Tuning the aqueous self-assembly process of insulin by a hydrophobic additive

Abstract

Biomolecular self-assembly is an efficient way of preparing soft-matter based materials. Herein we report a novel method, based on the use of insoluble additives in aqueous media, for influencing the self-assembly process. Due to their low solubility, the use of hydrophobic additives in aqueous media is problematic; however, by mixing the additive with the biomolecule in the solid state, prior to solvation, this problem can be circumvented. In the investigated self-assembly system, where bovine insulin self-assembles into spherical structures, the inclusion of the hydrophobic material α-sexithiophene (6T) results in significant changes in the self-assembly process. Under our reaction conditions, in the case of materials prepared from insulin-only the growth of spherulites typically stops at a diameter of 150 μm. However, by adding 2 weight% of hydrophobic material, spherulite growth continues up to diameters in the mm-range. The spherulites incorporate 6T and are thus fluorescent. The method reported herein should be of interest to all scientists working in the field of self-assembly as flexible material preparation, based simply on co-grinding of commercially available materials, adds another option to influence the structure and properties of products formed by self-assembly reactions.

---

Introduction

Self-assembly of molecular building blocks is a powerful and widely used method to prepare soft-matter based materials.¹ In order to avoid the often laborious synthesis of building blocks, an attractive approach is to use materials of biological origin, such as proteins, with a known capacity to self-assemble.^2–5 Self-assembly of proteins occurs in aqueous environment and is largely driven by a delicate balance of hydrophilic and hydrophobic interactions.⁶ It is possible to influence such interactions by water soluble substances/additives such as co-solvents miscible with water, salts, or polyelectrolytes, and these types of additives often have dramatic effects on the outcome of self-assembly processes. In addition, hydrophobic substances/substituents may influence self-assembly and aggregation of biomolecules. However, the use of strongly hydrophobic substances as an additive is complicated by the low solubility of such substances in water thus making it is difficult to ensure proper mixing of the biomolecules and the hydrophobic materials.

A common way to enable solubilization/dispersion in water of hydrophobic substances is the use of detergents. Many proteins have properties similar to detergents, and even strongly hydrophobic materials, such as carbon nanotubes, have been dispersed in water by proteins.⁷ Such systems, however, are rarely investigated regarding the influence of the hydrophobic materials on the self-assembly properties of proteins. For such studies it is more common to attach the hydrophobic unit by a covalent bond to the biomolecule, e.g. by amino acid substitution in polypeptides.⁸ Moreover, in biomolecule inspired systems such as amphiphilic peptides, the relative proportions of hydrophilic and hydrophobic parts can be modified by covalent synthetic chemistry.⁹ The same goes for amphiphilic block copolymers where the relative size of the hydrophobic and hydrophilic blocks can be tuned in order to influence the self-assembly process.¹⁰ It has also been shown that hydrophobic surfaces may influence the aggregation of proteins into amyloid materials.^11,12 However, the influence of non-covalently attached strongly hydrophobic organic molecules on the self-assembly process of biomolecules such as proteins is a largely unexplored area of research. Herein we investigate the effect of a hydrophobic additive on the self-assembly of bovine insulin into amyloid materials.^13–17 Bovine insulin is a small globular protein that when heated in aqueous acidic solutions forms fibrillar objects, known as amyloid fibrils that typically have diameters in the nm-range and lengths in the μm range.^13–17 Moreover there is often concomitant formation of spherical objects, known as spherulites, made up of an amorphous core surrounded by radially oriented amyloid fibril-like materials.^18–31 The formation of spherulites is thus a process competing with fibril formation. In fact, under many reaction conditions employed for amyloid formation, sperulites rather than fibrils constitute the main product.²⁹ A typical diameter of insulin spherulites is 50 μm with a maximum diameter of 150 μm. However, it should be noted that a recent report demonstrated the growth of unusually large insulin spherulites.³¹ Due to their size, the protein spherulites can be conveniently analyzed by microscopy. Moreover, under plane polarized light spherulites exhibit a typical Maltese cross pattern which facilitates identification of the spherulitic structures.¹⁸ Both amyloid fibrils and spherulites have been observed in amyloid plaques in tissue from patients with Alzheimer's disease; it is thus important to understand factors influencing the formation of both types of structures.^32,33

Herein we utilize mechanochemistry to mix insulin and the hydrophobic additive, enabling us to examine the effect of the hydrophobic compound on the self-assembly process where insulin forms protein spherulites. Mechanochemistry has in recent years seen a huge upsurge of interest, manifested for example by several recent review articles.³⁴ A common approach in mechanochemistry is to mix dry powders by mechanical means such as grinding. In this manner a wide range of chemical transformations can be achieved ranging from covalent bond formation to cocrystal formation and formation of mixed powders. The use of mechanochemistry should thus in principle provide a way to use strongly hydrophobic molecules as additives for biomolecular self-assembly processes. By co-grinding powders of the biomolecule and the hydrophobic substance a composite material can be formed. If such a composite material is soluble in water the effect of the hydrophobic substance on the self-assembly process could be evaluated. We base our approach on previously developed methodology,^35–38 where a premixing of the reagents is performed in the solid state by grinding. As hydrophobic additive we choose to investigate the strongly hydrophobic molecule α-sexithiophene (6T). 6T is a material of considerable interest for materials science applications as a semi-conductor or light emitting material.^39,40 However, the use of 6T is complicated by a low solubility even in organic solvents. Thus 6T is an interesting candidate to investigate the influence of strongly hydrophobic additives on the self-assembly processes of biomolecules. Furthermore, as 6T is a fluorescent material it is easy to verify the presence of 6T in the assembled materials by fluorescence microscopy. Moreover, 6T fluorescence is sensitive to the molecules aggregations state,³⁹ making it possible to detect if 6T is present in dispersed or aggregated form. We have previously reported the preparation of amyloid fibrils incorporating 6T, where agitation promoted formation of amyloid fibrils.³⁷ Herein we report that 6T can dramatically influence the self-assembly process where insulin forms spherulites under quiescent conditions. In the presence of 6T extremely large spherulitic structures are formed with diameters in the mm range, whereas under identical reaction conditions, but in the absence of 6T, normal protein spherulites are formed. In addition, the presence of 6T enable merging growth of spherulites. These results demonstrates the dramatic effect of hydrophobic molecules on insulin spherulite growth, and points to an additional tool whereby aqueous self-assembly processes can be influenced.

Materials and methods

Materials

Bovine insulin was purchased from Sigma-Aldrich, and α-sexithiophene was purchased from TCI, and used as received. In a typical procedure, 25 mg of bovine insulin was ground with 0.5 mg of α-sexithiophene. The grinding was done with a mortar and pestle for 10 minutes. The resulting composite material was dissolved in 5 ml 25 mM hydrochloric acid and filtered through a 0.2 μm PVDF filter. The samples were then heated at 65 °C and left standing for up to several days in order to induce spherulite formation. Control experiments for the insulin-only system were performed in a similar manner by dissolving 25 mg of bovine insulin in 5 ml 25 mM hydrochloric acid followed by filtering through a 0.2 μm PVDF filter prior to heating at 65 °C in order to induce spherulite formation.

Photographs

The photographs in Fig. 1b and c were taken using an Olympus E-300 digital camera equipped with a ZUIKO DIGITAL ED 40–150 mm 1 : 4.0–5.6 lens.

Fig. 1 The effect of 6T on the self-assembly of insulin in 25 mM HCl. The scalebars represent 300 μm. (a) Polarized light microscopy images of reaction solutions with and without 6T, after 12 hours heat treatment at 65 °C. At the bottom is shown the corresponding images when the samples are irradiated with 420 nm light. (b) Photograph of a 6T-containing reaction mixture after being heated at 65 °C for several days. (c) Photograph of a macroscopic spherulite with a diameter of 1.4 mm. (d) Polarized light microscopy image of a 6T-containing spherulite.

The fluorescent stereomicroscope images of spheres in Fig. 3a and b, were obtained by directing an externally mounted 420 nm emitting lamp source towards the samples.

Polarized light microscopy

The polarized light microscopy images of spherulites in solution were recorded using Olympus BH-2 microscope with a top mounted Olympus E-300 digital camera. For the images in Fig. 2a–d a quartz glass cuvette with an interior width of 1 mm and a length of 10 mm was used as a reaction chamber.

Fig. 2 Spherulite growth and isolated merged structures. The white scalebars represent 500 μm. (a–d) Polarized light microscopy images from a 6T-containing insulin solution (a and c) and the corresponding images 24 hours later (b and d). (e) Light microscopy images of isolated merged spherulites. (f) Polarized light microscopy image of structure shown in (e).

Fluorescence lifetime imaging

FLIM data as well as the correlated fluorescence microscopy images were recorded on a Zeiss LSM 780 microscope at an excitation wavelength of 405 nm.

Fluorescence microscopy

The fluorescence microscope images were recorded with an epifluorescence microscope (Zeiss Axiovert inverted microscope A200 Mot) equipped with a CCD camera (Axiocam HR).

Microtome cutting

The slices in Fig. 3c, d, f and g was obtained from a 6T-containing sphere using a microtome. Spherulites were separated from a sample solution using a pipette and pretreated with paraffin prior to cutting with the microtome. Paraffin was subsequently removed from the slices by repeated soaking of the sample glass slides in tissue clear (from Histolab) and finally EtOH.

Fig. 3 Emission data obtained from a 6T-containing spherulite. Data obtained by excitation at 420 nm for (a) and (b) and 405 nm for (c). (a) Surface view of the sphere shown in Fig. 1c. (b) An in close photograph of a large spherulite. (c) FLIM image of a 10 μm thick spherulite slice. Lifetime distribution is shown in a color gradient from red (300 ps) to blue (1100 ps). (d) Fluorescence microscopy image of the slice shown in (c). (e) Intensity weighted mean lifetimes of slice shown in (c). (f) FLIM image of sliced merged spherulites. Lifetime distribution is shown in a color gradient from red (300 ps) to blue (1100 ps). (g) Fluorescence microscopy image of the sliced spherulites shown in (f). (h) Intensity weighted mean lifetimes of sliced spherulites in (f).

Dynamic light scattering

Correlation curves were recorded using an ALV/DLS/SLS-5022 compact goniometer system from ALV Gmbh using a HeNe-laser (632 nm) as light source. To determine size distributions CONTIN analysis was performed using the ALV-500/E/EPP&ALV 60X0-win software. Prior to measuring the samples were filtrated using a 0.2 μm PVDF filter.

Results and discussion

In order to test the effect of 6T on the self assembly process of insulin we set up parallel reactions run at identical conditions, apart from the presence of 6T. 6T–insulin samples were prepared by grinding 0.5 mg 6T with 25 mg bovine insulin. The resulting composite material was then dissolved in 5 ml 25 mM HCl. After filtration, 0.5 ml of the solution was put in a capped vial and was thereafter heated at 65 °C. The corresponding insulin-only samples were prepared in an identical way except that the grinding step with 6T was omitted. The presence of 6T resulted in a dramatic difference in product structure, as can be seen in the images of reaction products shown in Fig. 1. In the left part of Fig. 1a is shown a plane polarized light microscopy image of an aliquot removed from an insulin-only reaction mixture after 12 hours. A large number of spherical objects, with typical diameters of 50 μm, and a maximum diameter of about 150 μm, can be observed (see also Fig. S1†). On the other hand, in the aliquot removed after 12 hours from the sample where insulin was ground with 6T, the structures are remarkably different. A comparatively small number of larger spherulites can be observed (Fig. 1a, see also Fig. S2†). Moreover, the spherulites are fluorescent when irradiated with 420 nm light, demonstrating the incorporation of 6T. On the other hand, insulin-only spherulites are not fluorescent, as demonstrated in the control experiment shown in the bottom left of Fig. 1a. Another difference between the insulin-only and insulin–6T reactions is the structural evolution over time. Thus, when the insulin–6T reaction mixture is kept at 65 °C for several days the spherulites grow into larger structures with diameters typically ranging between 0.3–1.4 mm (a photograph of a typical reaction mixture is shown in Fig. 1b). This result has been reproduced in several independent runs.

The formation of such large structures is in stark contrast with the self-assembly process observed under the same conditions for the insulin-only system, which even if kept for weeks rarely result in spherulites with a diameter wider than ∼150 μm (c.f. Fig. S1 and S2 as well as Fig. S3†). The result for the insulin-only reaction is in agreement with a large number of studies on insulin spherulite formation.^18–23 We have earlier reported the preparation of amyloid fibrils incorporating 6T. In this case we found it beneficial to agitate the solutions by magnetic stirring in the initial stages of the aggregation process. In this manner we could suppress the formation of spherulites.³⁷ However, when heating the 6T–insulin under quiescent conditions the main product is large spherical objects incorporating 6T. The 6T-containing spherical objects were investigated by various microscopy techniques. The spheres are stable structures and can be kept in aqueous solvent for at least two years without any visible structural changes. Moreover, the practical handling of the structures is facile and spherulites can be transferred from the reaction mixture onto e.g. microscope slides simply by pipetting. A photograph of an isolated spherical object with a diameter of 1.4 mm is shown in Fig. 1c. Protein aggregates typically have a white color; however, these spherical objects have a distinct orange color, demonstrating the presence of 6T. In addition, in Fig. 1d is shown an enlarged image of a spherical object illuminated under crossed polarizers; in this image a core region, surrounded by a Maltese cross pattern can be observed. This shows that the structure of the spherical object is related to that of previously reported spherulites, with a core region containing unordered protein surrounded by a region containing radially oriented fibrils.¹⁸

The structural development of the self-assembly process over time was investigated for both insulin-only and 6T–insulin systems. When comparing the outcomes, a typical insulin-only sample has a noticeably higher number of spherulites than an equivalent 6T–insulin sample at any given time point during growth (c.f. Fig. S1 and S2†). 6T-containing spherulites of sizes up to 150 μm that are indistinguishable from insulin-only spherulites of a similar size (apart from being fluorescent) are present throughout the growth process. However, 6T-containing insulin spherulites have a markedly higher tendency to form clusters than insulin-only spherulites (see Fig. S2b†). Finally, the self-assembly process in 6T-containing insulin solutions consistently result in formation of considerably larger spherulites, with diameters up to 1.4 mm.

In order to investigate the growth mechanism of individual spherulites we performed light microscopy studies on spherulite growth in a closed reaction chamber. In this manner we could obtain sequential images of the same area and thereby follow the structural evolution over time. Two examples of spherulite growth obtained from such measurements are shown in Fig. 2. In the bottom right of Fig. 2a an isolated spherulite can be observed with a diameter of about 200 μm. In Fig. 2b, the same structure is shown after 24 h of heating at 65 °C, where the spherulite structure has undergone growth to reach a diameter of about 500 μm. Statistical data on such isolated spherulite growth over time indicate that the mean diameter increases over time going from a mean diameter of approximately 250 μm after 4 hours to stabilize at a mean diameter of approximately 400 μm after 79 hours (see ESI, Fig. S10†).

Another type of growth process occurring in addition to that of isolated spherulite growth can be observed in the middle part of Fig. 2c and d, where a cluster of spherulites seems to grow into each other, thus resulting in a merging of spherulites. This is a feature we have not observed when investigating the insulin-only system. In the 6T–insulin system, this merging growth was observed repeatedly during a large number of experiments, and thus represents a unique process enabled by the hydrophobic additive. An example of an isolated structure of merged spherulites is shown in Fig. 2e and f. Additional images of merged spherulites are shown in the ESI Fig. S4–S8.† In addition, microscope images from the isolated structure in Fig. 2e and f seen from different angles (Fig. S6†) and fluorescence microscope images (Fig. S7) are available in the ESI.† An additional isolated merged structure is shown from different angles in Fig. S8.† It is thus obvious that the inclusion of the hydrophobic additive opens up novel reaction pathways. Under the investigated reaction conditions, insulin-only spherulites form stable structures with diameters up to ∼150 μm. Although at times clustering of individual insulin-only spherulites does occur, we have never observed such spherulites form merged structures of the type shown in Fig. 2e. In contrast, the 6T-containing insulin spherulites, in addition to isolated growth, undergoes merging. The presence of 6T promotes spherulite growth allowing formation of structures with a diameter significantly larger than ∼150 μm, which under our reaction conditions represent the endpoint of growth in the insulin-only case. The addition of 2 weight% of hydrophobic 6T thus has dramatic effects on the insulin self-assembly process, and in order to obtain mechanistic insight structural development as a function of time was investigated.

Due to difficulties in obtaining reliable data from a single reaction-chamber (caused by solvent evaporation and spherulites obscuring one another), we opted for setting up 7 parallel reactions, and analysing these one after the other at the appropriate time. A stock solution was prepared, from which 100 μl samples were distributed into 7 vials, which were subsequently heated at 65 °C. After the allotted time the entire content of the vial was transferred to a microscope slide. Analysis was then performed on the still wet sample by polarized light microscopy (Fig. S9†). In this manner we could estimate of the size distribution of spherulites, as well as the number of spherulites at the time when the reaction was stopped (Fig. S10a†). Moreover, in Fig. S10b† is also shown the size distribution of spherulites for reaction mixtures at 12, 22, and 79 hours. For practical reasons we only included spherulites with a diameter larger than 150 μm in the statistics. The graphs in Fig. S10† show a decrease in spherulite size over time with a concomitant increase of spherulite diameter.

It should be noted that the factors influencing growth in the 6T–insulin system are complex; the fact that 6T is hydrophobic does not in itself explain the new phenomena. For example, in an earlier study where we used hydrophobic iridium-complexes in place of 6T, we could only observe the formation of relatively normal amyloid-like materials.³⁵ To elucidate this matter, we further investigated the growth process and structural properties of the 6T-containing spherulites utilizing their fluorescent properties. If left to dry, the surface region of the spheres often will crack open, affording opportunities to investigate the interior parts of the spherulite. Typically a non-homogenous fluorescent emission is observed for such spherulites when excited with a 420 nm light source (Fig. 3a and b). On closer examination, the surface region has a yellow/orange fluorescence, as opposed to the immediate interior that exhibits a green fluorescence. In order to further characterize the difference of the surface and the interior of spherulites, a spherulite with a diameter of 415 μm was cut into slices with a thickness of 10 μm using a microtome apparatus. The slices were then analyzed by fluorescence microscopy and fluorescence lifetime imaging microscopy (FLIM). Both types of measurements were performed using an excitation wavelength of 405 nm. A typical FLIM-image of such a slice is shown in Fig. 3c. The corresponding fluorescence microscope image, obtained from the same slice as shown in Fig. 3d, displays strong green emission from the area corresponding to longer lifetimes (∼800–900 ps) in the FLIM-image (Fig. 3c). In contrast, there is weak emission in the outer region of the fluorescence microscope image corresponding to the region corresponding to shorter lifetimes (∼400–600 ps). FLIM and fluorescence microscopy thus shows that the slice consists of two regions with notably different properties regarding 6T fluorescence. Note that during preparation of slices the 6T-enriched surface tends become outstretched and to fold into the image-plane as a result of mechanical stress during the microtome processing. The yellow-orange colored region in Fig. 3c, corresponding to a population with a lifetime between ∼400–600 ps can thus not be used to assess the thickness of the outer region. We have instead estimated the thickness of the outer region from microscopy images, such as the ones shown in Fig. 3b, according to which the outer region has an apparent thickness in the single digit μm range. The different fluorescence characteristics observed in Fig. 3 stems from the sensitivity of 6T fluorescence to the aggregation state of 6T.^39,40 It is well established that 6T fluorescence is heavily dependent on the 6T packing. For example, when 6T is dispersed in a polymer matrix 6T fluorescence is of relatively high intensity and blue shifted compared to when 6T is present in a more aggregated state.^39,40 In contrast, in the aggregated state, 6T fluorescence typically has a maximum in the orange region of the spectrum with a relatively low intensity.^39,40 In the microtome slices, the lifetime of the 6T excited state is longer in the interior region than in the exterior region. Moreover, the interior region of the slice displays green fluorescence typical of dispersed 6T. In contrast, the fluorescence from the exterior region is red-shifted and weak. The fluorescence characteristics in the interior region of the slice thus corresponds to dispersed 6T, whereas the outer regions fluorescence corresponds to aggregated 6T.³⁹ We have previously demonstrated that 6T is organized in amyloid fibrils along the long 6T axis parallel to the long axis of the fibril.³⁷ The reason for this alignment is the fine structure of amyloid fibrils which as a result of extensive β-sheet formation display hydrophobic channels oriented along the fibril long axis. In addition other hydrophobic molecules are known to be intercalated in a similar way.^26,41,42 The binding of 6T to fibrillar regions of the spherulite can be expected to occur in a similar fashion, with 6T intercalating into hydrophobic grooves present in the fibril structure.³⁷

The diversity in fluorescence observed for large 6T-containing spherulites, demonstrates that 6T is not homogeneously distributed in the spherulite, with an interior region comprising 6T dispersed in the protein matrix, and an outer region where 6T is present in aggregated form.

We also obtained a slice from a merged spherulite complex as shown in Fig. 3f and g, which was investigated by means of FLIM and fluorescence microscopy. The fluorescence properties of 6T in the merged structure were similar to the ones obtained for isolated spherulites. Interestingly, 6T is enriched in the region where the two spherulites contact each other. Upon drying, the decrease in diameter, as well as change of mass of individual spherulites, before and after drying, showed that the water content in 6T-containing spherulites was roughly 50% of the total volume (for an example of a SEM-image of a dried spherulite, see Fig. S11†). The interior parts of the spherulite thus contains large amount of water. Therefore, with respect to hydrophobicity there is no particular thermodynamic advantage for 6T to remain within the interior part of the sphere. The exception is 6T that has been intercalated into the β-sheet framework. In contrast, 6T that has not undergone intercalation into the β-sheet framework may undergo migration to the spherulite surface, where it can form semi-crystalline aggregates.

We have noted that the presence of 6T also influences the nucleation process during spherulite growth. Generally we observe fewer and larger spherulites in the presence of 6T. The hydrophobic surface of 6T crystals may promote denaturation of the protein as well as formation of amorphous protein nuclei that act as nucleation points for spherulite growth. This process occurs in competition with amyloid fibril growth. The effect of the presence of 6T on the starting material for the reaction was investigated by Dynamic Light Scattering (DLS). For a mixture of 6T and insulin, particles with a size close to an estimated radius of 60 nm dominate the solution. It is likely that 6T is present in the solution in the form of clusters encapsulated by surrounding insulin molecules. In contrast, the solution of only insulin has a dominant estimated particle radius of about 1.5 nm, as expected for a solution of insulin in an acidic environment (Fig. 4).⁴³

Fig. 4 DLS data. (a) Correlation curves of insulin only (solid line) and 6T–insulin (dashed line). (b) Calculated size distributions corresponding to the correlation curves in (a).

The various data leads us to propose the mechanism outlined in Fig. 5 for the formation of the macroscopic spherulites. During grinding of 6T and insulin, 6T is mixed with insulin forming a composite material that can be dissolved in aqueous acid. Upon heating in the acidic medium, insulin unfolds and forms structures rich in β-sheet content, capable of self-assembly into amyloid-like structures. It should be noted that by co-grinding the hydrophobic molecule 6T with insulin, we are able to form a water soluble composite material, where the 6T component is completely insoluble in water. In the aqueous environment, 6T thus has to stay associated with insulin molecules, and is carried along into whatever self-assembled structures that insulin may form, including spherulites. It is thus likely that the formation of spherulites results in structures with an irregular distribution of 6T. The details of spherulite formation in amyloid–protein systems are not well understood even for the insulin-only system.^17,18 The mechanism of spherulite growth for the insulin–6T system is even more complex with several simultaneous pathways. We speculate that the initial stage of 6T–spherulite formation is similar to that for the formation of protein spherulites. The formation mechanism thus likely starts with the formation of an amorphous protein core that can act as a nuclei for spherulite growth. This amorphous core then acts as a platform where fibril-like materials can grow extending from the core in a radial fashion. The reason for the unusually large spherulitic structures may be related to formation of a smaller number of larger amorphous cores. After formation of the amorphous core, spherulites may grow by an isolated growth process by incorporating insulin–6T material from the surrounding solution, as illustrated in Fig. 3a. Moreover, there is the tendency of 6T-containing spherulites to form clusters that merge upon further growth, as is demonstrated by the structures of Fig. 2e and f (see ESI Fig. S4–S8†). The fact that the outer layer of the spherulite is rich in hydrophobic 6T should facilitate this merging process as the establishment of sphere–sphere contacts minimizes the amount of 6T exposed to water. Overall, 6T thus appears to have the effect of a hydrophobic “molecular glue” enabling both formation of large spherulites and merging of clustered spherulites.

Fig. 5 Comparison of insulin-only and 6T-containing reaction pathways. Schematic representation of the protein spherulite formation occurring upon heat treatment of insulin in acidic water. (a) Insulin with 6T (b) insulin-only.

We thus speculate that the macroscopic self-assembly observed in the case of 6T is the result of a delicate balance between molecular structure (shape) – enabling 6T to be dispersed into the amyloid matrix – in combination with a high crystallinity giving a strong driving force for the aggregation of 6T. The fact that 6T–molecules are not attached with covalent bonds to the protein enables dispersion of 6T in the protein matrix. In addition, 6T is highly hydrophobic, promoting aggregation processes in the aqueous solvent. Both these phenomena are likely to be important in promoting the growth of the large 6T–spherulites.

Conclusions

We have demonstrated the dramatic effect hydrophobic additives can have on the self-assembly process of insulin. Under our reaction conditions native insulin forms spherulites with a maximum diameter of 150 μm. Moreover, such spherulites does not undergo merging processes. On the other hand, when the hydrophobic material 6T is incorporated as a hydrophobic additive, this has a large effect on spherulite growth. Thus, mm-sized spherulites are formed in the presence of the hydrophobic additive. Moreover, the presence of 6T enables merging growth of spherulites. The results reported herein are of relevance to a large number of fields. Due to their association with a large numbers of diseases, it is important to understand factors influencing the aggregation behavior of proteins. We have demonstrated that hydrophobic substances can have a significant influence on self assembled protein structures. Moreover, the results show that the difference in fluorescence of 6T upon interaction with amyloid-like materials is an intrinsic property of the oligo-thiophene structural unit. Finally, the results reported herein are highly relevant to methodology development, as the method of using hydrophobic additives, incorporated by mixing in the solid state of the building block and the additive, should be applicable in a wide range of self-assembly processes, thereby providing a new opportunity for chemists to prepare novel structures from already known self-assembly systems. The results reported thus address an inherent drawback of the self-assembly approach: the difficulty of forming novel structures from a given building block. Furthermore, the methodology employed herein addresses the need for methodology development in order to functionalize biomaterials with other components, such as conducting or light emitting materials, in order to give the structures various desired properties.^3–5,17,44 Due to the simplicity of the preparative method, a wide range of commercially available materials are possible to use. This is important as the majority of readily available light emitting or semiconducting organic materials at the moment are of the hydrophobic variety. In addition, hydrophobically modified spherulites may have unique properties for applications such as drug delivery.⁴⁵ Further studies on the effect of hydrophobic additives on protein self-assembly, as well as studies on the application of the resulting materials, are currently underway in this laboratory.

Acknowledgements

NS acknowledges financial support from the Swedish Government Strategic Research Area in Materials Science on Functional Materials at Linköping University (Faculty Grant SFO-Mat-LiU # 2009-00971). We acknowledge funding via the Wallenberg Scholar grant to Olle Inganäs from the Knut and Alice Wallenberg foundation. Susanna Lönnqvist is acknowledged for help with microtome cutting and sample preparation, and Peter Nilsson is acknowledged for assistance with FLIM measurements.

References

1. Supramolecular soft matter: applications in materials and organic electronics, ed. T. Nakanishi, Wiley, Hoboken NJ, 2011.
2. M. Kushner and Z. Guan, Angew. Chem., Int. Ed., 2011, 50, 9026–9057.
3. T. P. J. Knowles, T. W. Oppenheim, A. K. Buell, D. Y. Chirgadze and M. E. Welland, Nat. Nanotechnol., 2010, 5, 204–207.
4. O. G. Jones and R. Mezzenga, Soft Matter, 2012, 8, 876–895.
5. D. N. Woolfson and Z. N. Mahmoud, Chem. Soc. Rev., 2010, 39, 3464–3479.
6. D. Chandler, Nature, 2005, 437, 640–647.
7. S. S. Karajanagi, H. Yang, P. Asuri, E. Selllitto, J. S. Dordrik and R. S. Kane, Langmuir, 2006, 22, 1392–1395.
8. W. Kim and M. H. Hecht, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 15824–15829.
9. X. Zhao and S. Zhang, Macromol. Biosci., 2007, 7, 13–22.
10. J. Rodríguez-Hernández, F. Chécot, Y. Gnanou and S. Lecommandaux, Prog. Polym. Sci., 2005, 30, 691–724.
11. J. Pronchik, X. He, J. T. Giurleo and D. S. J. Talaga, J. Am. Chem. Soc., 2010, 132, 9797–9803.
12. V. Slutzky, J. A. Tamada, A. M. Klibanov and R. Langer, Proc. Natl. Acad. Sci. U. S. A., 1991, 88, 9377–9381.
13. M. Groenning, S. Frokjaer and B. Vestergaard, Curr. Protein Pept. Sci., 2009, 10, 509–528.
14. D. F. Waugh, J. Am. Chem. Soc., 1946, 68, 247–250.
15. M. Bouchard, J. Zurdo, E. J. Nettleton, C. M. Dobson and C. V. Robinson, Protein Sci., 2000, 9, 1960–1967.
16. C. M. Dobson, Nature, 2003, 426, 884–890.
17. I. Cherny and E. Gazit, Angew. Chem., Int. Ed., 2008, 47, 4062–4069.
18. M. R. H. Krebs, C. E. MacPhee, A. F. Miller, I. E. Dunlop, C. M. Dobson and A. M. Donald, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 14420–14424.
19. M. R. H. Krebs, E. H. C. Bromley, S. S. Rogers and A. M. Donald, Biophys. J., 2005, 88, 2013–2021.
20. S. S. Rogers, M. R. H. Krebs, E. H. C. Bromley, E. van der Linden and A. M. Donald, Biophys. J., 2006, 90, 1043–1054.
21. K. R. Domike, E. Hardin, D. N. Armstead and A. M. Donald, Eur. Phys. J. E: Soft Matter Biol. Phys., 2009, 29, 173–182.
22. K. R. Domike and A. M. Donald, Biomacromolecules, 2007, 8, 3930–3937.
23. V. Foderà and A. M. Donald, Eur. Phys. J. E: Soft Matter Biol. Phys., 2010, 33, 273–282.
24. M. R. H. Krebs, K. R. Domike, D. Cannon and A. M. Donald, Faraday Discuss., 2008, 139, 265–274.
25. K. R. Domike and A. M. Donald, Int. J. Biol. Macromol., 2009, 44, 301–310.
26. M. R. H. Krebs, E. H. C. Bromley and A. M. Donald, J. Struct. Biol., 2005, 149, 30–37.
27. N. Yagi, N. Ohta, T. Iida and K. Inoue, J. Mol. Biol., 2006, 362, 327–333.
28. J. S. Lee, E. Um, J.-E. Park and C. B. Park, Langmuir, 2008, 24, 7068–7071.
29. M. I. Smith, V. Fodera, J. S. Sharp, C. J. Roberts and A. M. Donald, Colloids Surf., B, 2012, 89, 216–222.
30. V. Foderà, A. Zaccone, M. Lattuada and A. M. Donald, Phys. Rev. Lett., 2013, 111, 108105.
31. M. I. Smith, J. S. Sharp and C. J. Roberts, Soft Matter, 2012, 8, 3751–3755.
32. A. Sukhanova, S. Poly, A. Shemetov, I. Bronstein and I. Nabiev, Biopolymers, 2012, 97, 577–588.
33. C. Exley, E. House, J. F. Collingwood, M. R. Davidson, D. Cannon and A. D. Donald, J. Alzheimer's Dis., 2010, 20, 1159–1165.
34. S. L. James and T. Friscic, Chem. Soc. Rev., 2013, 42, 7494–7496.
35. A. Rizzo, O. Inganäs and N. Solin, Chem.–Eur. J., 2010, 16, 4190–4195.
36. B. V. Andersson, C. Skoglund, K. Uvdal and N. Solin, Biochem. Biophys. Res. Commun., 2012, 419, 682–686.
37. F. G. Bäcklund, J. Wigenius, F. Westerlund, O. Inganäs and N. Solin, J. Mater. Chem. C, 2014, 2, 7811–7822.
38. F. G. Bäcklund and N. Solin, ACS Comb. Sci., 2014, 16, 721–729.
39. E. Da Como, M. A. Loi, M. Murgia, R. Zamboni and M. J. Muccini, J. Am. Chem. Soc., 2006, 128, 4277–4281.
40. M. A. Loi, E. Da Como, F. Dinelli, M. Murgia, R. Zamboni, F. Biscarini and F. Muccini, Nat. Mater., 2005, 4, 81–85.
41. D. B. Carter and K. C. Chou, Neurobiol. Aging, 1998, 19, 37–40.
42. A. K. Schütz, A. Soragoni, S. Hornemann, A. Aguzzi, M. Ernst, A. Böckmann and B. H. Meier, Angew. Chem., Int. Ed., 2011, 50, 5956–5960.
43. L. Hovgaard, H. Jacobs, N. A. Mazer and S. W. Kim, Int. J. Pharm., 1996, 132, 107–113.
44. A. Rizzo, N. Solin, L. J. Lindgren, M. R. Andersson and O. Inganäs, Nano Lett., 2010, 10, 2225–2230.
45. U. Shimanovich, I. Efimov, T. O. Mason, P. Flagmeier, A. K. Buell, A. Gedanken, S. Linse, K. S. Åkerfeldt, C. M. Dobson, D. A. Weitz and T. P. J. Knowles, ACS Nano, 2015, 9, 43–51.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra16144d

---