Interactions between inorganic materials and biomolecules at the molecular level, although complex, are commonplace. Examples include biominerals, which are, in most cases, facilitated by and in contact with biomolecules; implantable biomaterials; and food and drug handling. The effectiveness of these functional materials is dependent on the interfacial properties, i.e. the extent of molecular level ‘association’ with biomolecules. The present article gives information on biomolecule–inorganic material interactions and illustrates our current understanding using selected examples. The examples include (i) mechanism of biointegration: the role of surface chemistry and protein adsorption, (ii) towards improved aluminium-containing materials, and (iii) understanding the bioinorganic interface: experiment and modelling. A wide range of experimental techniques (microscopic, spectroscopic, particle sizing, thermal methods and solution methods) are used by the research group to study interactions between (bio)molecules and molecular and colloidal species that are coupled with computational simulation studies to gain as much information as possible on the molecular-scale interactions. Our goal is to uncover the mechanisms underpinning any interactions and to identify ‘rules’ or ‘guiding principles’ that could be used to explain and hence predict behaviour for a wide range of (bio)molecule–mineral systems.
- inorganic material
Interactions between inorganic materials and biomolecules at the molecular level, although complex, are common occurrences. Examples include biominerals, implanted biomaterials and the processing and passage of food and drugs through an organism. The effectiveness of these functional materials is to a large extent dependent on the interfacial properties, i.e. the extent of molecular level ‘association’ of biomolecules with the mineral phase. One goal of our research is to uncover the mechanisms underpinning any such interactions and to identify ‘rules’ or ‘guiding principles’ that could explain and, furthermore, predict structure and properties for a wide range of (bio)molecule–mineral systems. An understanding of these interactions would be highly fruitful not only to understand biological mineralization processes, but also to design novel materials and processing technologies for applications in fields as diverse as biological imaging and biosensors, implant integration, food and drug processing and delivery, and electronic materials [1–9].
Lessons from biomineralization
Biominerals are widely formed by bacteria, single-celled protists, plants, invertebrates and vertebrates including humankind. They consist of an inorganic phase (or phases) (usually simple salts or oxides) and a range of biomolecules that are often proteins, but may be carbohydrates, lipids or low-molecular-mass (<1 kDa) molecules such as polyamines [10–12]. Biominerals are always formed from an aqueous environment that is globally undersaturated with respect to the required mineral or mineral phase and yet is rich in ions and biomolecules. Reaction conditions are such that protons or hydroxyl ions are not readily available (i.e. circumneutral pH where H+ or OH− concentrations are of the order of 10−6–10−8 M) and temperatures for reaction are rather low (4–70°C). The biological processes that control and regulate biomineralization are able to select for particular building blocks/entities from a plethora of ions and biomolecules, and also ensure that the ‘correct’ inorganic and biochemical species combine into a composite material(s) in a spatially and temporally defined fashion. The narrow range of reaction conditions available to biological organisms does not hinder the formation of complex structures or the development of physical and mechanical properties that have evolved to provide support to a largely water-based structure (e.g. bones in humans), withstand stress (e.g. some silicified structures in plants), act as armour (e.g. mollusc shells) or act as ion supplies (e.g. ferritin for iron and calcium and phosphate for bones), etc. The composites, with their wide-ranging structures and properties, serve as exquisite guides to the synthesis possibilities for the controlled fabrication of materials having a wide range of mechanical and other properties.
The formation of a biomineral requires the transport and concentration of the desired element(s), nucleation, growth and aggregation/assembly of the mineral building blocks, together with possible phase transformations. Finally, growth inhibition is necessary to limit the dimensions and tune the properties of the resulting materials. Many minerals are formed within membrane-bounded compartments that limit the volume of the chemical/biochemical environment, and the biomolecules present within this space may act as a scaffold (usually an insoluble array of biomolecules) on which the inorganic mineral may form, and they can be involved in nucleation, and can control and inhibit growth. In addition, they may also serve to protect the mineral from dissolution during the life of the organism. It should be noted that the activity (and perhaps structure) of the biomolecules involved in these distinct phases of biomineral formation and stabilization may be modified by the volume restrictions imposed on them, leading to perhaps far from conventionally anticipated behaviour.
Probing the interface
One approach to understanding the role(s) played by biomineral associated molecules in structure regulation is to extract, isolate and characterize the biomolecules and then assess their activity using model in vitro assays. Information so obtained is then extrapolated back to the living organism, generally with very little ‘actual’ biological evidence for the validity of the claims made. Using this approach, only a very few biomolecules have been fully characterized and even less information is available concerning the complex matrices in which they are found [10,13–19]. Similarly, their activity when part of a composite structure is also not understood [10,13–19]. In addition, the approach is limited to a few materials only and excludes many commercially relevant materials such as CdS, oxides of titanium, tin, etc., and metals. An alternative molecular biology approach to understanding this phenomenon is to identify key molecules involved in mineralization, to make point mutations of residues thought to be important and to investigate the consequences of these changes in vivo [20,21].
Although it is very difficult to probe the interface between the largely inorganic materials that form and the biomolecular matrices/molecules involved in structure regulation, information obtained by experimenters is being used by scientists implementing computational techniques to probe the interface between the organic and inorganic components [22–27]. As biomineralization takes place in the presence of water, it is crucially important that the chemistry (including the effects of solution composition) and physics (the effect of biomolecules and inorganic surfaces on the detailed structure of water and similarly how the structure of water affects interfacial processes) of the aqueous interface are considered as fully as possible in order to generate models that are realistic. Factors such as charge (including distribution of charge within and on water molecules and the biomolecules themselves) are an issue, as is a sensible description of changes in charge during a binding process .
An alternative use for the structural information obtained for mineral-associated biomolecules is to try to understand the effect that chemical groupings, singly or in combination, have on the mineral formation process [28–31]. The aim is, for an individual biomolecule, to identify the possible role(s) of functional groups and their spatial distribution during mineral formation. As an additional benefit, this approach may also identify simple structural motifs important in structure control that may then be used to construct a new generation of materials with improved structure and function. The principles learnt could be applied both to materials that organisms routinely produce and to other materials that are only produced in the laboratory or by geological processes.
A further approach that we are using to increase our understanding of interfacial interactions between inorganic materials and biomolecules has been to make use of a method typically used in the pharmaceutical industry for drug discovery. It consists of combinatorial approaches such as phage and cell display methods to identify biomolecules (peptides) that interact with a wide range of natural and non-natural mineral surfaces and to then study the interaction of these molecules with the minerals themselves [27,32–35]. The combinatorial method involves the selection of peptides capable of interacting with inorganic materials from a random library of sequences [even within over a billion different sequences, the length of the peptide determines whether the whole of the possible peptide reaction space is explored (e.g. heptamers) or not (e.g. dodecamers and higher) (Table 1)], then taking this information and attempting to use it to generate the same or similar materials in the laboratory.
The ‘biopanning’ approach has generally been used to isolate biological ‘catalysts’ and ‘templates’ and then to use the peptides for the ‘bottom-up’ microfabrication of materials. However, the protocol used to synthesize the target material for biopanning is thought to play some role in determining which peptide compositions selectively bind to the particular surfaces expressed. For example, the peptide sequences identified to strongly bind silica differ significantly in their sequences and chemical properties depending on whether the target material was amorphous silica (prepared using a solution synthesis)  or quartz (crystalline form of silica) [37–39] (Table 2). The range of surfaces that have been selected for by this approach is large and includes metals, oxides, salts, polymers and carbonaceous materials used in many key applications. We believe that the combinatorial approach can be used to obtain useful information concerning interactions between minerals and biomolecules that would be of use to both experimental and computational scientists to increase our understanding of these complex processes, to produce predictive rules governing such interactions and to generate new materials including ordered assemblies of materials. Concerning the interactions, questions that need answering include the significance of (i) the atoms/ions present on or within a particular material surface and/or their disposition with respect to one another in the bulk and on the surface of the material, (ii) the fundamental chemistry of the peptides (chemical functionality and spatial disposition), and (iii) the impact of either or both of the above on characteristics of the media such as water structure and the disposition of ions. Answers to these questions would assist in using such biomolecules and their derivatives to direct materials synthesis, including phase choice, provide morphology control, stabilize and/or dissolve particles and could also increase our understanding of aspects of mineral production during biomineralization.
In our research, we have extensive experience working with biominerals including silica [40–44] and the sulfates of strontium [45–48] and barium [46,49]. Some of our most recent work includes extensive studies of silica formation in the presence of small and large (bio)molecules aimed at understanding how such molecules interact with silica species (are the observed processes driven by ionic, hydrophobic and/or hydrogen-bond interactions?). In the case of silica, this understanding can be used to customize important properties of the final product, including the size, porosity and surface chemistry. We are exploring interactions between natural proteins and nanometre-sized particles, and between model peptides and particles. We now give short summaries of these two seemingly neighbouring areas to present the distinctive elements that they introduce into our research strategy.
Mechanism of biointegration: the role of surface chemistry and protein adsorption
We are interested in the mechanisms by which an implanted material can be integrated or rejected by the human body. In this respect, the physicochemical interactions between living tissue and a biomaterial are of utmost importance both during and after implantation. To understand and tune bone-implant interactions, one must consider the overlapping length and timescales at which the underlying processes occur. The interactions involving molecules including water, ions and small proteins occur on the shorter nanometre scale (we will restrict this to 100 nm) and occur within timescales that range from nanoseconds to hours. The larger scale (1–100 μm) interactions involving cells tend to occur from several hours after implantation onwards. Our experiments performed using functionalized flat surfaces and spheres from 15 to 165 nm in diameter, together with those of other researchers, have shown that proteins adsorb in differing quantities, densities, conformation and orientations depending on the chemical and physical characteristics of the surface [6,8]. For the model proteins investigated, BSA and Fg (fibrinogen), surfaces that are hydrophobic in nature caused the proteins to adhere more strongly and deform the most. The two model proteins responded differently to surfaces with differing curvature with BSA deforming most on the flatter surfaces (largest particles), and Fg deforming most on the most curved surfaces (smallest particles). Given that protein adsorption and replacement on surfaces is a prerequisite for an implanted material to be recognized by the body, an understanding of protein–surface interactions and control over the conformation and orientation of immobilized species may ultimately allow the fabrication of materials/surfaces with tailored bioactivity.
Towards improved aluminium-containing materials
In another area of research, we are investigating the interactions of aluminium hydrolytic species from 1 nm upwards with biomolecules including proteins. Our research aims to understand how charge, morphology and the structure of the aluminium species influence how they interact with proteins [50,51]. In studies with model globular proteins (BSA, pI ∼4.5, and lysozyme, pI ∼11.4), gels from aluminium hydroxide sols [with both proteins, albeit of vastly varying strength due to the differing abilities of BSA (excellent) and lysozyme (poor) to induce an electrostatically driven coagulation] and particles of distinct sizes (Al13-mer, Al30-mer polycations with both proteins) are formed . As an example, for the lysozyme–polycation complexes measured by light scattering, the sizes obtained suggest the formation of a complex between one protein molecule with either three Al30-mers or six to seven Al13-mers. Evidence for atomistic interactions between the mineral and the biomolecule was obtained from detailed analysis of protein/aluminium species FTIR (Fourier transform infrared) correlation maps. Polycations were demonstrated to bind to specific acidic and hydroxy-containing amino acids and, in some cases, evidence could be obtained that the binding process resulted in further functional groups being deprotonated, thereby making them available for binding to the aluminium-containing surfaces. The spectroscopic study also showed that interactions between the proteins and nanometre-sized aluminium hydroxide particles to form gels arose largely from hydrogen-bonding interactions between hydroxy groups on the particle surfaces and acidic amino acid side chains. An understanding of the variety of interactions between aluminium-containing species and proteins with varying physicochemical properties provides a useful base for the fabrication of novel aluminium-based materials and could minimize the risks of release or improper use of harmful forms of the metal, e.g. in water treatment  and vaccine adjuvants creating aluminium sensitization .
Understanding the bioinorganic interface: experiment and modelling
In our studies of peptide–mineral interactions, we are working towards generating a predictive model that will explain why certain peptides, for example those generated by phage display methods, interact with and, in many cases, promote the formation of specific materials under mild conditions. We are studying both crystalline (ZnO) and amorphous (SiO2) oxide phases and metallic (silver) surfaces. Our approach is to use experimental studies to quantitatively measure the binding patterns and correlate the resulting information with computational studies of peptide behaviour in solution and binding to specific surfaces. The role of different surface arrangements of atoms/ions, water, charge states and concentration effects are being explored in silico and compared with the experimental binding and morphology data obtained. The effects of point mutations within a peptide sequence and putative smaller binding fragments on peptide conformation and energy are also being explored, with the in silico data giving us an informed lead as to the peptides to synthesize and explore in our experimental studies.
Our studies are directed towards understanding the interface between inorganic and biomolecular species and the generation of a predictive model applicable to a wide range of material phases.
We are grateful for support, past and present, from Agriculture and Food Research Council (AFRC), Science and Engineering Research Council (SERC), Biotechnology and Biological Sciences Research Council (BBSRC), the European Union (EU), Air Force Office of Scientific Research (AFOSR), Unilever, Ineos Chemicals and Smith and Nephew, which has enabled us to perform a wide range of studies both in vivo and in vitro.
The Perry group, past and present, are thanked for their many contributions in this area of science.
Bionanotechnology II: from Biomolecular Assembly to Applications: Biochemical Society Focused Meeting held at Robinson College, Cambridge, U.K., 7–9 January 2009. Organized and Edited by Tony Cass (Imperial College London, U.K.) and Dek Woolfson (Bristol, U.K.).
Abbreviations: Fg, fibrinogen
- © The Authors Journal compilation © 2009 Biochemical Society