Sample Preparation with Molecularly Imprinted Polymers (MIPs)

The first report of using molecularly imprinted polymers (MIPs) as a selective sorbent in sample preparation was in 1994 for the solid-phase extraction (SPE) cleanup of urine to analyze pentamidine. Now, 30 years later, the applications for MIPs are widespread. The molecular recognition abilities of these stable polymers offer high selectivity for a sample preparation method. MIP synthesis is inexpensive, easy, and a clear alternative to the use of natural receptors. Sample preparation techniques using MIPs are broad. Beyond SPE, MIPs are also used in microextraction, matrix solid-phase dispersion, coated fibers, monoliths, and stir bar sorptive extraction, to name a few. This article will describe the current highlights of using MIPs in sample preparation.

W

hat is a molecularly imprinted polymer (MIP)? Simply, it is a tailor-made polymer which recognizes an analyte of interest, following the same principles of the lock and key enzyme mechanism. Many recent advances have brought MIP technology to a wider audience, specifically for sample preparation, and applications now abound. If the inception of MIPs is taken as Sellegrin’s landmark paper in Analytical Chemistry on the analysis of pentamidine (1), this is a 30-year-old technology, and even with the breadth of publications on preparation and applications, it has been slow to develop on a commercial scale. Likely, this is because producing a tailor-made polymer is a tedious and time-consuming process. Has the technology reached the precipice of commercial viability? MIPs’ time may finally be here.

How MIPs Are Made

The most common process to make a MIP through polymerization is called the noncovalent process. Procedurally, the template molecule (analyte of interest) and functional monomers are put in a solution (the solvent is called the porogen, so named because it infiltrates the pores) and complexed via noncovalent bonds. A cross linker and an initiator are then added to initiate polymerization of the monomers around the template molecule. The whole system is washed extensively to remove the template molecule, and cavities or binding sites are then created. The binding sites are similar in size, shape, and position of functional moieties to the target molecule. The necessity of a large number of organic solvents under acidic or basic conditions is a limitation to the noncovalent MIP polymerization process. Relatively speaking, the experimental approach is simple; monomers of many types are readily available for a variety of desired template analytes. The template-monomer interactions are equilibrium-driven in noncovalent processes, and typically, a monomer content is added in excess; this can create non-selective bonding sites in the MIPs (2).

Two other processes are used to make MIPs: covalent and semi-covalent. The covalent approach, which can be limited, was introduced by Wulff and Sarchan (3) in the early 1970s, but not used for sample preparation methods until much later. It differs from the noncovalent approach in that the bonds between the template molecule and the monomers are reversible covalent bonds in the first step of the polymerization process. To remove the template molecule, the covalent bonds that were formed need to be broken, and mild conditions are required. These conditions are often difficult to determine, which is the most prevalent limitation to the covalent approach.

In the semi-covalent approach, before polymerization, the target analyte is covalently bound to the chosen functional monomer, or the imprinting step. The rebinding step relies only on classic noncovalent interactions, such as hydrogen bonding and hydrophobic, as well as ionic interactions.

The polymerization technique most frequently used for all three of these polymerization processes is bulk polymerization; however, if regularly shaped particles are required, suspension or precipitation polymerization processes are better choices (4). Bulk polymerization includes a milling step, which can produce irregularly-shaped particles and destroy binding sites. Suspension and precipitation polymerization, by nature of the processes themselves, generate spherical MIP particles. Spherical particles provide optimal surface contact for the target analytes; irregular surfaces are less ideal.

It is important to note that not all MIPs are made with a single target molecule. There is a double template approach, where two analytes of interest are used and complex sample extraction is made easier. Reports have shown higher adsorption capacity and a reduction in sample preparation steps when double templates are used (5). Dummy molecules can also serve as the targeted molecule. Generally chosen by their resemblance to the specific desired target in terms of shape, size, and functional groups, the advantage of using a dummy is the elimination of high recovery values caused by residual targeted molecules not removed with the wash step. This potential interference is more important when low-level quantitative analysis is the end analysis goal.

From an environmentally conscious perspective, synthesis procedures used to develop MIPs have only recently considered the 12 principles of Green Chemistry, and many improvements can be made. MIPs frequently use chemicals such as acrylic acid, vinyl pyridine, and styrene as the functional monomer, and ethylene glycol dimethacrylate or divinylbenzene as the cross linker. The reduction of organic solvent use and the choice of greener polymerization strategies, as well as greener poragens, are just starting to be mentioned in the literature (4,6). The use of ionic liquids and deep eutectic solvents are on the rise, as is the use of biopolymers and natural monomers.

MIP in SPE

Batch SPE, conventional offline SPE, and online SPE have all used MIP technology (2). Offline SPE with MIPs simply substitutes the classical C-18, C-8, and other conventional phases in the cartridge with the MIP phase. The column is then loaded with the sample after conditioning, washed with the selected solvent, and the target analyte is eluted. The extract is then analyzed chromatographically, as in conventional SPE. Online SPE places a cartridge or column loaded with the prepared MIP in a direct line between the injection port and the chromatographic column when a switching valve is turned. In this way, the sample can be loaded onto the MIP cartridge, and then rinsed with the mobile phase. Depending on the valving configuration, the MIP cartridge can be rinsed with a different eluent to eliminate interferences or a sample matrix before being introduced to the chromatographic column. Batch SPE has been replaced by the offline and inline procedures described above, except for the case of magnetic molecularly imprinted particles (MMIP).

Magnetic Molecularly

Imprinted (MMI) SPE

The main advantage of magnetic molecularly imprinted (MMI) polymers is the enhanced selectivity for a target molecule available through MIP technology coupled with the ability to be quickly isolated from sample matrices with an external magnetic field (6). The procedure to make these MMI materials consists of a sol-gel reaction with different types of magnetic sorbents; Fe₃O₄ is a frequently used magnetic sorbent. Particles, nanotubes, and nanosheets have all been tried and reported as potential magnetic sources. The goal of these systems is to maintain enough magnetic properties that a magnet can be used to extract molecules of interest after the separation occurs on a material of high specific surface area which still contains adequate molecularly imprinted binding sites or cavities for the target analyte.

MIP in Dispersive SPE (dSPE)

Dispersive SPE (dSPE) is conducting by adding the adsorbent particles directly to the sample matrix, allowing an extraction to occur over a selected extraction time, after which the adsorbent is recovered either by centrifugation and filtering, or, if a magnetic core particle was used, magnetic field. A desorption step follows with a suitable solvent. Choice of desorption solvent, extraction time, and stirring rate, all control the extraction efficiency, and therefore must be determined. This is a very common approach used in MIP sample cleanup. For MIPs prepared by bulk polymerization techniques, a specific amount of the MIP is added to the sample matrix, and the process ensues as for standard dSPE. Pichon and associates reported that over 50% of MIP sample extraction was conducted by dispersive SPE. Since single lab-made MIP materials are the most common, dSPE is the easiest approach to use for sample cleanup (6).

MIP Thin Films and Fibers

High-throughput analysis and lower consumption of organic solvents than conventional SPE methods have been obtained using in situ synthesis of MIPs on the surface of microfiltration glass fiber membranes in multi-well filter plates or onto polyethylene frits (4,7). Similar surface imprinting methods have been utilized for MIP-coated fibers and stir bars for SPME and stir bar sorptive extraction (SBSE). Activation by a silylation reaction, followed by immersion in the prepolymer solution, enables the polymerization reaction to take place on the fibers or glass magnets. The coating thickness and porosity of the final imprinted fiber are controlled by the polymerization time and the choice of porogen. There are polymerization reactions that allow for better control of the thickness of the polymer; for example, surface reversible addition–fragmentation chain transfer (RAFT). Additional work on these techniques is needed to advance MIPs on thin films.

MIP Monoliths

Fibers, disks, and columns have been prepared as monolithic imprinted polymers for sample extraction (8). Monoliths can be both organic polymer-based or inorganic-based materials. Both types offer the absence of frits to retain the column material in the column. The monoliths are a single piece of material in the column, fiber, or disk, and show good permeability, low pressure drop, and allow for preparation within the structures. The selection of the porogen and the template need to be chosen carefully during preparation, since both affect the final morphology of the monolith, as well as the extraction efficiency. Additional flexibility, high stability, and enhanced mechanical properties have been seen when hybrid organic-inorganic MIPs were created. MIP monoliths have potential for both proteins and peptides in bioanalysis.

MIP in SPME

As early as 2001, solid-phase microextraction (SPME) tubes have been coated with MIPs (6), but more applications have been seen from 2016 to the present. Typical SPME tubes are coated with polydimethylsiloxane (PDMS), polyacrylate (PA), or divinylbenzene (DVB). Both stainless steel and silica glass have been used to create MIP-based SPME tubes, the capillaries (200 to 300 µm diameter) are coated by immersing them in a larger tube with a diameter of at least 1 mm (6). This larger tube is filled with the polymerization solution and serves as an outer mold to contain the MIP layer. As for all extraction techniques, the measure of effectiveness is based on temperature, extraction time, the sorbent itself, and the analyte. For MIPs that are not stable at elevated temperatures (the temperature of a GC inlet), desorption can take place in a liquid solvent, which is then transferred to a sample vial, and a standard injection is made. For trace level analysis, this is less desirable, as dilution can result using a solvent for the desorption step. Frequently, a non-imprinted polymer (NIP) is used as the control. With an NIP, no target analyte is added in the polymerization process, but the process is identical to the one used when the MIP is created. Precision for fibers produced using this coating polymerization process have been reported at less than 5% for one fiber used repeatedly, and less than 6% for five fibers used for the same target analyte. Reusability of the fibers showed they remained effective up to 100 cycles.

MIP SPME fibers prepared in this way showed higher extraction efficiencies than fibers based on the chemistries of PDMS, PA and DVB, but the extraction phase amount on the MIP SPME phases was higher based on the larger amount of extraction phase present—this trend is, therefore, as should be expected (7).

Commercial Availability

Figure 1 from (6) shows the groups of molecules for which MIPs have been developed during the two-year period from 2018 to 2020; development has touched every area of sample preparation from simple and complex matrices. However, most of the MIPs that have been created were generated for use in individual labs by select researchers. The commercially available MIPs, although growing, remain more limited. Supelco, with their SupelMIP SPE cartridges, currently offer the following target analytes: clenbuterol, beta agonists, beta blockers, beta receptors (which are a combination of beta agonists and beta blockers), and tobacco metabolines, such as 4-(methynitrosamino)-1-(3-pyridyl)-1-butanol, (MMAL), N-nitrosonornicotine (NNN), 4-(methynitrosamino)-1-(3-pyridyl)-1-butanone (NNK), N ‘-nitrosonanabasine (NAB), and N’-nitrosoanatabine (NAT). They also have MIP SPE cartridges for chloramphenicol, triazine, riboflavin, and amphetamines. Likewise, Affinisep have developed their own separate line, AFFINIMIP SPE, particularly for food and pharma analysis. Thus far, their target molecules include mycotoxins (patulin, zearalenone, ochratoxin A, fumonisins), drugs (amphetamines), and endocrine-disrupting compounds (bisphenol A and estrogens [estradiol]).

Based on all the developed MIPs in the literature, the number available for commercial purchase is relatively small. There is good news, however—both Affinisep and MIP Technologies, a subsidiary of Biotage, will custom make MIP phases for desired target analytes. This service opens the field to those in commercial laboratories that do not have the time or resources required to develop MIP separation phases for SPE.

Summary

Without a doubt, MIP technology offers an exclusive selectivity for analytes of interest^; simply, the MIP is designed specifically for the target analyte. This selectivity comes at the cost of a labor-intensive process to produce the MIP. Extensive research and a wide variety of sample preparation options are available, and current efforts are focused on making the established synthesis and polymerization pathways simpler and more compliant with the principles of Green Chemistry. In addition, commercial offerings for high-value target analytes and MIP synthesis for hire should broaden the viability of MIPs in the routine laboratory. After 30 years, this technology appears to be on the precipice of making a significant contribution to the routine analytical laboratory.

References

Sellergren, B.; Swietlow, A.; Arnebrant, T.; Unger, K. Consecutive Selective Adsorption of Pentamidine and Phosphate Biomolecules on a Self-Assembled Layer: Reversible Formation of a Chemically Selective Coating. Anal. Chem. 1996, 68 (2), 402–407. DOI: 10.1021/ac9508356

Turiel, E.; Martin-Esteban, A. Molecularly Imprinted Polymers for Sample Preparation: A Review. Anal. Chim. Acta 2010, 668, 87–99. DOI: 10.1016/j.aca.2010.04.019

Wullf, G.; Sarhan, A. Uber die Anwendung on enzymanalog gebauten Polymeren zur Racemattrennung. Angew Chemie1972, 84 (8), 364. DOI: 10.1002/ange.19720840838

Diaz-Alvarez, M.; Turiel, E.; Martin-Esteban, A. Recent Advances and Future Trends in Molecularly Imprinted Polymers-Based Sample Preparation J. Sep. Sci. 2023, 46 (12), 2300157. DOI: 10.1002/jssc.202300157

Tang, W.; Li, G.; Row, K. H.; Zhu, T. Preparation of Hybrid Molecularly Imprinted Polymer with Double Templates for Rapid Simultaneous Purification of Theophylline and Chlorogenic Acid in Green Tea. Talanta2016,152, 1–8. DOI: 10.1016/j.talanta.2016.01.046

Pichon, V.; Delaunay, N.; Combes, A. Sample Preparation Using Molecularly Imprinted Polymers. Anal. Chem. 2020, 92, 16–33. DOI: 10.1021/acs.analchem.9b04816

Koster, E. H. M.; Crescenzi, C.; Den Hoedt, W.; Ensing, K.; de Jong, G. J.; Fibers Coated with Molecularly Imprinted Polymers for Solid Phase Microextraction. Anal. Chem. 2001, 73, 3140–3145. DOI: 10.1021/ac001331x

Gama, M. R.; Bottoli, C. B. G. Molecularly Imprinted Polymers for Bioanalytical Sample Preparation. J. Chromatogr. B2017, 1043, 107–121. DOI: 10.1016/j.jchromb.2016.09.045

About the Author

Mary Ellen McNally is an FMC Fellow at the Stine Research Center for FMC Corporation. Dr. McNally was named to the Analytical Scientist Power List as one of the Top 50 most influential women in the analytical sciences, has received the American Microchemical Society Steyermark Award in the field of microanalysis, the Chromatography Forum of Delaware Valley Award for contributions to the field of chromatography, and has been recognized for her contributions to the field of supercritical fluids by the Midwest SFC Discussion and the Tri-State Analytical Supercritical Fluid Discussion Groups.