Formation and stability of food foams and aerated emulsions: Hydrophobins as novel functional ingredients

Current Opinion in Colloid & Interface Science 18 (2013) 292–301

Contents lists available at SciVerse ScienceDirect

Current Opinion in Colloid & Interface Science journal homepage: www.elsevier.com/locate/cocis

Formation and stability of food foams and aerated emulsions: Hydrophobins as novel functional ingredients☆ Ali J. Green a, Karen A. Littlejohn a, Paul Hooley b, Philip W. Cox a,⁎ a b

Chemical Engineering, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK School of Applied Sciences, University of Wolverhampton, Wolverhampton WV1 1LY, UK

a r t i c l e

i n f o

Article history: Received 5 March 2013 Received in revised form 17 April 2013 Accepted 18 April 2013 Available online 25 April 2013 Keywords: Hydrophobin Protein Emulsion Foam Stability Food Hydrocolloid Bioinformatics

a b s t r a c t Foams remain an invaluable part of the food engineer's arsenal. Unfortunately the number of new molecules available to stabilise foams is starting to dwindle. Partially, this is due to the difficulties of finding new species with favourable properties and, in many respects, this trend is led by a commercial need to make food labels ‘green’. Food grade proteins offer a number of potential solutions, as well as some excellent physical properties, when at the air–water interface. This review will use the example of hydrophobins as useful proteins finding applications within the food industry. It will also serve as a case study to examine potential methods to identify other new and potentially useful molecules. © 2013 The Authors. Published by Elsevier Ltd. All rights reserved.

1. Introduction 1.1. Food foams and emulsions Food foams still form a major part of the foodstuffs produced and sold today. Such edible foams range from the head of an English ale to the air cell structure in bread. However, stabilisation of the foams remains non-trivial and is the subject, or co‐subject along with emulsions, of a number of recent reviews [1 ••,2–5]. In the main, this continued interest is driven by the sensory pleasure foamed products can deliver. For example the air in ice‐cream makes it sensorially more interesting than ice–slush alone [6]. However, the development of new products, or product differentiation, places continuing demands upon interfacial designs to give differentiated new structures or provide cheap functionality with a constant demand for exceptional stability. The challenges faced when designing colloidal systems in general, and air filled systems such as foams, in particular, are reasonably familiar and are covered in many reviews, papers

☆ This is an open-access article distributed under the terms of the Creative Commons Attribution–NonCommercial–No Derivative Works License, which permits non-commercial use, distribution, and reproduction in any medium, provided the original author and source are credited. ⁎ Corresponding author at: School of Chemical Engineering, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK. Tel.: +44 121 414 5304. E-mail address: [email protected] (P.W. Cox).

and textbooks. Good examples of this include Dickinson [1 ••] and Campbell & Mougeot [7]. A good starting point to look at foam destabilisation can be found in Dutta et al. [8]. These reviews provide excellent descriptions of the creation, and engineered persistence, of foams and demonstrate the continued interest, in foams, by the food industry. In particular Campbell & Mougeot [7] provide a fascinating introduction into the mechanisms of foamed food manufacture and give an excellent summary of the benefits of using and producing food foams. In a similar manner Damodaran [9] gives a concise but very complete description of proteins at foam interfaces and the mechanisms of foam destabilisation when proteins are used as the stabilisation molecule. This specificity to molecules when describing the general mechanisms of maintaining stability conveniently introduces key questions to the literature on food foams. What is the range and number of molecules available for foam stabilisation? What are the advantages of using one molecule over another? Here there is a divergence from the exact realm of the food scientist or amphiphile chemist into the demands of the food industry, and the limitations set by regulation and the desires of the consumer. These amorphous constraints are incredibly difficult to quantify as, in addition to absolute toxicity, local preference, law and culture constantly serve to ‘move the goal posts’ for the manufacturing community. A familiar example of this is the inclusion of genetically modified (GM) products/ingredients into food. Within North America this meets little or no resistance, while large parts of Europe are vehemently opposed to the use of GM in the food manufacturing chain

1359-0294/$ – see front matter © 2013 The Authors. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.cocis.2013.04.008

A.J. Green et al. / Current Opinion in Colloid & Interface Science 18 (2013) 292–301

[10]. Thus for the food technologist/engineer there is a limited palette of molecules with which they can stabilise foamed foods and new individual molecules, let alone classes of molecules, are very difficult to identify. Rodriguez Patino et al. [2,11], Murray et al. [12] and Dickinson [1••] have exhaustively covered the practicalities of using the common and more recently introduced molecules/mechanisms. In essence, though, these four excellent papers can be distilled down into having reviewed the various classes of the commonly used molecules. Even when including the ‘new kids on the block’ the list of foam or emulsion forming molecules, acceptable to the food industry, is surprisingly short.

1.2. Choices of molecules The key players within the list of industrially relevant molecules are the low molecular weight emulsifiers and foaming agents, such as the very common Tweens, the mono- and di‐glycerides, proteins, especially milk derived proteins [13] and quite frequently fragments from digested soy or otherwise non‐functional proteins [14]; fat crystals [6] are also used along with polysaccharide based hydrocolloids that are also becoming more common. Although it should be said that hydrocolloids are tending to be used to bridge the gap between the original list and that of the newer molecules introduced below [15]. The ‘new kids on the block’, as it were, now include Pickering particles [16,17] and mixtures from the above list. Once again each of these molecules is part of a relatively short, specific and food acceptable list, which has received a considerable amount of attention in the press. Therefore they do not need to be revisited again in individual detail. However, if we take our very loose, but industrially relevant, criteria we can possibly identify some trends in the application of each class of molecules to foamed systems. The small molecular weight chemicals are exactly that: chemicals; and a current thrust for many food manufacturers is for the ‘Green Labelling’ of products. Indeed, the ability to reduce the chemical load on a food's label, foamed or not, is currently keeping many university and company laboratories active at present. Similarly, the use of fat crystals may be limited for many products beyond the indulgent commodities such as ice-cream and chocolate due to the health implications in their use. Commodities with enhanced levels of fat, as well as salt and sugar, despite potential functionality may well be difficult to market in an ever more obesity aware context. Hydrocolloids help form the bridge between the traditional and the new. Of all the hydrocolloids it is probably gelatin and ovalbumin that form the most familiar foams. Here hydrocolloids are taken to be biopolymers suspended in water and to conveniently encompass a number of molecular classes. Indeed, a special place is reserved for gelatin in the food engineer's palette due to its all but unique physical properties, such as its mouth relevant melting point and innate elasticity [18]. Unfortunately, despite its physical advantages; gelatin, and to a lesser extent albumin [19], are losing favour with consumers due to their animal origin. Also the memory of the potential implications of BSE has, along with the consumer desire for ‘free-range’, seen a drive to find replacements. In addition the functional contribution from the majority of gums is minimal, and centred on modifying the drainage rates of foams by affecting viscosity. The interfacial activity of gums can also be limited. This then opens up the continued use of proteins for foam generation and maintenance [20]. Dickinson [15] gives an excellent review, albeit a decade old now, in which the use of hydrocolloids, of all denominations, is described at interfaces. Ostensibly, Dickinson's paper [15] is concerned with hydrocolloids at emulsion interfaces but tries to indicate the opportunity for a survey of new applications. However and almost without exception the molecules within this review have not been added to significantly since the review was originally written, and this is due to the conservative nature of the food industry.

293

Despite ‘first’ being described by Pickering in 1907, Pickering particles are very much the newcomers to the foam arena, by design if not just serendipitous practice [6,16,21]. The basic physics of the action of particles at interfaces and its extension into foams is explored by Binks [17]. From this work the principle of particle size and energy of attachment should be borne in mind when considering particulate stabilisers. Principally the lag in industrial uptake of an ostensibly excellent method of foam control has been because of the lack of enough plausible, edible, particles of the right size for food manufacture. This is beyond the use of oil droplets in, for example, creams and ices. As an area of research, particle stabilisation is, nevertheless, moving forward rapidly never the less. An increasing number of potential Pickering particles are starting to be proposed [16]. Despite a number of plausible, truly edible, particles [22], including Janus particles [23], being proposed, the lag between discovery and getting these particles into the industry still exists. Principally this is due to the difficulties in the manufacturing of particles [16]. Indeed, the final statement in Dickinson's 2010 [16] review is: “The preparation of cheap and effective colloidal particles based on food-grade ingredients, especially proteins, is the key technological challenge”. The implications of this will be expanded upon in the remainder of this review. This view was also shared and supported by Murray et al. [24] when considering bubble stability as well as formation.

1.3. Mixed foaming agents Lastly the short list of foaming molecules within the food-processing arena includes the use of mixed foaming agents. This approach has already been touched upon when discussing protein/biopolymer mixtures. Recently a number of papers have appeared which consider the use of emulsifiers with other particles as a new mechanism for emulsion and foam stabilisation. As an excellent example, Murray et al. [12] demonstrated the use of particles and emulsifiers in foam situations. Again a separate review of this approach is fast becoming a necessity. What has become clear through a couple of remarkable papers is that, when using mixed systems, care should be taken, due to the fact that unless designed well these systems can cause particular problems. For instance Pichot et al. [25], for emulsions, and Mackie et al. [26], for foams, showed how the competitive adsorption of the components could change or detrimentally alter the form and function of the stabilising system. Interestingly, the understanding of the control of colloidal structures by the food industry is also finding a place in other disciplines such as Pharmaceuticals [27].

1.4. Criteria for molecule selection Throughout this introduction a number of excellent reviews have been cited that provide a breadth and depth of information concerning almost all aspects of food foam formation and stability. In an effort to provide a level of differentiation for this work, a number of themes have been identified. There remains a need to identify new stabilising agents for food foams within the industry and these molecules and structures need to be accepted by the consumer. The new structures must be robust enough for manufacturing and other physical processes, be adaptable enough for mixed emulsifier situations and have a prospect of future, large scale and economic production. These criteria can all be associated with using proteins [28]. The potential for wholly new molecules is most likely to see proteins as the answer as well. This may be a bold statement; however, the remaining thrust of this article hopes to show that it is a defendable position. It is intended to illustrate the ability for proteins to meet each of the criteria raised above. The prime example that will be used to achieve this will be hydrophobins, a useful class of small amphiphilic proteins found within filamentous fungi [29].

294

A.J. Green et al. / Current Opinion in Colloid & Interface Science 18 (2013) 292–301

2. Hydrophobins as a case study 2.1. Hydrophobins Citing a single molecular class to illustrate the breadth of potential that proteins offer as a whole, may again be bold, but it is well justified by the level of recent interest these molecules have received. Hydrophobins are ideal as a case study for foams and emulsions, over other small proteins used for emulsion and foam stability, because of their highly surface active nature, and self assembling properties, thus far unique to the hydrophobin protein class. Even tiny concentrations of aqueous hydrophobins can quickly saturate an air–water interface, and then non-covalently form a strong, elastic film that is both strong and dynamic. Fig. 1 shows micrographs from Cox et al. [30] and Tchuenbou-Magaia et al. [31]. The two images show the ability of hydrophobins to form well defined, persistent, air cells (the definition of foams will be challenged later) across a large difference in length scale and therefore the mechanism of formation and stabilisation can cope with highly varied internal Laplace pressures, manufacturing principles and longevity. These two structures provide a foundation for the remainder of this document as an extended case study; to see if proteins can indeed answer all of the criteria derived above. Firstly, this class of protein certainly represents a new entry onto the palette of the food engineer and their self-assembling nature suggests a reduced cost of manufacture, certainly in contrast to making Janus particles. This is because their fungal origin appears to be receiving immediate acceptance from the consumer, the scientific community and even the popular press [Fig. 2] [32 •,30]. At the moment a number of papers are appearing that show that these molecules exhibit a variety of interesting characteristics when mixed with other molecules or structures [33]. Finally, a potential for mass production is required, for new molecules to be industrially relevant. There is evidence of industrial production for these molecules having started, which might be applicable for food. Thus, hydrophobins are now starting to reveal a great variability within their structure and even greater potential as new surface active molecules [34]. The structures seen in Fig. 1 have been shown to have remarkable physical properties and stability. To help show, on a molecular level, why the hydrophobin molecules are able to provide such remarkable structures and functionality, their action at interfaces will be considered [35 •]. In a broader context this will be compared to the action of most other proteins, particularly the mechanisms of their absorption. Mixed systems will then be examined, both in the context of diffusion to interfaces and the way this may affect future manufacturing processes. Finally and most uniquely, the use of bioinformatics will be introduced as a tool to identify new potentially technologically useful molecules. In addition to this new paradigm, bioinformatics can also relate mechanisms to three dimensional structure, which is often the rate limiting step in identifying new molecules [34,36,37••].

Fig. 2. A short article from “The Sun” news paper (UK) 25/3/2008.

2.2. Film formation Hydrophobins are secreted in a monomeric form, but once in solution they form dimers and in some cases tetramers [38–41]. The tetramers are usually seen at higher concentrations [39,42•] although they have been observed throughout the concentration range [43 •]. These oligomers help to protect the hydrophobins' hydrophobic regions from contact with the polar substrate [44] sufficiently that no further aggregation is formed, even at concentrations as high as 100 mg/ml [42•,45], but not so far that film formation is not preferable. As can be seen from ‘Fig. 1’ these molecules form films at interfaces very readily. It is argued, particularly for Class II hydrophobins, that the tetramers formed in solution are an intrinsic part of the self assembly process that hydrophobins undergo when exposed to a hydrophobic/ hydrophilic interface [39,42 •,44]. The ability of hydrophobins to self assemble into films at a hydrophobic/hydrophilic interface is what makes hydrophobins of such great interest to those looking at the problem of stabilising food foams and emulsions as the processing of fine structures can be simplified. Being very small proteins they

Fig. 1. Showing approximately 150 μm air cells (A) and approximately 2 μm air cells (B) stabilised by hydrophobin protein HFBII [31,30] (with permission).

A.J. Green et al. / Current Opinion in Colloid & Interface Science 18 (2013) 292–301

diffuse to a surface relatively fast [31,46 •], combined with their highly amphiphilic nature, this makes them the most highly surface active proteins known [47,48]. Their tight core, stabilised by disulphide bonds, prevents the proteins from either; folding their hydrophobic regions away from the polar substrate, or unfolding at the interface [32 •]. The self assembling nature of hydrophobins can be split into two processes; firstly, concentrating at the hydrophobic/hydrophilic interface, and secondly, formation into elastic films. The concentration step has been shown to be primarily driven by the surface tension of the interface [49–51]. While there is no absolute understanding of the cause of the elastic film formation, Basheva et al. [52 ••] give two possible explanations; one being the interactions of polar side groups, the second being that there are hydrophobic interactions between the hydrophobic patches within the hydrophilic region of the hydrophobin. The latter is the more likely option due to the consistency of film formation throughout the full pH range [52 ••]. This implies that, in contrast to other protein systems, hydrophobins diffuse rapidly to interfaces, and once there hold a globular shape while also self associating into films. This behaviour is more like that of small molecular emulsifiers, rather than that of a globular protein. This self assembly potential also gives an interface which, remarkably, can protect from coalescence and Ostwald ripening [8,31]. 2.3. Key characteristics adding stability to emulsions and foams When talking about emulsions and foams, especially when stabilised by hydrophobins, it is important to note that the boundary between the two terms is no longer easily defined [Table 1]. It is impossible to describe the stability characteristics of one without touching on the other. The line becomes yet more blurred when discussing air filled emulsions (AFEs), a term introduced by the Birmingham group, and

Table 1 Table defining several words and phrases used throughout the review. Word/Acronym Emulsion Foam

Definition

A colloidal suspension of one liquid in another liquid. Small, frothy bubbles formed in or on the surface of a liquid, as from fermentation or shaking. or A colloid in which particles of a gas are dispersed throughout a liquid. Air filled emulsion A colloidal suspension of small, protein coated, air cells (AFE) dispersed in water. Cell A small particle, drop or bubble of fluid. Hydrophobin (HF) Short chain amphiphilic proteins, with a conserved eight cysteine motif. Class I Form highly insoluble, elastic, rodlet layers when at hydrophilic/hydrophobic interfaces. Class II Form solubility resistant, rodlet free, elastic films when at hydrophilic/hydrophobic interfaces. Monomer A single sub unit. Oligomer A grouping of an unspecific number of sub units. Dimer A pair of (2) associated sub units. Tetramer A quad of (4) associated sub units. Floc/flocculation A collection or grouping of cells. Aggregate A collection or mass of solid particles. Coalesce When two, or more, cells come into contact in such a way that they become one larger cell. Duplex or triple Where an emulsion is itself emulsified in such a way that there emulsion are smaller cells within the larger dispersed cells. e.g. Water cells within oil cells within a continuous water phase. Triphasic Where two different emulsions are mixed such that there are emulsion two different types of dispersed cells. e.g. air and oil cells, both within a continuous water phase. Coacervation The process of liquid separation into two distinct phases, each containing a different concentration of materials. Gel emulsion A gel formed under shear such that the gel forms into small ‘hairy’ particles within a less concentrated gel solution.

295

foams [Fig. 1]. These are both air and water systems, and there is no one point at which an air filled emulsion becomes a foam, as the difference between them is only the air cell size and the air to water ratio, which are frequently interdependent. Hydrophobins are uniquely suited to the production of emulsions and foams [48, 53]. They not only reduce the surface tension [32 •,31,54,55], allowing for smaller air or oil cells to be produced, but also coat the surface quickly [31,46 •], thus producing a buffer against coalescence. Hydrophobins also bond to one another, producing an elastic membrane at the interface [32 •,30]. This not only creates a barrier to oil or air particles escaping, but also adds a rigidity to the cell that prevents distortion and weakening of the cell and the loss of the hydrophobin from the surface. Alexandrov et al. [35 •] have shown that the elasticity of the hydrophobin coat is caused by a solidification of the hydrophobin protein film, brought about at a surface tension of about 50 mN/m (a surface pressure of around 22 mN/m), which has also been modelled by Cheung [56]. They have also shown that the elastic membrane, when stretched, will ‘relax’ indicating an increase in the surface area covered by the hydrophobin membrane, thus showing that the membrane is dynamic and changeable. It is this dynamism that may explain the longevity and robustness of foams and emulsions stabilised by hydrophobin proteins. This aspect also appears to help minimise the effect of some of the manufacturing issues, such as the coat being knocked from the interface, seen with other proteins. It has been noted by several authors that the adsorption of hydrophobins as they assemble at the interface may be too rapid for controlled use in current industrial process equipment and protein may be lost as inactive oligomerised particles [49,50 •]. However, as it is thought that their self assembly is driven primarily by surface tension [44,49,50•,57], if the surface tension is reduced, the speed at which the hydrophobins assemble at the interface is reduced [49,50 •]. This combined with the possibility of mixing several hydrophobins together or adding other contaminating proteins [58,59••], such as whey [33] or β-casein [60 •], and also controlling the pH around the pI [31,46 •], could allow for high levels of control to a process involving the formation of hydrophobin films. This has the potential to allow for the ideal rate constant to be achieved. Ideally formation should not occur so fast that the hydrophobins form on interfaces such as those produced during cavitations, and are lost as hollow aggregates, but fast enough that air cells do not have a chance to coalesce. Once made, a hydrophobin protein coat, when applied to an air filled emulsion (AFE) or a familiar oil and water emulsion, shows exceptional stability for upwards of four months, at ambient conditions, with little or no deterioration [30,31,32 •,61]. Thus another key requirement, for the food industry, has been realised. A hydrophobin stabilised emulsion would be stable comfortably within the time frame of a food's manufacturing and distribution time window. Determining the physical activities controlling these processes will be of immense interest in the future. This will be particularly true for food engineers as they strive to match the length scales of the products they wish to manufacture (foams, emulsions or AFEs) with the timescales of the adsorption of hydrophobins [31,39,59 ••]. For example, Wohlleben et al. [62] added 1 ppm of a Class I hydrophobin to a 300 ppm standard surfactant solution and noted the ability of the buffer/oil solution to maintain an emulsification state after 24 h. This was measured by the amount of oil that had separated out of the emulsion. This decreased from 100% to 0%. It was also noted that the addition of the hydrophobin protein again stabilised the interface against coalescence and phase separation via the enhancement of the interfacial elastic modulus. Also Tchuenbou-Magaia et al. [31] described a general method for the making of air filled emulsions. This method, although developed for use with hydrophobins, has successfully been used, with minor alterations, by this group for other cysteine rich surface active proteins such as egg white and bovine serum albumin [58]. Both of these suspensions are stabilised, like those of hydrophobins

296

A.J. Green et al. / Current Opinion in Colloid & Interface Science 18 (2013) 292–301

at the correct length scales for the production and distribution processes required within the food industry [63]. Most protein interfaces can be destabilised by competitive adsorption, particularly by low molecular weight surfactants [26,64 •]. Thus far hydrophobin interfaces appear to be resistant to this form of destabilisation and to the presence of oil, which is normally an efficient anti-foaming agent [24, 31]. Wohlleben et al. [62] describe the early question as to why hydrophobins have exceptional surface properties while they themselves are physically quite inert. The elastic membrane is stabilised via non-covalent inter-molecular bonds [46 •,52 ••,65], allowing for the hydrophobins to be highly stable, but, given the right conditions, re-solubilised and re-used with no ill effects [44, 47, 54] due to the unique film-forming mechanism discussed earlier. Interestingly, due to their rigid tertiary structure and amphiphilic nature, hydrophobins have been described as small molecular emulsifiers [16] and likened to Janus particles [66]. ‘Fig. 1’ shows the different length scales to which hydrophobins have been used. However, what also requires attention is the time scale of adsorption and how these match to potential unit operations in industry. Cox et al. [32 •] and Tchuenbou-Magaia et al. [31] use very different mechanisms, for example, in Cox et al. [32 •] whipping of the proteins was used, a process with relatively long time constants for adsorption. Whereas, in Tchuenbou-Magaia et al. [31] sonication was used. This was a more novel approach but again the kinetics of the protein adsorption could be tailored for the process. Cox & Hooley [29], however, pointed out that care should be taken to prevent partial formation, as these ‘plaques’ if swept from the interface might be rendered inactive and thus the protein wasted. Due to the small size of hydrophobins, they will not distort the surface textural properties of the dispersed phase, therefore maintaining traditional mouth feel, taste and flavour in food based emulsions [31,63]. Fortunately, as introduced above, hydrophobins are acceptable to the food production industry. This is due to their abundance in fungi, which have been eaten safely by humanity for millennia [34,54]. Remarkably fungi use hydrophobins to avoid detection by a host cell's immune response [67], this makes hydrophobins likely inert agents for drug delivery. Emulsions and, particularly, foams are susceptible to film drainage, or creaming, where the liquid flows between the dispersed cells, leaving the cells to be forced closer together, eventually forming a thin film [9]. This film then gets steadily thinner until it breaks and the two adjacent cells coalesce into one larger cell. However, hydrophobins have the ability to form a bi-layer once two films come together. This bi-layer limits further drainage, and is strong enough to be significantly more stable, even to relatively high shear environments [52 ••], than a comparable Tween stabilised interface. Although not related directly to food foams' and emulsions' stability, hydrophobins also have the ability to reverse the hydrophobicity of a surface [54,55], allowing for surfaces such as steel and glass to be wetted more easily, reducing bubbles while filling a tank, which could prove useful to some sensitive or particularly fast processes. This could be one of the reasons that Reger et al. [68, 69] were able to produce Pickering emulsions by combining solid particles, such as Boehmite and silica based clay, with hydrophobins to stabilise mono-dispersed oil in water emulsions that were stable for months. They were also able to show that the emulsification properties of the combined particles were more effective than the individual components. The use of hydrophobins to stabilise both the smaller dispersed phase and the foam could then lead to further stabilisation, with the smaller cells of the dispersed phase drawn into the junctures between the larger air cells of the foam where the hydrophobin bi-layers would hold flocs of cells together, as well as in place [Fig. 3]. With the help of various innovative methods hydrophobins have been shown to be far more numerous, and their classification far more complex, than first thought [34,36]. As well as identifying how many hydrophobin proteins are produced by a single fungal genus,

these methods also have the potential to describe the characteristics of these proteins; where to find them, their physical properties in solution, their interfacial properties and even the likelihood of unfolding or cross linking. This type of information can help direct the search for an “ideal” protein for any particular foam or emulsion system. The approaches described briefly here, and expanded upon below, are used to identify this potential. The relatively small size and the specificity of conserved regions, centred on four pairs of cysteines, within the hydrophobins, make them the ideal vehicle with which to start this form of directed discovery. 2.4. Extraction and expression It is proteins that potentially hold the largest reservoir of new molecules with the highest likelihood to penetrate the food industry. However, identifying potential candidates for commercial use can be difficult. Once again hydrophobins are to be used as a case study as an emerging application in the food industry [32 •,30]. This process started with biologists first noticing proteins with interesting properties [47], which were then used in innovative technical applications [32 •,31,30] and this has led to the high level of current activity in the search for new surface active forms [34,36]. It has to be said that it is not all plain sailing for the hydrophobin story. Several ‘icebergs’ still potentially lie ahead for the technology. One is the legal position; as either a novel product or classification as a novel use/concentration must be established. This will not be discussed here. However, other potential problems include the processing of the molecules, such as hydrophobins, are challenging molecules to work with due to their affinity for interfaces. Once hydrophobins have been expressed the extraction process currently uses strong acids; such as TFA (trifluoroacetic acid) [31,65,70,71] to isolate monomeric proteins. This process is inherently hazardous and therefore preferable to keep separated from food processes. It is also still difficult to extract and isolate some hydrophobins, even with the use of TFA. This lack of process knowledge stems from the incredibly small number of hydrophobins that have been characterised to date. This leaves the generally accepted method for the extraction of hydrophobins from fungi which has been described using TFA or SDS (sodium dodecyl sulphate). Thus far it has been noted that this method, despite the excitement over the hydrophobins' properties, has not been optimised for the food industry. To ensure that there is no contamination in the final food product [31,38,48] it is desirable for a substitute to the TFA extraction process to be found, or for a more easily extracted hydrophobin or hydrophobinlike protein to be discovered, before hydrophobins can be safely used in foods destined for consumption by humans. One such potential route has been touched upon in the use of Tween to ‘lift’ hydrophobins from Polytetrafluoroethylene (PTFE) surfaces [72,73]. It is a common aim to want continuous or larger batches of production, and indeed it has been reported that a large chemical company has scaled up production to kilogramme quantities of hydrophobins [62]. Elsewhere only very small batches can be produced. This is due, in part, to the difficulty with the process and the high energy costs involved. It has also been expressed that the high surface activity and speed with which hydrophobins assemble at an interface would render the majority of current processing equipment, which tends to have relatively long residence times or high surface areas, unacceptable for hydrophobin production [32 •]. Due to the excitement around hydrophobin proteins, allied with the difficulties of production, there have been many cloning methods described using a variety of expression vectors. These approaches seek to isolate the hydrophobin encoding gene and place it into a more amenable host, allowing direct control over expression times and quantities. These experiments include DewA, an Aspergillus hydrophobin cloned into a Trichoderma vector [74]. There are two main successful systems described, firstly, the cloning of a gene encoding a hydrophobin,

A.J. Green et al. / Current Opinion in Colloid & Interface Science 18 (2013) 292–301

297

Fig. 3. A diagram of the drainage of liquid from between foam air cells.

perhaps as a fusion protein [62, 75], into Escherichia coli [76]; secondly, the cloning of genes encoding hydrophobin proteins into the commercially available expression system Pichia pastoris [77,78], allowing efficient extracellular secretion. For more information on the Pichia expression system versus the E. coli system see the review by Daly & Hearn [79]. Hydrophobins are relatively short molecules of 80 to 100 amino acids and are therefore encoded by relatively short lengths of DNA, opening the possibility of artificially synthesising DNA constructs that encode a chosen molecule directly from a computer database sequence. Artificial gene synthesis can be remarkably cheap — for example, an entire cloned hydrophobin construct can be provided for around $100 (US) (Geneart, Life Technologies). Stübner et al. [77] described this approach to optimise codon usage for expression of hydrophobins using a simple inducible promoter system in the yeast Pichia. A simple induction with a low concentration of methanol instructs the host cell to over-express the hydrophobin and secrete it efficiently. This system can scale up quickly to a pilot or commercial scale, thus rapidly expanding the potential number of hydrophobins available for food applications and allowing for the modification of proteins to produce specific characteristics. This general approach should be also applicable to other small molecules too, whatever their original host organism.

significant overlap [Fig. 4] where almost 25% of the Class IIs and almost 10% of Class Is are either above or below the 100aa boundary. A variety of programmes, available online, allow the prediction of secondary structures of proteins directly from the primary sequence of amino acids (for example, PSIPRED, http://bioinf.cs.ucl.ac.uk/psipred/ [82]). Primary sequences are usually predicted directly from translations of particular DNA sequences without any direct characterisation of a purified protein itself. Such programmes can predict structures such as beta sheets and alpha helices. It is far more difficult to derive tertiary structures or protein folds from such genomics data [83]. Significant similarity at a primary structure level is assumed to predict similar tertiary structure. Only a small number of models for hydrophobin tertiary structures are available and predicting the function and 3D structure can remain difficult. Another major issue with the classification of hydrophobins is that it was established using so few proteins, only two of which are

2.5. Classification Industrial interest has been piqued by the recent research into the classification of hydrophobins. Until 2010 it was thought that hydrophobins fit into two distinct classes [34,36]. This traditional system for the classification of hydrophobins is said to be based on both the solubility of the aggregates and the hydrophobicity profiles of the proteins [80,81]. However it has been shown that the hydrophobicity profiles for hydrophobins are not as clear cut as previously thought [34,36]. It has also been generally accepted that Class IIs are smaller than Class Is and although this is generally the case, there is a

Fig. 4. The size distribution of the currently identified Class I and Class II hydrophobins.

298

A.J. Green et al. / Current Opinion in Colloid & Interface Science 18 (2013) 292–301

Class IIs from the same fungus [84]. Even now only about half a dozen Class I hydrophobins have been used in many experiments [SC3 — [40,85], HGFI — [46 •], Vmh2 — [86], EAS — [49,87], DewA — [49,88], RodA — [89]] while that number drops to only two for Class II hydrophobins [HFBI — [32 •,42 •,46 •,57] and HFBII — [32 •,31,42 •,44,50 •,52 ••]] which is probably only about 1–2% of the potential hydrophobins given the hundreds of proteins found via bioinformatics in just one genus [34,36,37 ••]. Littlejohn et al. [34] and Jensen et al. [36] introduced the concept that it is desirable to establish a classification system that can explain the full potential and subtle differences found within the hydrophobin proteins. This was after establishing that there are Aspergillus proteins that cannot be classified easily into either of the two current classes. It is the opinion of the authors here that there may be as many as five hydrophobin classes, with extra classes both between and extending beyond the scope of the current class system. It also appears, that the differences between classes occur more on a sliding and overlapping scale. Again this presents the food engineer with a potential for future growth in target molecules if they can be patient for the science to catch up.

2.6. Genomics in the identification and characterisation of novel molecules Hydrophobins have a distinctive eight cysteine motif (X-C-X-CCX-C-X-C-X-CC-X-C-X), forming 4 disulfide bonds (C1–C6, C2–C5, C3–C4 & C7–C8) [36,81] which have already been implicated in the unique properties of these proteins. It is this motif that allows hydrophobins to be identified. The cysteines bridge to form four loops with variable lengths [81], giving a final tertiary structure showing a four leaf clover shape with hydrophobic and hydrophilic ‘leaves’ [90]. Although the N-terminus, C-terminus and C3–C4 loop can all be altered, with no effect to functionality, the C7–C8 loop is integral to the formation of rodlets in the traditional Class I hydrophobin [57,91,92]. By using modern genome projects the food engineer has unparalleled opportunities to identify novel molecules. Automated DNA sequencing coupled to rapid data analysis allows the use of the internet to interrogate publically available data. Bioinformatics techniques employ computer algorithms to identify genes, and their encoded products, from raw DNA sequence data. Many databases are already well annotated with simple keyword searches available for chosen classes of gene or protein. Jensen et al. [36] and Littlejohn et al. [34] demonstrate how even the well-established genome databases for the Aspergillus species can yield evidence for new hydrophobins. Crucially, it is possible to carry out initial categorisation and characterisation of molecules in silico, for example via molecular mass and pI predictions along with hydrophobicity plots. All of which is applicable to predictions of foam stability. Phylogenetics can then be used to quickly identify clusters of related proteins and so focus on those proteins with novel properties or, conversely, identify relatives of molecules with known, desirable features. Hence, a rational selection of target molecules may be made before expensive and lengthy laboratory experimentation is undertaken. This is an approach that has been widely used in the pharmaceutical industry for some time, both to identify and target drug molecules [67,93,94] but is only now finding applications in food science.

2.7. Practical genomics and bioinformatics The potential discussed above can be realised via a number of resources. The Kegg database (http://www.genome.jp/kegg/) shows that the number of genome projects being completed is steadily increasing year on year. As of February 28th 2013 a total of 180 eukaryote genomes, 149 Achaea and 2143 bacteria genomes have been completed. The same bioinformatics approaches used in the current hydrophobin example, employing genome databases to identify desirable molecules with similarities to ideal sequences, can be applied to find other proteins with potential for foam formation. It must be noted however that there are still some significant errors inherent with this type of data [95–97]. The few papers written on the identification of novel food proteins by the use of DNA sequence data from genome projects [34,36,37 ••] are quoted extensively here. However, there are several differences between the criteria used for the identification of a probable hydrophobin [98]. The most promising identification method is the use of similarity search engines, such as BLASTp (for proteins) and BLASTn (for nucleotides) [34,36,37 ••,99], followed by either manual annotation [34] or the use of further automated programmes that search for particular consensus sequences [36,37 ••]. All authors have noted the difficulty automated search programmes have had in the identification of novel hydrophobins, as the overall similarity across the length of the proteins, excluding the distinctive eight cysteine motif, is limited [Figs. 5 & 6]. The more automation used in each procedure, the higher the likelihood of a novel target protein being missed. For an example of these approaches Fig. 5, shows the high similarity of the cysteine residues, and the cooler colours of Fig. 6 show the complete sequence, emphasising the dissimilarity throughout the protein as a whole and, hence, the difficulty that is faced by the automated search programmes. As a final point, the recent (2013) solution NMR structure of the DewA protein of Aspergillus nidulans, recently published by Morris et al. [100], is only the second Class I structure to be published, after EAS of Neurospora crassa. DewA shows unusual arrangements of secondary structure that allows dimer as well as fibril formation. There is an urgent need to resolve more tertiary structures of hydrophobins to truly appreciate the diversity of molecules available to engineers as predicted by bioinformatics [34].

3. Conclusion This review has attempted to highlight several important areas: Primarily, it has hoped to show that foamed structures, and the molecules that stabilise them, are still highly relevant to the food industry. A list of new and emerging foam stabilisers has been given and explored. Central to this list was the use of proteins within the food industry. The remainder of the review used hydrophobins, as a case study, to chart the entry of a novel protein from being a scientific curiosity to a plausible ingredient worthy of further exploitation by the food industry. The other important area, this review has tried to address, is the likely future of food foams and emulsions. In doing so we have tried to introduce new classes of molecules, into which the food industry

Fig. 5. An example of the use of a multiple sequence alignment TCoffee (Notre dame et al. [101]) to confirm the identity of a novel hydrophobin. Hydrophobin sequences taken from the Aspergillus comparative database matched to ‘Afu2g17340’, previously unidentified as a hydrophobin showing similarity around the conserved cysteine motif. Adapted from sequences identified in Littlejohn et al. [34].

A.J. Green et al. / Current Opinion in Colloid & Interface Science 18 (2013) 292–301

Fig. 6. An example of the use of a multiple sequence alignment TCoffee (Notre dame et al. [101]) to confirm the identity of a novel hydrophobin. Hydrophobin sequences taken from the Aspergillus comparative database matched to ‘Afu2g17340’, previously unidentified as a hydrophobin but showing reduced similarity across the complete sequence as would be found on the genome database. Adapted from sequences identified in Littlejohn et al. [34].

299

300

A.J. Green et al. / Current Opinion in Colloid & Interface Science 18 (2013) 292–301

could tap. The hydrophobin example has been extrapolated with the use of bioinformatics, to indicate tertiary structure and function, in an attempt to illustrate the full potential of proteins in general. Acknowledgements The authors would like to thank Miss K. Preece for her contribution throughout the writing of this review. References and recommended reading •, •• [1] Dickinson E. Stabilising emulsion‐based colloidal structures with mixed food •• ingredients. J Sci Food Agr 2012.An excellent review that encapsulates all major aspects of food colloid science. [2] Rodríguez Patino JM, Carrera Sánchez C, Rodríguez Niño MR. Implications of interfacial characteristics of food foaming agents in foam formulations. Adv Colloid Interface Sci 2008;140:95–113. [3] Kralova I, Sjöblom J. Surfactants used in food industry: a review. J Dispersion Sci Technol 2009;30:1363–83. [4] Dickinson E. Hydrocolloids as emulsifiers and emulsion stabilizers. Food Hydrocoll 2009;23:1473–82. [5] Palzer S. Food structures for nutrition, health and wellness. Trends Food Sci Technol 2009;20:194–200. [6] Crilly JF, Russell AB, Cox AR, Cebula DJ. Designing multiscale structures for desired properties of ice cream. Ind Eng Chem Res 2008;47:6362–7. [7] Campbell GM, Mougeot E. Creation and characterisation of aerated food products. Trends Food Sci Technol 1999;10:283–96. [8] Dutta A, Chengara A, Nikolov AD, Wasan DT, Chen K, Campbell B. Destabilisation of aerated food products: effects of Otswald ripening and gas diffusion. J Food Eng 2004;62:177–84. [9] Damodaran S. Protein stabilization of emulsions and foams. J Food Sci 2005;70: R54–66. [10] Dannenberg A. The dispersion and development of consumer preferences for genetically modified food — a meta-analysis. Ecol Econ 2009;68:2182–92. [11] Rodríguez Patino JM, Dolores Naranjo Delgado M, Linares Fernández J. Stability and mechanical strength of aqueous foams containing food proteins. Colloids Surf A 1995;99:65–78. [12] Murray BS, Durga K, Yusoff A, Stoyanov SD. Stabilization of foams and emulsions by mixtures of surface active food-grade particles and proteins. Food Hydrocoll 2011;25:627–38. [13] Raikos V. Effect of heat treatment on milk protein functionality at emulsion interfaces. A review. Food Hydrocoll 2010;24:259–65. [14] Martínez KD, Sánchez CC, Rodríguez Patino JM, Pilosof AMR. Interfacial and foaming properties of soy protein and their hydrolysates. Food Hydrocoll 2009;23:2149–57. [15] Dickinson E. Hydrocolloids at the interface and the influence on the properties of dispersed systems. Food Hydrocoll 2003;17:25–39. [16] Dickinson E. Food emulsions and foams: stabilization by particles. Curr Opin Colloid Interface Sci 2010;15:40–9. [17] Binks BP. Particles as surfactants — similarities and differences. Curr Opin Colloid Interface Sci 2002;7:21–41. [18] Karim A, Bhat R. Gelatin alternatives for the food industry: recent developments, challenges and prospects. Trends Food Sci Technol 2008;19:644–56. [19] Pernell CW, Foegeding EA, Luck PJ, Davis JP. Properties of whey and egg white protein foams. Colloids Surf A 2002;204:9–21. [20] Miquelim JN, Lannes SCS, Mezzenga R. pH influence on the stability of foams with protein–polysaccharide complexes. Food Hydrocoll 2012;24:398–405. [21] Spyropoulos F, Cox PW, Fryer PJ, Norton IT. Immiscible liquid liquid mixing. Food Mixing 2008:175–98. [22] Murray BS, Ettelaie R. Foam stability: proteins and nanoparticles. Curr Opin Colloid Interface Sci 2004;9:314–20. [23] Perro A, Meunier F, Schmitt V, Ravaine S. Production of large quantities of “Janus” nanoparticles using wax-in-water emulsions. Colloids Surf A 2009;332: 57–62. [24] Murray BS, Dickinson E, Wang Y. Bubble stability in the presence of oil-in-water emulsion droplets: influence of surface shear versus dilational rheology. Food Hydrocoll 2009;23:1198–208. [25] Pichot R, Spyropoulos F, Norton IT. O/W emulsions stabilised by both low molecular weight surfactants and colloidal particles: the effects of surfactant type and concentration. J Colloid Interface Sci 2010;352:128–35. [26] Mackie AR, Gunning AP, Wilde PJ, Morris VJ. Orogenic displacement of protein from the air/water interface by competitive adsorbtion. J Colloid Interface Sci 1999;210:157–66. [27] Beija M, Salvayre R, Lauth-de Veguerie N, Marty J-D. Colloidal systems for drug delivery: from design to therapy. Trends Biotechnol 2012;30:485–96. [28] Boland MJ, Rae AN, Vereijken JM, Meuwissen MPM, Fischer ARH, van Boekel MAJS, et al. The future supply of animal-derived protein for human consumption. Trends Food Sci Technol 2012;29:62–73.

• ••

Of special interest. Of outstanding interest.

[29] Cox PW, Hooley P. Hydrophobins: new prospects for biotechnology. Fungal Biol Rev 2009;23:40–7. [30] Cox AR, Aldred D, Russel A. Exceptional stability of food foams using class II hydrophobin HFBII. Food Hydrocoll 2008;23:366–76. [31] Tchuenbou-Magaia FL, Norton IT, Cox PW. Hydrophobins stabilised air filled emulsions for the food industry. Food Hydrocoll 2009;23:1877–85. [32] Cox AR, Cagnol F, Al Russel, Izzard M. Surface properties of class II hydrophobins • from Trichoderma reesei and influences on bubble stability. Langmuir 2007;23: 7995–8002.Excellent paper on the physical properties of food colloid science. This paper gives a very balanced view of the technology in terms of science and application to food production. [33] Blijdenstein TBJ, Ganzevles RA, De Groot PWN, Stoyanov SD. On the link between surface rheology and foam disproportionation in mixed hydrophobin HFBII and whey protein systems. Colloids Surf A 2013:1–8 [in press]. [34] Littlejohn KA, Hooley P, Cox PW. Bioinformatics predicts diverse Aspergillus hydrophobins with novel properties. Food Hydrocoll 2012;27:503–16. [35] Alevandrov NA, Marinova KG, Gurkov TD, Danov KD, Kralchevshy PA, Stoyanov • SD, et al. Interfacial layers from the protein HFBII hydrophobin: dynamic surface tension, dilation elasticity and relaxation times. J Colloid Interface Sci 2012;376: 296–306.This paper condenses the current understanding of the properties of the hydrophobin's elastic membrane and greatly extends the thinking with regard to the uses of the molecule. Experimental data on formation and time dependant mechanisms of folm properties is also given. [36] Jensen BG, Andersen MR, Pedersen MH, Frisvad JC, Søndergaard I. Hydrophobins from Aspergillus species cannot be clearly divided into two classes. BMC Res Notes 2010;3:344–50. [37] Yang K, Deng Y, Zhang C, Elasri M. Identification of new members of hydrophobin •• family using primary structures analysis. BMC Bioinforma 2006;7:16–23.Pioneering, paper on using bioinformatics to identify potential hydrophobins. [38] Bailey MJ, Askolin S, Norhammer N, Tenkanen M, Linder M, Penttilä M. Process technological effects of deletion and amplification of hydrophobins I and II in transformants of Trichoderma reesei. Appl Microbiol Biotechnol 2002;58: 721–7. [39] Torkkeli M, Serimaa R, Ikkala O, Linder M. Aggregation and self-assembly of hydrophobins from Trichoderma reesei: low-resolution structural models. Biophys J 2002;83:2240–7. [40] Wösten HA, de Vries OM, Wessels JG. Interfacial self-assembly of a fungal hydrophobin into a hydrophobic rodlet layer. Plant Cell Online 1993;5:1567–74. [41] Stroud PA, Goodwin JS, Butko P, Cannon GC, McCormick CL. Experimental evidence for multiple assembled states of Sc3 from Schizophyllum commune. Biomacromol 2003;4:956–67. [42] Paananen A, Vuorimaa E, Torkkeli M, Penttilä M, Kauranen M, Ikkala O, et al. • Structural hierarchy in molecular films of two class II hydrophobins. Biochem 2003;42:5253–8.A good paper from a leading group in the field, looking at the formation of hydrophobin films at the interface. The examination is not only of the macro-scale properties but also the micro-scale formation. [43] Szilvay GR, Nakari-Setälä T, Linder MB. Behaviour of Trichoderma reesei • hydrophobins in solution: interactions, dynamics and multimer formation. Biochem 2006;45:8590–8.This paper importantly covers the pre-formation characteristics of hydrophobins, and presents arguments for how they might relate to the post-formation characteristics. [44] Hakanpää J, Paananen A, Askolin S, Nakari-Setälä T, Parkkinen T, Penttilä W, et al. Atomic resolution structure of the HFBII hydrophobin a self-assembling amphiphile. J Biol Chem 2004;279:534–9. [45] Corvis Y, Brezensinski G, Rink R, Walcarius A, van der Heyden A, Mutelet F, et al. Analytical investigation of the interactions between SC3 hydrophobin and lipid layers: elaborating of nanostructured matrixes for immobilizing redox systems. Analyt Chem 2006;78:4850–64. [46] Wang Z, Lienemann M, Qiau M, Linder MB. Mechanisms of protein adhesion on sur• face films of hydrophobin. Langmuir 2010;26:8491–6.Thorough research demonstrating (indirectly) great future potential for food engineering. This work provides excellent base data for the initial design of a number of technological process and not just food applications. [47] Wösten HA, de Vocht ML. Hydrophobins, the fungal coat unravelled. Biochem et Biophys Biomem 2000;1469:79–86. [48] Linder M. Hydrophobins: proteins that self assemble at interfaces. Curr Opin Colloid Interface Sci 2009;14:356–63. [49] Morris VK, Ren Q, Macindoe I, Kwan AH, Byrne N, Sunde M. Recruitment of class I hydrophobins to the air:water interface initiates a multi-step process of functional amyloid formation. J Biol Chem 2011;286:15955–63. [50] Zhang XL, Penfold J, Thomas RK, Tucker IM, Petkov JT, Bent J, et al. Adsorbtion • behavior of hydrophobin and hydrophobin/surfactant mixtures at the air–water interface. Langmuir 2011;27:11316–23.This paper covers the complexities of the adsorption of mixed systems, introducing well the mechanisms and potential problems this adds to the overall system. [51] Lumsdon SO, Green J, Stieglitz B. Adsorption of hydrophobin proteins at hydrophobic and hydrophilic interfaces. Colloids Surf B 2005;44:172–8. [52] Basheva ES, Kralchevshy PA, Danov KD, Stoyanov SD, Blijdenstein TBJ, Pelan EG, •• et al. Self-assembled bilayers from the protein HFBII hydrophobin: nature of the adhesion energy. Surf 2011;17:22–4.An excellent study on the mechanism of molecular adhesion. Also including some lateral thinking about the basic mechanisms of film formation as well as inter-film adhesion. [53] Sarlin T, Nakari-Setälä T, Linder M, Penttilä M, Haikara A. Fungal hydrophobins as predictors of the gushing activity of malt. J Inst Brew 2005;111:105–11. [54] Wessels JG. Hydrophobins: proteins that change the nature of the fungal surface. Adv Microb Ecol 1996;38:1–45.

A.J. Green et al. / Current Opinion in Colloid & Interface Science 18 (2013) 292–301 [55] Wösten HA, Schuren FH, Wessels JG. Interfacial self-assembly of a hydrophobin into an amphipathic protein membrane mediates fungal attachment to hydrophobic surfaces. EMBO 1994;13:5848. [56] Cheung DL. Molecular simulation of hydrophobin adsorption at an oil–water interface. Langmuir 2012;28:8730–6. [57] Szilvay GR, Paananen A, Laurikainen K, Vuorimaa E, Lemmetyinen H, Peltonen J, et al. Self-assembled hydrophobin protein films at the air–water interface: structural analysis and molecular engineering. Biochem 2007;46:2345–54. [58] Tchuenbou-Magaia FL, Cox PW. Tribological study of suspensions of cysteine-rich protein stabilised microbubbles and subsequent triphasic A/O/W emulsions. J Texture Stud 2011;42:185–96. [59] Askolin S, Linder M, Scholtmeijer K, Tenkanen M, Penttilä M, de Vocht ML, •• et al. Interaction and comparison of a class I hydrophobin from Schizophyllum commune and class II hydrophobins from Trichoderma reesei. Biomacromol 2006;7:1295–301.Possibly the best paper in the field. [60] Burke J, Cox AR, Petkov J, Murray BS. Interfacial rheology and stability of air bub• bles stabilised by mixtures of hydrophobin and β-casein.Food Hydrocoll 2012: 1–9 [in press].This paper now starts to show the use of the molecule to be maturing in the literature and as a potentially industrially useful molecule. [61] Blijdenstein T, de Groot PWN, Stoyanov SD. On the link between foam coarsening and surface rheology: why hydrophobins are so different. Soft Matter 2010;6: 1799–808. [62] Wohlleben W, Subkowski T, Bollschweiler C, Vacano B, Liu Y, Schrepp W, et al. Recombinantly produced hydrophobins from fungal analogues as highly surfaceactive performance proteins. Eur Biophys J 2010;39:457–68. [63] Norton IT, Fryer PJ, Moore S. Product/process integration in food manufacture: engineering sustained health. AIChE J 2006;52:1632–40. [64] Pugnaloni LA, Dickinson E, Ettelaie R, Mackie AR, Wilde PJ. Competitive adsorp• tion of proteins and low-molecular-weight surfactants: computer simulation and microscopic imaging. Adv Colloid Interface Sci 2004;107:27–49.A useful paper for the understanding of the mechanisms of how proteins and emulsifiers interact and respond to the hydrophobic/hydrophilic interface. [65] de Vries OMH, Fekkes MP, Wösten HAB, Wessels JGH. Insoluble hydrophobin complexes in the walls of Schizophyllum commune and other filamenteous fungi. Arch Mircobiol 1993;159:330–5. [66] Walther A, Müller AH. Janus particles. Soft Matter 2008;4:663–8. [67] Aimanianda V, Bayry J, Bozza S, Kniemeyer O, Perruccio K, Elluru S, et al. Surface hydrophobin prevents immune recognition of airborne fungal spores. Nature 2009;460:1117–21. [68] Reger M, Sekine T, Okamoto T, Watanabe K, Hoffmann H. Pickering emulsions stabilized by novel clay–hydrophobin synergism. Soft Matter 2011;7:11021–30. [69] Reger M, Hoffmann H. Hydrophobin coated boehmite nanoparticles stabilizing oil in water emulsions. J Colloid Interface Sci 2012;368:378–86. [70] de Vocht ML, Scholtmeijer K, van der Vegte EW, de Vrier OM, Sonveaux N, Wösten HA, et al. Structural characterization of the hydrophobin SC3, as a monomer and after self-assembly at hydrophobic/hydrophilic interfaces. Biophys J 1998;74:2059–68. [71] Beever RE, Redgwell RJ, Dempsey GP. Purification and characterization of the rodlet layer of Neurospora crassa conidia. J Bacteriol 1979;140:1063–70. [72] de Vocht ML, Reviakine I, Ulrich W-P, Bergsma-Schutter W, Wösten HAB, Vogel H, et al. Self-assembly of the hydrophobin SC3 proceeds via two structural intermediates. Protein Sci 2002;11:1199–205. [73] Scholtmeijer K, de Vocht ML, Rink R, Robillard GT, Wösten HA. Assembly of the fungal SC3 hydrophobin into functional amyloid fibrils depends on its concentration and is promoted by cell wall polysaccharides. J Biol Chem 2009;284:26309–14. [74] Schmoll M, Seibel C, Kotlowski C, Wöllert F, Vendt G, Liebmann B, et al. Recombinant production of an Aspergillus nidulans class I hydrophobin (DewA) in Hypocrea jecorina (Trichoderma reesei) is promoter-dependent. Appl Microbiol Biotechnol 2010;88:95–103. [75] Joensuu J, Conley A, Lienemann M, Brandle MJ, Linder M, Menassa R. Hydrophobin fusions for high-level transient protein expression and purification in Nicotiana benthamiana. Plant Physiol 2010;152:622–33. [76] Wang Z, Feng S, Huang Y, Li S, Xu H, Zhang XL, et al. Expression and characterization of a Grifola frondosa hydrophobin in Pichia pastoris. Protein Expr Purif 2010;72:19–25. [77] Stübner M, Lutterschmid G, Vogel RF, Niessen L. Heterologous expression of the hydrophobin FcHyd5p from Fusarium culmorum in Pichia pastoris and evaluation

[78]

[79]

[80] [81] [82] [83]

[84]

[85]

[86]

[87]

[88] [89]

[90]

[91]

[92]

[93]

[94] [95] [96] [97] [98]

[99] [100]

[101]

301

of its surface activity and contribution to gushing of carbonated beverages. Int J Food Microbiol 2010;141:110–5. Niu B, Wang D, Yang Y, Xu H, Qiao M. Heterologous expression and characterization of the hydrophobin HFBI in Pichia pastoris and evaluation of its contribution to the food industry. Amino Acids 2012;43:763–71. Daily R, Hearn MT. Expression of heterologous proteins in Pichia pastoris: a useful experimental tool in protein engineering and production. J Mol Recognit 2004;18:119–38. Hektor HJ, Scholtmeijer K. Hydrophobins: proteins with potential. Curr Opin Biotechnol 2005;16:434–9. Wessels JGH. Hydrophobins: proteins that change the nature of the fungal surface. Adv Microb Ecol 1997;38:1–45. Jones DM. Protein secondary structure prediction based on position-specific scoring matrices. J Mol Biol 1999;292:195–202. Lobley A, Sadowski MI, Jones DT. pGenTHREADER and pDomTHREADER: new methods for improved protein fold recognition and superfamily discrimination. Bioinforma 2009;25:1761–7. Kubicek C, Baker S, Gamaif C, Kenerley C, Druzhinina I. Purifying selection and birth-and-death evolution in the class II hydrophobin gene families of the ascomycete Trichoderma/Hypocrea. BMC Evol Biol 1998;23:18–33. Scholtmeijer K, Janssen I M, Gerssen B, de Vocht ML, van Leewen BM, van Kooten TG, Wessels JG. Surface modifications created by using engineering hydrophobins. Appl Environ Microbiol 2002;68:1367–73. Houmadi S, Rodriguez RD, Longobardi S, Giardina P, Fauré MC, Giocondo M, et al. Self-assembly of hydrophobin protein rodlets studied with atomic force spectroscopy in dynamic mode. Langmuir 2012;28:2551–7. Kwan AHY, Winefield RD, Sunde M, Mathews JM, Heverkamp RG, Templeton MD, et al. Structural basis for rodlet assembly in fungal hydrophobins. Proc NASUSA 2006;103:3621–6. Stringer MA, Timberlake WE. DewA encodes a fungal hydrophobin component of the Aspergillus spore wall. Mol Microbiol 1995;16:33–44. Stringer MA, Dean RA, Sewall TC, Timberlake WE. Rodletless, a new Aspergillus developmental mutant induced by direct gene inactivation. Gene Dev 1991;5: 1161–71. Hakanpää J, Szilvay GR, Kaljunen H, Maksimainen M, Linder M, Rouvinen J. Two crystal structures of Trichoderma reesei hydrophobin HFBI—the structure of a protein amphiphile with and without detergent interaction. Protein Sci 2009;15:2129–40. Kwan AH, Macindoe I, Vukašin PV, Morris VK, Kass I, Gupte R, et al. The Cys3–Cys4 loop of the hydrophobin EAS is not required for rodlet formation and surface activity. J Mol Biol 2008;382:708–20. Macindoe I, Kwan AH, Ren Q, Morris VK, Yang W, Mackay JP, et al. Self-assembly of functional amphipathic amyloid monolayers by the fungal hydrophobin EAS. Proc NAS 2012;109:E804–11. Haas Jimoh Akanbi M, Post E, Meter-Arkema A, Rink R, Robillard GT, Wang X, et al. Use of hydrophobins in formulation of water insoluble drugs for oral administration. Colloids Surf B 2010;75:526–31. Beija M, Salvayre R, Lauth-de Viguerie N, Marty JD. Colloidal systems for drug delivery: from design to therapy. Trends Biotechnol 2012;30:485–96. Poptsova MS, Gorgarten JP. Using comparative genome analysis to identify problems in annotated genomes. Microbiol 2010;156:1909–17. Bork P, Bairoch A. Go hunting in sequence databases but watch out for the traps. Trends Genet 1996;12:425. Devos D, Valencia A. Intrinsic errors in genome annotation. Trends Genet 2001;17: 429–31. Beauvais A, Schmidt C, Guadagnini S, Roux P, Perret E, Henry C. An extracellular matrix glues together the aerial-grown hyphae of Aspergillus fumigatus. Cell Microbiol 2007;9:1588–600. Altschul SF. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res 1997;25:3389–402. Morris VK, Kwan AH, Sunde M. Analysis of the structure and conformational states of DewA gives insight into the assembly of the fungal hydrophobins. J Mol Biol 2013;425:244–56. Notredame C, Higgins DG, Heringa J. T-coffee: a novel method for fast and accurate multiple sequence alignment. J Mol Biol 2000;302:205–18.