kohlhaas, A. Kleinhammes, Y. Jia, Y. Wu, S. B. T. Nguyen and R. S. Ruoff, Carbon, 2007, 45, 1558.
38. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
39. J. Kim, L. J. Cote, F. Kim, W. Yuan, K. R. Shull and J. Huang, J. Am. Chem. Soc., 2010, 132, 8180.
40. Q. Li, L. Liu, S. Zhang, M. Xu, X. Wang, C. Wang, F. Besenbacher and M. Dong, Chem.–Eur. J., 2014, 20, 7236.
41. M. Mahmoudi, O. Akhavan, M. Ghavami, F. Rezaee and S. M. A. Ghiasi, Nanoscale, 2012, 4, 7322.
42. Y. Liu, L. P. Xu, W. Dai, H. Dong, Y. Wen and X. Zhang, Nanoscale, 2015, 7, 19060.
43. Z. Yang, C. Ge, J. Liu, Y. Chong, Z. Gu, C. A. Jimenez-Cruz, Z. Chai and R. Zhou, Nanoscale, 2015, 7, 18725.
44. L. Baweja, K. Balamurugan, V. Subramanian and A. Dhawan, J. Mol. Graphics Modell., 2015, 61, 175.
45. X. Yu, Q. Wang, Y. Lin, J. Zhao, C. Zhao and J. Zheng, Langmuir, 2012, 28, 6595.
46. X. Wang, J. K. Weber, L. Liu, M. Dong, R. Zhou and J. Li, Nanoscale, 2015, 7, 15341.
47. G. Qing, S. Zhao, Y. Xiong, Z. Lv, F. Jiang, Y. Liu, H. Chen, M. Zhang and T. Sun, J. Am. Chem. Soc., 2014, 136, 10736.
48. M. Jayaraman, G. Kannayiram and J. Rajadas, Arch. Biochem. Biophys., 2008, 474, 15.
49. L. Liu, N. Xia, J. Zhang, W. Mao, Y. Wua and X. Gea, Anal. Methods, 2015, 7, 8727.
50. F. Wang, C. Liu, Y. Fan, Y. Wang and Z. Li, Chem. Commun., 2014, 50, 8161.
51. H. Jang, J. Lee and D.-H. Min, J. Mater. Chem. B, 2014, 2, 2452.
52. M. De, S. S. Chou and V. P. Dravid, J. Am. Chem. Soc., 2011, 133, 17524.
53. J. I. Paredes, S. Villar-Rodil, A. Martinez-Alonso and J. M. D. Tascon, Langmuir, 2008, 24, 10560.
54. I. Roy, D. Rana, G. Sarkar, A. Bhattacharyya, N. R. Saha, S. Mondal, S. Pattanayak, S. Chattopadhya and D. Chattopadhyay, RSC Adv., 2015, 5, 25357.
55. M. Biancalana, K. Makabe, A. Koide and S. Koide, J. Mol. Biol., 2009, 385, 1052.
56. M. R. H. Krebs, E. H. C. Bromley and A. M. Donald, J. Struct. Biol., 2005, 14, 30.
57. H. LeVine, Methods Enzymol., 1999, 309, 274.
58. R. Sabate, L. Rodriguez-Santiago, M. Sodupe, S. J. Saupec and S. Venturad, Chem. Commun., 2013, 49, 5745.
59. L. Nielsen, R. Khurana, A. Coats, S. Frokjaer, J. Brange, S. Vyas, V. N. Uversky and A. L. Fink, Biochemistry, 2001, 40, 6036.
60. 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.
61. H. Naiki and K. Nakakuki, Lab. Invest., 1996, 74, 374.
62. L. Whitmore and B. A. Wallace, Nucleic Acids Res., 2004, 32, W668.
63. J. F. Labbé, T. Lefèvre, A. A. Guay-Bégin and M. Auger, Phys. Chem. Chem. Phys., 2013, 15, 7228.
64. M. Del Mar Martínez-Senac, J. Villalaín and J. C. Gómez-Fernández, Eur. J. Biochem., 1999, 265, 744.
65. D. Naumann, Infrared spectroscopy in microbiology, John Wiley & Sons Ltd, Chichester, 2000.
66. P. K. Smith, R. I. Krohn, G. T. Hermanson, A. K. Mallia, F. H. Gartner, M. D. Provenzano, E. K. Fujimoto, N. M. Goeke, B. J. Olson and D. C. Klenk, Anal. Biochem., 1985, 150, 76.
67. X. Wang, X. Wang, C. Zhang, Y. Jiaoa and Z. Guo, Chem. Sci., 2012, 3, 1304.
68. T. M. Ryan, J. Caine, H. D. Mertens, N. Kirby, J. Nigro, K. Breheney, L. J. Waddington, V. A. Streltsov, C. Curtain, C. L. Masters and B. R. Roberts, PeerJ, 2013, 1, e73.
69. M. Mahmoudi, M. P. Monopoli, M. Rezaei, I. Lynch, F. Bertoli, J. J. McManus and K. A. Dawson, ChemBioChem, 2013, 14, 568.
70. Z. Fu, Y. Luo, P. Derreumaux and G. Wei, Biophys. J., 2009, 97, 1795.
71. C. Cabaleiro-Lago, F. Quinlan-Pluck, I. Lynch, K. A. Dawson and S. Linse, ACS Chem. Neurosci., 2010, 1, 279.
72. M. Ghavami, M. Rezaei, R. Ejtehadi, M. Lotfi, M. A. Shokrgozar, B. A. Emamy, J. Raush and M. Mahmoudi, ACS Chem. Neurosci., 2013, 4, 375.
73. J. Kyte and R. F. Doolittle, J. Mol. Biol., 1982, 157, 105.
74. G. D. Rose and R. Wolfenden, Annu. Rev. Biophys. Biomol. Struct., 1993, 22, 381.
75. S. Li, J. J. Mulloor, L. Wang, Y. Ji, C. J. Mulloor, M. Micic, J. Orbulescu and R. M. Leblanc, ACS Appl. Mater. Interfaces, 2014, 6, 5704.
76. S. Li, A. N. Aphale, I. G. Macwan, P. K. Patra, W. G. Gonzalez, J. Miksovska and R. M. Leblanc, ACS Appl. Mater. Interfaces, 2012, 4, 7069.
77. M. Zhang, B. C. Yin, X. F. Wang and B. C. Ye, Chem. Commun., 2011, 47, 2399.
78. J. Zhang, F. Zhang, H. Yang, X. Huang, H. Liu, J. Zhang and S. Guo, Langmuir, 2010, 26, 6083.
79. W. B. Stine, L. Jungbauer, C. Yu and M. J. LaDu, Methods Mol. Biol., 2011, 670, 13.

---

Footnote

† Electronic supplementary information (ESI) available: Synthesis and physical characterisation of the prepared graphene oxide (GO). See DOI: 10.1039/c6ra23570k

---