Electrochemical MIP Sensor for Butyrylcholinesterase

Molecularly imprinted polymers (MIPs) mimic the binding sites of antibodies by substituting the amino acid-scaffold of proteins by synthetic polymers. In this work, the first MIP for the recognition of the diagnostically relevant enzyme butyrylcholinesterase (BuChE) is presented. The MIP was prepared using electropolymerization of the functional monomer o-phenylenediamine and was deposited as a thin film on a glassy carbon electrode by oxidative potentiodynamic polymerization. Rebinding and removal of the template were detected by cyclic voltammetry using ferricyanide as a redox marker. Furthermore, the enzymatic activity of BuChE rebound to the MIP was measured via the anodic oxidation of thiocholine, the reaction product of butyrylthiocholine. The response was linear between 50 pM and 2 nM concentrations of BuChE with a detection limit of 14.7 pM. In addition to the high sensitivity for BuChE, the sensor responded towards pseudo-irreversible inhibitors in the lower mM range.


Introduction
Specific interactions between reaction partners, e.g., the binding of an antigen to the respective antibody in the immune system, the conversion of substrates by enzymes and the sequence-specific binding of nucleic acids are involved in key events of living systems. The specificity of biomacromolecules is routinely exploited in the biochemical analysis including clinical diagnostics, environmental control, and food analysis. However, the advantage of applying biomacromolecules is accompanied by restricted stability and high prices. In order to overcome the same disadvantages and to reduce the reagents cost, the concept of biomimetic recognition elements has been created: Synthetic organic polymers, which possess specific binding cavities and nucleotide-based aptamers, have been developed as binders for target analytes. The concept of fully synthetic so-called molecularly imprinted polymers (MIPs) has been realized by Wulff and Mosbach [1,2]. MIPs mimic antibodies by substituting the amino-acid-scaffold of proteins by a synthetic polymer. Organic monomers are polymerized in the presence of the target molecule, which is removed after the formation of a polymeric matrix. This process results in the formation of cavities complementary to the size, shape, and position of functional groups of the template molecule, respectively parts of the target molecule [3][4][5][6][7][8]. Typically, MIPs are made up of two to six functional monomers, but they have been also successfully synthesized from only one monomer and even without a cross-linker. This is a real technological breakthrough and

Preparation of BuChE Imprinted Electrodes
Glassy carbon disk electrodes (3 mm in diameter; CH Instruments, Austin, TX, USA) were used for the MIP-synthesis and for the voltammetric and amperometric measurements. Prior to electropolymerization GCEs were cleaned with 30% nitric acid for 15 min. Afterward, mechanical cleaning was performed with 1.0, 0.3, and 0.05 µm alumina slurry, respectively, and electrodes were rinsed with ethanol and Millipore water by ultrasonication.
Electropolymerization was performed by scanning the potential of the GCE between 0 and 0.8 V with a scan rate of 50 mV/s in a solution of 100 mM acetate buffer, pH 5.2 containing 5 mM o-PD and 25 µg/mL BuChE. The number of scans was varied between 5, 10, and 20. Template molecules were removed from the MIP by overnight shaking the electrodes in 100 mM NaOH solution with a Polymers 2019, 11, 1970 4 of 11 speed of 300 rpm. Cyt c imprinted electrodes (Cyt c-MIP) and non-imprinted polymer (NIP) modified electrodes were prepared as control electrodes in a similar manner, but in the presence of 25 µg/mL Cyt c and in the absence of BuChE, respectively.

Electrochemical and Scanning Electron Microscope Measurements
Voltammetric and amperometric measurements were performed by using a PalmSens potentiostat (Netherlands). Electropolymerization, cyclic voltammetry (CV), and an amperometric indication of the enzymatically generated thiocholine (TC) were carried out using a one-compartment three-electrode electrochemical cell with a volume of 2 mL. A spiral platinum wire was used as the counter electrode and an Ag/AgCl electrode (3 M KCl) as the reference. All the measurements were carried out at room temperature.
All steps of MIP preparation and rebinding were characterized by the diffusional permeability of the redox marker ferricyanide, using cyclic voltammetry between −0.2 and 0.7 V with a scan rate of 50 mV/s in 100 mM KCl containing 5 mM ferricyanide solution. Furthermore, the enzymatic activity of the MIP-bound BuChE was indicated by amperometric measurements of the product thiocholine (TC) at 0.4 V in a stirred solution of 100 mM phosphate buffer pH 7.4. For the BuChE inhibition studies, 40 mM stock solutions of respective drugs (rivastigmine, galantamine, memantine, in 100 mM phosphate buffer, pH 7.4) were used. BuChE inhibition was followed using a continuous addition of 1 mM drug to 2.5 mM BTC containing electrochemical cells.
Scanning electron microscopic (SEM) measurements were performed by ZEISS EVO 40 (Merlin, Carl Zeiss, Oberkochen, Germany) EV018 vacuum oven was used for drying the electrode.

Electropolymerization and Template Removal
Electrosynthesis of the BuChE-MIP by scanning the potential of the GCE between 0 and 0.8 V in a solution containing o-PD and BuChE brings about the well-known irreversible peak at 0.4 V for o-PD in the first scan. The current decreases gradually in the following scans approaching almost complete suppression, as it is expected for the formation of the non-conducting polymer covered layer [34] ( Figure 2).

Electrochemical and Scanning Electron Microscope Measurements
Voltammetric and amperometric measurements were performed by using a PalmSens potentiostat (Netherlands). Electropolymerization, cyclic voltammetry (CV), and an amperometric indication of the enzymatically generated thiocholine (TC) were carried out using a one-compartment three-electrode electrochemical cell with a volume of 2 mL. A spiral platinum wire was used as the counter electrode and an Ag/AgCl electrode (3 M KCl) as the reference. All the measurements were carried out at room temperature.
All steps of MIP preparation and rebinding were characterized by the diffusional permeability of the redox marker ferricyanide, using cyclic voltammetry between −0.2 and 0.7 V with a scan rate of 50 mV/s in 100 mM KCl containing 5 mM ferricyanide solution. Furthermore, the enzymatic activity of the MIP-bound BuChE was indicated by amperometric measurements of the product thiocholine (TC) at 0.4 V in a stirred solution of 100 mM phosphate buffer pH 7.4. For the BuChE inhibition studies, 40 mM stock solutions of respective drugs (rivastigmine, galantamine, memantine, in 100 mM phosphate buffer, pH 7.4) were used. BuChE inhibition was followed using a continuous addition of 1 mM drug to 2.5 mM BTC containing electrochemical cells.
Scanning electron microscopic (SEM) measurements were performed by ZEISS EVO 40 (Merlin, Carl Zeiss, Oberkochen, Germany) EV018 vacuum oven was used for drying the electrode.

Electropolymerization and Template Removal
Electrosynthesis of the BuChE-MIP by scanning the potential of the GCE between 0 and 0.8 V in a solution containing o-PD and BuChE brings about the well-known irreversible peak at 0.4 V for o-PD in the first scan. The current decreases gradually in the following scans approaching almost complete suppression, as it is expected for the formation of the non-conducting polymer covered layer [34] (Figure 2). Ten cycles of polymer deposition lead to almost complete suppression of the current for the redox marker ferricyanide for both the NIP-and the MIP-covered electrodes indicating the formation  Ten cycles of polymer deposition lead to almost complete suppression of the current for the redox marker ferricyanide for both the NIP-and the MIP-covered electrodes indicating the formation of a dense polymer film ( Figure 3a). Incubation of the MIP-covered electrode in 100 mM NaOH causes a marked increase of the peaks for ferricyanide in the CVs (Figure 3b) which indicates the formation of pores in the polymer film after the removal of the target. It was found out, that the layer was not tight when only five CV scans were applied for the preparation of the MIPs. The large current for the redox marker of both the NIP and the MIP after electropolymerization indicated that the layer was not tight for ferricyanide after five scans, i.e., did not block the permeation of redox marker. On the other hand, after 20 scans dense films were obtained, which hinders effective template removal. Hence, throughout this study, a fixed number of 10 scans were implemented for the MIP preparation.
Polymers 2019, 11, x FOR PEER REVIEW 5 of 12 of a dense polymer film (Figure 3a). Incubation of the MIP-covered electrode in 100 mM NaOH causes a marked increase of the peaks for ferricyanide in the CVs (Figure 3b) which indicates the formation of pores in the polymer film after the removal of the target. It was found out, that the layer was not tight when only five CV scans were applied for the preparation of the MIPs. The large current for the redox marker of both the NIP and the MIP after electropolymerization indicated that the layer was not tight for ferricyanide after five scans, i.e., did not block the permeation of redox marker. On the other hand, after 20 scans dense films were obtained, which hinders effective template removal. Hence, throughout this study, a fixed number of 10 scans were implemented for the MIP preparation. The SEM investigations of the MIP and NIP samples were also performed to reveal differences in the surface topographies. SEM images of the MIP and NIP samples taken prior to and after template removal are shown in Figure 4. The surface topographies of the samples are significantly different from each other: The SEM images of NIP samples (Figures 4a and b) show less pronounced changes in surface appearance than that of the MIP samples (Figure 4c and d). There is no drastic change at the NIP surface after treatment with 100 mM NaOH and the polymerized coating on the GCE preserves its physical appearance. However, as shown in Figures 4b and d, the MIP surface possesses more porosity than the NIP surface. Since the only difference between MIP and NIP is the presence of BuChE template in the MIP film, the change in porosity after NaOH treatment indicates the removal of the BuChE moieties.
The procedures of template removal after MIP-synthesis are compromises between complete removal and intactness of the polymer structure. The treatment of the MIP by incubation in 100 mM NaOH obviously leads to partial removal of the polymer which results in the changes of surface morphology. Only enzymatic digestion of the protein target does not harm the polymer; however, problems arrive from the remaining protein [10]. The SEM investigations of the MIP and NIP samples were also performed to reveal differences in the surface topographies. SEM images of the MIP and NIP samples taken prior to and after template removal are shown in Figure 4. The surface topographies of the samples are significantly different from each other: The SEM images of NIP samples (Figure 4a,b) show less pronounced changes in surface appearance than that of the MIP samples (Figure 4c,d). There is no drastic change at the NIP surface after treatment with 100 mM NaOH and the polymerized coating on the GCE preserves its physical appearance. However, as shown in Figure 4b,d, the MIP surface possesses more porosity than the NIP surface. Since the only difference between MIP and NIP is the presence of BuChE template in the MIP film, the change in porosity after NaOH treatment indicates the removal of the BuChE moieties.
The procedures of template removal after MIP-synthesis are compromises between complete removal and intactness of the polymer structure. The treatment of the MIP by incubation in 100 mM NaOH obviously leads to partial removal of the polymer which results in the changes of surface morphology. Only enzymatic digestion of the protein target does not harm the polymer; however, problems arrive from the remaining protein [10].

Rebinding
Incubation of the bare GCE in 50 pM BuChE had almost no effect on the CVs of the redox marker indicating that the enzyme adsorbed only weakly on the bare electrode surface. On the other hand, incubation of the MIP after template removal in 50 pM BuChE caused a suppression of the ferricyanide signal at 0.7 V of 34% whereas 250 pM BuChE caused a decrease of 63% as compared with the value after template removal (Figure 3c and d).
Furthermore, the enzymatic activity of BuChE after rebinding to the MIP was indicated by measuring the anodic oxidation of thiocholine-the reaction product of the enzymatic conversion of BTC (see Figure S1 in the Supporting Information). The current increase after additions of BTC reflects the activity of the BuChE, bound to the MIP. Each concentration was three times determined and the error bars are presented ( Figure 5).
The extrapolated anodic currents increased almost linearly upon the addition of 2.5 mM BTC in the concentration range between 50 pM and 2 nM BuChE ( Figure 5). The limit of detection was calculated to be 14.7 pM using the relation, LOD = 3.3 s/m. For the limit of quantification, a value of 44.6 pM was obtained using the relation, LOQ = 10 s/m, where s is the standard deviation of the lowest concentration (50 pM) and m is the slope of the calibration curve.

Rebinding
Incubation of the bare GCE in 50 pM BuChE had almost no effect on the CVs of the redox marker indicating that the enzyme adsorbed only weakly on the bare electrode surface. On the other hand, incubation of the MIP after template removal in 50 pM BuChE caused a suppression of the ferricyanide signal at 0.7 V of 34% whereas 250 pM BuChE caused a decrease of 63% as compared with the value after template removal (Figure 3c,d).
Furthermore, the enzymatic activity of BuChE after rebinding to the MIP was indicated by measuring the anodic oxidation of thiocholine-the reaction product of the enzymatic conversion of BTC (see Figure S1 in the Supporting Information). The current increase after additions of BTC reflects the activity of the BuChE, bound to the MIP. Each concentration was three times determined and the error bars are presented ( Figure 5).
The extrapolated anodic currents increased almost linearly upon the addition of 2.5 mM BTC in the concentration range between 50 pM and 2 nM BuChE ( Figure 5). The limit of detection was calculated to be 14.7 pM using the relation, LOD = 3.3 s/m. For the limit of quantification, a value of 44.6 pM was obtained using the relation, LOQ = 10 s/m, where s is the standard deviation of the lowest concentration (50 pM) and m is the slope of the calibration curve.
The current signals for the MIP after rebinding of 250 pM BuChE were almost 20-fold and 50-fold higher than those after electropolymerization and template removal, respectively. On the other hand, for the NIP the anodic current after addition of BTC was almost identical after electropolymerization, incubation in NaOH and incubation in 50 or 250 pM BuChE. It was very small as compared with that after the rebinding of BuChE to the MIP ( Figure 5). The comparison between MIP and NIP clearly indicated the higher affinity of the MIP towards BuChE by specific binding as compared with the nonspecific adsorption to the polymer surface of the NIP. In addition, using Cyt c as a dummy template for MIP synthesis the nonspecific binding to the poly-oPD film was investigated. After incubation of the Cyt c-MIP in a solution containing 0.1 µg/mL of BuChE or the same amount of Cyt c, the injection of BTC generated almost the same current signal for both proteins. This value was only 1-3 percent of the respective value for the BuChE-MIP. This finding indicates the low nonspecific binding of BuChE to the polymer film ( Figure 5). The current signals for the MIP after rebinding of 250 pM BuChE were almost 20-fold and 50fold higher than those after electropolymerization and template removal, respectively. On the other hand, for the NIP the anodic current after addition of BTC was almost identical after electropolymerization, incubation in NaOH and incubation in 50 or 250 pM BuChE. It was very small as compared with that after the rebinding of BuChE to the MIP ( Figure 5). The comparison between MIP and NIP clearly indicated the higher affinity of the MIP towards BuChE by specific binding as compared with the nonspecific adsorption to the polymer surface of the NIP. In addition, using Cyt c as a dummy template for MIP synthesis the nonspecific binding to the poly-oPD film was investigated. After incubation of the Cyt c-MIP in a solution containing 0.1 μg/mL of BuChE or the same amount of Cyt c, the injection of BTC generated almost the same current signal for both proteins. This value was only 1-3 percent of the respective value for the BuChE-MIP. This finding indicates the low nonspecific binding of BuChE to the polymer film ( Figure 5). Furthermore, the cross-reactivity of the BuChE-imprinted film was investigated in competitive binding experiments using bovine serum albumin (BSA, MW: ∼66.5 KDa) as a competitor. The BuChE-concentration was constant at 1 μg/mL whereas BSA was increased from 0 to 0.5 μg/L. The current which reflects the enzymatic activity of the MIP-bound BuChE gradually decreased with increasing concentration of BSA. This behavior indicates the partial displacement of BuChE by BSA. The current signal was reduced by at equimolar concentrations of BuChE (1 μg/L) and BSA (0.15 μg/L) by 45% ( Figure 6) and the decrease was less pronounced at a higher amount of BSA. This crossreactivity is not sufficient for measurements of BuChE in blood since serum albumins have an almost 10,000-fold excess in relation to typically 70 nM of the enzyme BuChE. Furthermore, the cross-reactivity of the BuChE-imprinted film was investigated in competitive binding experiments using bovine serum albumin (BSA, MW:~66.5 KDa) as a competitor. The BuChE-concentration was constant at 1 µg/mL whereas BSA was increased from 0 to 0.5 µg/L. The current which reflects the enzymatic activity of the MIP-bound BuChE gradually decreased with increasing concentration of BSA. This behavior indicates the partial displacement of BuChE by BSA. The current signal was reduced by at equimolar concentrations of BuChE (1 µg/L) and BSA (0.15 µg/L) by 45% ( Figure 6) and the decrease was less pronounced at a higher amount of BSA. This cross-reactivity is not sufficient for measurements of BuChE in blood since serum albumins have an almost 10,000-fold excess in relation to typically 70 nM of the enzyme BuChE.

Inhibition of Butyrylcholinesterase by Anti-Alzheimer Drug Rivastigmine
BuChE can be inhibited by several pharmaceuticals that are used in Alzheimer′s disease treatment. In this study, the effect of rivastigmine was examined to demonstrate the inhibitory effects

Inhibition of Butyrylcholinesterase by Anti-Alzheimer Drug Rivastigmine
BuChE can be inhibited by several pharmaceuticals that are used in Alzheimer s disease treatment. In this study, the effect of rivastigmine was examined to demonstrate the inhibitory effects towards the MIP-bound BuChE. Rivastigmine is considered a pseudo-irreversible cholinesterase inhibitor that forms a carbamoylated complex with the enzymes. After single-dose administration, enzyme inhibition was reported to persist for 10 to 12 h. This longer duration of action is unique among cholinesterase inhibitors. Rivastigmine fits into the enzyme's active site in a similar fashion to acetylcholine and has been reported to inhibit both AChE and BuChE with equal potency [35]. Interaction of 10 and 22.5 mM rivastigmine with the BuChE-MIP decreased immediately the sensor response by 23% and 47.5%, respectively (Figure 7). Comparable results have been obtained for galantamine and memantine (see Figure S2 in the Supporting Information).
BSA concentration (µg/mL) Figure 6. Current decrease for competitive binding of BuChE and BSA to the BuChE-MIP (BuChEconcentration was constant at 1 μg/mL whereas concentration of BSA was increased from 0 to 0.5 μg/L).

Inhibition of Butyrylcholinesterase by Anti-Alzheimer Drug Rivastigmine
BuChE can be inhibited by several pharmaceuticals that are used in Alzheimer′s disease treatment. In this study, the effect of rivastigmine was examined to demonstrate the inhibitory effects towards the MIP-bound BuChE. Rivastigmine is considered a pseudo-irreversible cholinesterase inhibitor that forms a carbamoylated complex with the enzymes. After single-dose administration, enzyme inhibition was reported to persist for 10 to 12 h. This longer duration of action is unique among cholinesterase inhibitors. Rivastigmine fits into the enzyme's active site in a similar fashion to acetylcholine and has been reported to inhibit both AChE and BuChE with equal potency [35]. Interaction of 10 and 22.5 mM rivastigmine with the BuChE-MIP decreased immediately the sensor response by 23% and 47.5%, respectively (Figure 7). Comparable results have been obtained for galantamine and memantine (see Figure S2 in the Supporting Information).  These results open the route to a reusable sensor for inhibitors by template removal after inhibition followed by reloading of the enzyme.
Among the enzyme-MIPs, the evaluation of diffusional permeability of a redox-active low-molecular species by CV has been frequently used [40][41][42]. As demonstrated in this work (Figure 3), it is a simple and highly sensitive approach for the characterization of each step of MIP synthesis. However, rebinding causes only small decreases in the large reference value after template removal. In addition, different composition, pH and ionic strength of the solutions for rebinding and the measurement of the redox marker can cause changes of the polymer layer and does not allow to discriminate for unspecific binding of other proteins.
In this paper, we applied the measurement of the enzymatic activity of the biocatalyst to characterize the performance of MIP-sensors. Electrochemical detection of an electroactive product allows the direct quantification of rebinding at the sensor surface. In addition to high sensitivity by the local substrate formation, the assay combines synergistically recognition by the MIP and substrate specificity of the target enzyme.

Conclusions
The BuChE-MIP sensor presented in this paper is based on electro-enzymatic readout and the straight-forward electrosynthesis without additional nanoparticle-textured surfaces or graphene. It realizes measurements of BuChE in the lower pM range without signal amplification thus reaching the sensitivity of immunoassays (see Table S2 in the Supporting Information). However, the cross-reactivity of the MIP-sensor towards abundant constituents in blood will influence the measurement of BuChE in blood because serum albumins have an almost 10,000-fold excess in relation to typically 70 nM of the enzyme BuChE. Limitations by cross-reactivity are a general challenge for affinity sensors because they have only one "separation plate" whilst lateral flow devices or chromatography columns amplify the separation by repeated binding and dissociation. Nevertheless, the correct measurement by MIP-sensors in real samples of biomarkers which are in the pM range has been claimed in literature [10].  Table S1: MIPs for Enzymes, Table S2