Molecularly imprinted polymers (MIPs) is a new family of porous materials capable of biomimetic molecular recognition. These polymers are prepared by mixing monomers and the second component, called template. After polymerization, where monomers link with each other, the template is removed, creating cavities and pores in the polymeric structure. Since the structure of the polymer forms around the template molecules (imprinting), these cavities and pores are complementary in shape to the template molecules. This molecular recognition functionality of MIPs can be exploited in a number of applications from sensing and separations to drug delivery.
The key challenge in the current MIP technologies is rational design and selection of the components of these materials, given a large number of possible polymeric precursors. Molecular simulations can substantially streamline and optimize this process, thus avoiding costly experimental effort. However, an important requirement for the molecular simulations to become a major tool in MIP studies is a model capable of molecular recognition.
The main objective of this project was to develop a realistic molecular model of MIP formation and function. Using this model we aimed at explaining how molecular recognition is induced in MIPs and how it depends on the chemical properties of the functional monomers and other components of the system.
The insights generated from this model would be invaluable in developing rational MIP design strategies.
Particular emphasis of the project was made on the templates relevant to the pharma industry and for drug delivery applications, where MIPs are expected to make the most profound impact.
The efficacy of drug therapies can be increased immensely by precise control of the location and rate of the drug release. One way to control drug delivery is to store the active agent in a porous material, or a matrix, which would slowly release the medicine through its pores. This can be useful in sustaining the desired level of a medicine in the body over an extended period of time for a range of conditions such as diabetes and some forms of cancer. This approach can be even further improved if there was a way to prepare porous materials which would selectively and strongly bind the desired drug molecules, making the release time even longer.
An exciting recent idea is that we can use new types of polymers for this purpose. These polymers are prepared by mixing monomers and the second component, called template. After polymerization, where monomers link with each other, the template is removed, creating cavities and channels in the polymeric structure. Since the structure of the polymer forms around the template molecules (imprinting), it is possible that many of the cavities formed are of the shape that is complementary to template molecules, like lock and key, or hand and glove. Therefore, one can hypothesize that this material should recognize and bind template molecules. In fact, this ideology is borrowed from biological systems, where the geometrical (and interaction) match between two molecular objects is called molecular recognition and plays a vital part in many processes, including enzymes functions and genetic information replication! A number of world renowned groups (Peppas and Langer in the US, Piletsky and Turner in the UK, Mosbach in Sweden) have been developing these new polymers with desired functions.
Although simple in principle, this concept is difficult to implement for controlled drug delivery. The final structure should combine selectivity (so it binds only desired molecules) and at the same time be accessible, that is the drug molecules should be able to go in and out of the structure. This is a difficult compromise to achieve and the final result depends on many experimental variables, such as components structure and concentrations, temperature and so on. Moreover, it is not even clear what this compromise should be for controlled drug delivery applications.
This problem is tedious to investigate in experiments, considering a large number of possible factors. A more efficient approach is to construct a simplified model that imitates a real system and use a computer to calculate its properties. This is called computer simulation approach. Sometimes, the behaviour of the model can be reduced to several simple (or not so simple!) mathematical formulas, which we would generally call a theory.
The purpose of this project was to construct models of possible materials for controlled drug delivery and to investigate their properties as a function of processing conditions using theory and computer simulations. This will help us understand the mechanism of molecular recognition and what features a material should have in order to exhibit this property. We will also show how this property depends on various factors involved in material synthesis and how it is interconnected with other important characteristics of the material such as accessibility and selectivity. Eventually, we aim at designing a faster, cheaper and more efficient route to synthesis of new materials for controlled drug delivery, hoping it will help to battle a range of illnesses, from arthritis to diabetes and cancer.
In this project we investigated the main factors that influence molecular recognition functionality of MIPs, using both theoretical methods and simplified models. Specifically, we demonstrated in a series of publications that molecular recognition in porous structures can not be explained solely based on geometric shape complementary arguments (with a specific exception of enantiomeric molecules) and the complete description of the phenomena must also involve hydrogen bonds and other polar contributions. We proposed the first simplified model of molecular recognition in imprinted porous structures and showed that the performance of MIPs is indeed a fine tuned function of various processing conditions such as relative concentrations of polymerization components. We believe this model can be used to further elucidate factors affecting molecular recognition functionality of MIPs and to propose new synthetic protocols.
A detailed atomistic model has been also developed to provide more quantitative predictions of the adsorption and binding in MIPs. This is the first model based on the protocol that realistically mimics all stages of MIP formation and function. From this perspective, Sarkisov's group holds a unique expertise in the field. Accurate atomistic force fields have been adopted and carefully validated in this study. We then explored adsorption and molecular recognition of volatile organic compounds, or VOCs, such as benzene, toluene and pyrazine, in MIPs. Small drug molecules are indeed comparable in size and structure to VOCs, and therefore the proposed model, if further extended to binding in solution, can be used to design and optimize MIPs for adsorption and release of small drug molecules.
The project resulted in 5 peer reviewed journal publications, 2 conference proceedings. It has been also presented at numerous international conferences including the most recent MIP2012 meeting in Paris.