This review is meant to describe synthetic strategies, from both a chemical and technological point of view, of Molecular Imprinted Polymers (MIPs), with particular emphasis on the chance of obtaining them under the shape of nanoparticles, citing and discussing examples and approaches taken from the recent literature. Must say that, nowadays, molecular recognition processes are playing a growing role in various scientific fields, such as diagnostics, chemical catalysis and separation systems, in addiction to biosensors and drug delivery. In order to pursue this aim, natural molecular recognition systems, like enzymes, antibodies and receptors, are available (Yan, 2002). These last are undoubtedly highly specific and selective for several kinds of chemical and biological moieties but, unfortunately, they also present some disadvantages. Problems connected with the usage of these molecules are their low stability and poor performance in organic solvents, and also at low and high pH values and at high temperature. All these factors cause them, in turn, to have a very short shelf-life, together with high costs. Moreover, not all the target analytes can be recognized by enzymes or receptors, because these last often simply do not exist in nature or are difficult to clone. Finally, these kinds of biomolecules encounter some problems with immobilization on suitable supports for their usage in the assays (Piletsky and Turner, 2006). Fortunately, other recognition systems can be used rather than biomolecules and, among these, molecular imprinted polymers have shown very promising. The increasing amount of papers published per annum in this area over the last ten years, which is almost tripled from 2000 to 2009, seems to reflect the potentiality and the broad interest regarding these materials (Figure 1) (Whitcombe, 2010).
Figure 1. Number of published papers in the area of molecular imprinted polymers for the decade 2000-2009 (Whitcombe, 2010).
In a recent review, molecular imprinting has been defined as "the construction of ligand selective recognition sites in synthetic polymers where a template (atom, ion, molecule, complex or a molecular, ionic or macromolecular assembly, including micro-organisms) is employed in order to facilitate recognition site formation during the covalent assembly of the bulk phase by a polymerization or polycondensation process, with subsequent removal of some or all of the template being necessary for recognition to occur in the spaces vacated by the templating species" (Alexander et al., 2006: 107). It is schematically represented in Figure 2.
Figure 2. Schematic representation of the molecular imprinting process: the reversible interactions between the template and the polymerizable functional monomer may involve one or more of the following mechanisms: [(A) reversible covalent bond(s), (B) covalently attached polymerizable binding groups that are activated for non-covalent interaction by template cleavage, (C) electrostatic interactions, (D) hydrophobic or van der Waals interactions; each one of these is formed with complementary functional groups or structural elements of the template, (a-d) respectively]. A subsequent polymerization in the presence of crosslinker, a cross-linking reaction or other process, results in the formation of an insoluble matrix in which the template sites reside. Template is then removed from the polymer through disruption of polymer-template interactions, and subsequent extraction of the template from the matrix. The template, or his analogues, may then be selectively rebound by the polymer in the sites vacated by template, or 'imprints' (Adapted from Alexander et al., 2006).
As the figure shows, the synthesis of molecular imprinted polymers involves one or more types of monomers, which present suitable functional groups to interact with the template, either covalently or, rather that, exploiting non-covalent interactions. The reaction mixture includes also a cross-linker and a porogenic solvent and, after mixing, it is cured to give a porous material, which contains sites that are complementary to the template molecules, both for shape and functional groups. After this step, the template is usually removed from the imprinted polymer by washing with solvent or through a combination of chemical treatments and washing. At this point, the imprinted sites should be available for rebinding with the template or its structural analogues (Mayes and Whitcombe, 2005).
Differently from biomolecules, MIPs, on the contrary, are stable at low and high pH values, pressure and temperature (<180°C). Moreover, they are less expensive and easier to obtain, and they can be exploited in organic mediums. Finally, they can be synthesized for functionally recognizing of several kind of substances, like ions (Esen et al., 2009), nucleic acids (Ogiso et al., 2007), proteins (Zeng et al., 2010), drugs (Cirillo et al., 2009) and even yeast cells and erythrocytes (Jenik et al., 2009). However, like biomolecules, they are also burdened with some limitations. For example, they generally show a poor performance in aqueous medium, and also lack of a standard common procedure for their preparation (Piletsky and Turner, 2006).
MOLECULAR IMPRINTING APPROACHES
In terms of imprinting strategies, mainly three kinds of approaches can be distinguished. They are all classified according to the nature of the bonds established between the template and the functional monomers, and in particular they are indicated as covalent, semi-covalent and non-covalent approaches.
COVALENT IMPRINTING
This first imprinting approach has been pioneered by Wulff and colleagues (1982), and it involves the covalent modification of the template, which is reversibly chemically bonded to the functional monomers. This modification is then followed by the polymerization step, during which the template is still covalently connected to the monomer, now polymerized. Finally, the template is cleaved through a mild chemical reaction (e.g. hydrolysis or reduction), leaving behind the binding cavities. Theoretically, this approach shows very important advantages, given the fact that, thanks to the strong interaction between the template and the functional monomer, it should lead to homogeneous binding sites (Umpleby II et al., 2000). However, both the removal of the chemically bonded template and also its rebinding are not simple processes, because they involve the disruption and the re-formation of covalent interactions, which in turn make these processes really slow. Moreover, the prior derivatization of the template could not be easy, also depending on its chemical nature and functional groups (Ye and Mosbach, 2002).
SEMI-COVALENT IMPRINTING
A little variation of the above-mentioned method consists in carrying out just the imprinting step using the polymerization of the template bonded to the functional monomer, usually through an ester bond, while the rebinding process is only due to non-covalent interactions. The template is usually removed by hydrolysis, allowing the following rebinding step to happen thanks to the establishment of mainly hydrogen and electrostatic bonds. Unfortunately, also this process is not so advantageous as it appears, because the template hydrolysis could not be so easy due to the steric hindrance, and the same steric aspects could also interfere with the non-covalent interactions in the rebinding step (Mayes and Whitcombe, 2005; Alexander et al., 2006). In order to overcome these problems, sometimes the cleavage of the template has been carried out through reduction of the covalent bond using LiAlH4, and then obtaining, for example, an alcoholic group instead of a carboxylic one (Ikegami et al., 2004). In a totally different and clever way, Whitcombe and co-workers (1995) connected the functional monomer to the template using a linker group, which has been "sacrificed" on the template removal step. That is why this approach was given the name of the "sacrificial spacer approach". In this first example, more in detail, the carbonyl group of a carbonate ester was used as spacer group in the imprinting of cholesterol. A 4-vinylphenyl carbonate ester was then exploited as a template covalently bound to the monomer, template which in turn can be easily cleaved by hydrolysis, releasing CO2. After this removal, recognition sites are obtained, sites which bear a phenolic group that can establish hydrogen bonds with the template. The role of this spacer is then double, because it has both the functions of connecting the template and the monomer each other, but also, properly working as a spacer, it has to avoid the steric hindrance which could take place during the rebinding step.
NON-COVALENT IMPRINTING
The third and last mentioned approach is the so-called non-covalent approach, which has been pioneered by Mosbach and co-workers (Arshady and Mosbach, 1981). It exploits several kinds of non-covalent bonds between the template and the monomers, such as hydrogen, electrostatic and also hydrophobic interactions like Van der Waals forces. Since all these types of interactions are not really strong, a way to obtain more stable template-functional monomer complexes is to use an excess of monomers but unfortunately this choice is far from being free from drawbacks, since it often leads to a wide distribution of heterogeneous binding sites (Ye and Mosbach, 2002). Nevertheless, thanks to its simplicity, this method is definitely the most widely exploited to prepare MIPs, also because several functional monomers are commercially available, while many other tailor-made ones have also been reported (Alexander et al., 2006).
DIFFERENT SHAPES FOR DIFFERENT APPLICATIONS
MIPs are usually prepared as monoliths, exploiting 'bulk' polymerization processes of vinyl monomers. Then, since in this way a unique block is obtained, it has to be grinded and adequately sieved before being used for the various applications. Even if this procedure is simple and convenient, it presents some limitations. First of all, the milling process causes a loss of a significant amount of material, but the main drawback probably is that, after the sieving step, irregularly shaped particles are obtained, which often do not consent to properly use them for the various applications (Alexander et al., 2006). For example, applications like chromatography or SPE require micron size beads in order to pursue an efficient packing into columns or cartridges, while, for the coating of sensor devices, MIPs under the format of thin films fit best. Then, in order to optimise both the performance and the imprinting procedure of MIPs, it is important to develop a synthetic method which allows obtaining them in a predefined structural format, functional to exalt those properties of the MIP material which are more useful for the considered application. Several methods have already been developed, in order to obtain MIPs under these various shapes, like films and membranes, micro- and nanoparticles. Nevertheless, it is not easy to fit the synthetic conditions with the operational parameters required to obtain an adequate imprinting (Pérez-Moral and Mayes, 2006).
MIP FILMS AND MEMBRANES
Two of the most important formats which has been investigated to obtain MIPs are films and membranes. In fact, in the last years, several approaches have been tried in order to obtain MIPs under these shapes. These last are undoubtedly quite advantageous for those applications which require a more or less thin imprinted layer, like in the case of coating sensors with films or for separation purposes using MIP membranes. Some of these synthetic approaches are polymerization in moulds or on layers, electropolymerization, and grafting from or to a support.
The first method has been used by Sergeyeva and co-workers (2010), who synthesized molecularly imprinted polymer membranes able to mimic the catalytic properties of the natural enzyme tyrosinase, in order to fabricate biosensors for phenols detection. The membranes have been obtained through a thermo-initiated radical polymerization which was lead for 12h between two glass slides, supposed to act as a mould. Wu and his group (2009) fabricated an optical sensor for formaldehyde based on a MIP film, which has been created by dropping the polymerization mixture on a mirror surface and then covering it by a 12mm diameter cover slip. The polymerization was then lead using UV light radiation in a nitrogen-purged glass bottle for 48h. In this way they obtained a MIP film about 1mm thick. The main drawbacks of this method, however, lay on the difficulty in managing both the film thickness and its porosity, which sometimes are very difficult to control.
Another method which is useful for preparing MIP films and membranes is electropolymerization. Aghaei and colleagues (2010) fabricated a cholesterol biosensor based on capacitive detection, using the electropolymerization of 2-mercaptobenzimidazole (2-MBI) on a gold electrode, in the presence of cholesterol as template. The electropolymerization process has been lead using cyclic voltammetry. This polymerization method is suitable for obtaining very thin films, which are better for improving the sensitivity of the sensor, but since in capacitive measurements the film has to be electrically insulating, in order to avoid undesirable redox reactions, it is a common procedure to use alkanethiols for filling the defects of the membrane after electropolymerization, thus increasing the insulating properties. This aim has been achieved, in this work, by immersing the electrode into a 100mM n-dodecanethiol ethanol solution for 12h. Another very good example of MIP films obtained by electropolymerization has been provided by Choong et al. (2009), who lead the electropolymerization of a caffeine-imprinted polypyrrole thin film on an array composed by vertically aligned carbon nanotubes (CNTs), which acted as a high surface 3D scaffold for the MIP deposition. Probably the greatest advantage in using this well organised, three-dimensional support is that the thickness of the MIP film which coats each nanotube can be properly regulated to fit target molecules of various sizes.
Since it is not easy to generate high affinity binding sites while controlling, at the same time, the porosity and other features of the polymers, a recent trend is to divide the imprinting process from the obtainment of a MIP characterized by a precise morphology (Rückert et al., 2002). This last aim can be achieved by grafting molecular imprinted polymers "to" or "from" a support, depending on where the grafting reaction starts from, if from the polymer chains (grafting to) or from the support (grafting from) (Minko, 2008). An example of this last approach has been given by Wu et al. (2006), who coated an Au electrode on a quartz crystal microbalance (QCM) with a polymer imprinted for bilirubin. The surface of the electrode had been previously treated with allyl mercaptan, in order to obtain the MIP film through a photo-graft surface polymerization technique, started by UV light irradiation and benzophenone as photo-initiator. However, since in this way it is not easy to avoid gelation and polymerisation processes to take place in solution, new kinds of initiators are starting to be used now, in which one of the radicals derived from the decomposition step is not capable of initiating the polymerisation, while it can terminate the growing polymers in solution (Pérez-Moral and Mayes, 2006). An example of this approach has been recently provided by Lakshmi and co-workers (2009), who prepared an electrochemical sensor for catechols. First they created a poly(aniline) layer on the golden electrode surface through electropolymerization of N-phenylethylene diamine methacrylamide (NPEDMA). This monomer bears orthogonal polymerizable functional groups, an aniline and a methacrylamide function, both independently polymerizable. After this step, the methacrilamide groups on the electropolymerized layer have been activated through an iniferter, N,N′-diethyldithiocarbamic acid benzyl ester, and, after that, the MIP components, chosen to obtain a mimic of the enzyme tyrosinase, have been added (Figure 3).
Figure 3. Scheme of the preparation of the catalytic MIP-hybrid electrodes imprinted with catechol (Adapted from Lakshmi et al., 2009)
MIP BEADS
Microparticles.
Nanoparticles. Importance and possible uses, examples and approaches, with advantages and disadvantages: suspension polymerization; two-stage swelling polymerization; precipitation polymerization; core-shell particles; emulsion polymerization; miniemulsion polymerization. Possibility of dealing with living radical polymerizations approaches.
Conclusions.
REFERENCES (Up to now):
