Article Cite This: Anal. Chem. 2019, 91, 7546−7553
pubs.acs.org/ac
“Gate Effect” in p‑Synephrine Electrochemical Sensing with a Molecularly Imprinted Polymer and Redox Probes Patrycja Lach,† Maciej Cieplak,*,† Marta Majewska,† Krzysztof R. Noworyta,† Piyush Sindhu Sharma,† and Wlodzimierz Kutner*,†,‡ †
Institute of Physical Chemistry, Polish Academy of Sciences, Kasprzaka 44/52, 01-224 Warsaw, Poland Faculty of Mathematics and Natural Sciences, School of Sciences, Cardinal Stefan Wyszynski University in Warsaw, Wóycickiego 1/3, 01-815 Warsaw, Poland
Downloaded via UNIV OF SASKATCHEWAN on August 23, 2019 at 06:46:27 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
‡
S Supporting Information *
ABSTRACT: The “gate effect” mechanism for conductive molecularly imprinted polymer (MIP) film coated electrodes was investigated in detail. It was demonstrated that the decrease of the DPV signal for the Fe(CN)64−/Fe(CN)63− redox probe with the increase of the p-synephrine target analyte concentration in solution at the polythiophene MIP-film coated electrode did not originate from swelling or shrinking of the MIP film, as it was previously postulated, but from changes in the electrochemical process kinetics. The MIP-film coated electrode was examined with cyclic voltammetry (CV), differential pulse voltammetry (DPV), electrochemical impedance spectroscopy (EIS), and surface plasmon resonance (SPR). The MIP-film thickness in the absence and in the presence of the psynephrine analyte was examined with in situ AFM imaging. Moreover, it was demonstrated that doping of the MIP film was not affected by p-synephrine binding in MIP-film molecular cavities. It was concluded that the “gate effect” was most likely caused by changes in radical cation (polaron) mobility in the film.
M
of analyte binding.6 This mechanism of enhanced or, more often, hindered diffusion of the redox probe caused by MIP film swelling or shrinking is widely adopted to account for operation of all MIP film-based chemosensors. Physical blocking of the electrode surface by adsorbed large nonconductive molecules (e.g., proteins) might be another reason for hindering redox probe diffusion through the MIP film.7 However, both of these mechanisms do not explain all possible phenomena that may be responsible for the redox probe current change corresponding to analyte binding, especially in terms of conductive MIP films. Herein, we present a detailed study of the “gate effect” mechanism accounting for operation of a conductive MIP film based chemosensor. As a model of the MIP chemosensing system, we used a functionalized polythiophene film imprinted with p-synephrine SYN (Scheme 1). As a dietary supplement, encountered in an extract from bitter orange, SYN effectively causes weight loss.8 It is becoming more and more popular since the United States Food and Drug Administration (FDA) has prohibited ephedrine as a dietary supplement. SYN is an efficient agonist of a β3-adrenergic receptor (ADRB3) that can mainly be found in a brown adipose tissue.9 Importantly, SYN overdosing is considered to cause such serious undesirable effects as elevated blood pressure and heart rate, cardiac
olecularly imprinted polymers (MIPs) are synthetic tailored materials dedicated, among others, for mimicking selective biological recognition.1 They provide binding of target analytes with selectivity almost as high as that of natural receptors. Due to their unique properties, they have found numerous applications in fabrication of selective chemosensors.2 MIP-based chemosensors are comparable to biosensors in terms of sensitivity, selectivity, and detectability. However, they are superior because of ease of fabrication, durability, and tolerance to harsh experimental conditions, such as elevated or lowered temperatures, high ionic strength, high solution acidity or basicity, as well as the presence of heavy metal ions and organic solvents. Especially, conductive MIP films are recently becoming more and more popular.3 This is mainly because these MIPs can readily be deposited on electrodes, by electropolymerization, as thin films.4 For electrochemical determination of electroinactive analytes with an MIP chemosensor, often some external redox probe is added to the test solution. This approach was firstly described for theophylline determination, using the theophylline imprinted poly(methacrylic acid-co-ethylene glycol dimethacrylate).5 Binding of target analyte molecules by molecular cavities caused MIP film swelling. This swelling resulted in the change of the film permeability to redox probe molecules or ions, thus significantly increasing faradaic currents corresponding to oxidation or reduction of the redox probe. Therefore, this effect was called the “gate effect”. Later, it appeared that the faradaic currents of the probes were also decreased because © 2019 American Chemical Society
Received: November 29, 2018 Accepted: May 24, 2019 Published: May 24, 2019 7546
DOI: 10.1021/acs.analchem.8b05512 Anal. Chem. 2019, 91, 7546−7553
Article
Analytical Chemistry
were carried out at room temperature, 20(±1) °C, by using an electrochemical V-shaped mini cell (Scheme S1). They were performed using 0.1 M phosphate buffered saline, PBS (pH = 7.4), in the presence of the redox probe, vis., 0.1 M K3[Fe(CN)6] and 0.1 M K4[Fe(CN)6] or 0.1 M Ru(NH3)6Cl3. Therefore, in the DPV measurements in the presence of the redox probe, the potential was scanned from 0 to 0.60 V or from −0.50 to 0.10 V vs Ag/Ag+ pseudoreference electrode, respectively, with a potential step of 5 mV. The amplitude of 50 ms pulses applied was 25 mV. In the EIS experiments, an alternating current excitation signal of frequency in the range of 1 MHz to 100 mHz and 10 mV sinusoidal amplitude was used at open-circuit potential (OCP). The electrochemical system was approximated with a modified Randles-Ershler equivalent circuit, Rs+CPE/(Rct + W0), with Z-View software (Scriber Associates, Inc.), in which Rs, Rct, CPE, and W0 are solution resistance, charge-transfer resistance, constant-phase element, and Warburg impedance, respectively. To evaluate selectivity of the devised chemosensor, we have constructed calibration plots for interfering compounds that have spatial distribution of functional groups similar to those of SYN, namely adrenalin and creatinine, or that may be found in real samples, vis., urea and glucose. Determinations of all interfering compounds were performed in the same way as for the SYN analyte, on the same MIP-SYN film-coated electrode. It means that the electrodes, the redox probe solution, and the concentration ranges were the same, just interfering compounds were determined instead of SYN. The selectivity factors were calculated by dividing the slope of the calibration curve of the MIP-SYN film-coated electrode for SYN by the slopes of the calibration curves separately for each interference. The imprinting factor was calculated in a similar way, that is, by dividing the slope of the calibration curve for SYN recorded on the MIP film coated electrode by the slope of calibration curve obtained on the nonimprinted (NIP) film coated electrode. Synthesis and Deposition of the MIP-SYN Film. The MIPSYN film was prepared by oxidative electropolymerization under potentiodynamic conditions (with five potential cycles) over the potential range of 0 to 1.30 V vs Ag/AgCl pseudoreference electrode at a scan rate of 50 mV s−1. A solution of 10 μM SYN (template), 30 μM 2,2′-bithiophene-5carboxylic acid functional monomer (FM), 300 μM 2,3′bithiophene cross-linking monomer (CM), and 100 mM tetrabutylammonium perchlorate [(TBA)ClO4] in acetonitrile was used for this electropolymerization. Before MIP-SYN film deposition, the electrode was cleaned with the “piranha” solution for 10 min and then mirror-finished with 0.05 mm alumina slurry. (Warning: the “piranha” solution is very dangerous upon contact with skin or eye.) SYN Template Extraction from the MIP-SYN Film. After electropolymerization, the template was extracted from the resulting MIP-SYN film by immersing the film coated electrode in 0.1 M NaOH for 60 min at room temperature, 20(±1) °C. After each SYN determination, the MIP-SYN filmcoated electrode was immersed in the same solution, until the SYN analyte was completely extracted, as confirmed by a stable, maintained at its maximum value, DPV peak of the redox probe. Synthesis of the Control Nonimprinted Polymer (NIP) Film. The NIP films were prepared by oxidative electropolymerization under the same conditions as those used for MIP-SYN film preparation but in the SYN template absence.
Scheme 1. (a) Structural Formula and (b) the DFT B3LYP/ 6-31g(d) Optimized Pre-Polymerization Complex Structure at Room Temperature in Acetonitrilea
a The box includes structural formulas of the p-synephrine SYN (template), 2,2′-bithiophene-5-carboxylic acid FM (functional monomer), and 2,3′-bithiophene CM (cross-linking monomer).
arrhythmia, myocardial infarction, and sudden death.10 On the other hand, there are suggestions that trace amounts of such alkaloids as octopamine, tyramine, or m-synephrine, are responsible for these adverse effects and SYN is completely safe to use.11 So far, several procedures have been developed to determine SYN in analyzed samples. In these procedures, mostly highperformance liquid chromatography (HPLC) with ultraviolet (UV) absorbance detection,12 gas chromatography-mass spectrometry (GC-MS),13 liquid chromatography (LC) with UV detection,14 or high-performance thin-layer chromatography (HPTLC) with densitometric detection15 are used. The HPLC and GC techniques are time-consuming as well as expensive equipment and trained operators demanding. Spectrophotometric methods typically require time-consuming separation of the analyte from its complex matrix. Apparently, there is a need for devising a chemosensor for fast and selective SYN determination. Moreover, SYN imprinting attempts are scarce and limited to the synthesis of an imprinted membrane for selective SYN separation16 and to resins for solid-phase extraction (SPE) of SYN from the samples.17
■
EXPERIMENTAL SECTION Procedures. Electrochemical Measurements. All cyclic voltammetry (CV), differential pulse voltammetry (DPV), and electrochemical impedance spectroscopy (EIS) measurements 7547
DOI: 10.1021/acs.analchem.8b05512 Anal. Chem. 2019, 91, 7546−7553
Article
Analytical Chemistry Moreover, the NIP-film coated electrode was immersed in 0.1 M NaOH for 60 min at room temperature, for similar treatment before measurements. Surface Plasmon Resonance (SPR) Measurements. All SPR measurements were accomplished using an AUTOLAB ESPIRIT spectrometer connected to the PGSTAT12 potentiostat/galvanostat, in a manner described above. Au-coated BK7 glass SPR chips and 0.1 M PBS (pH = 7.4) were used. After each analyte solution injection, 10 to 30 CV cycles in the range of 0 to 0.40 V vs Ag/Ag+ pseudoreference electrode, at a scan rate of 50 mV s−1, were applied to ensure stable SPR readout.
■
RESULTS AND DISCUSSION Selection of the Functional Monomer. Selecting appropriate functional monomers is the crucial step of successful MIP film preparation. In solution, these monomers should form a prepolymerization complex with the SYN template, sufficiently stable to survive electrochemical copolymerization with cross-linking monomers. After SYN template extraction, molecular cavities of the desired shape and size are generated in the resulting MIP. Careful inspection of possible interaction patterns provided by the SYN molecule led us to select the functional monomer bearing the carboxyl group, FM (Scheme 1). This group is dissociated in a neutral aprotic polar solvent solution. Therefore, it may serve as a hydrogen bond acceptor for hydroxyl groups of SYN. Moreover, the negatively charged carboxyl group may electrostatically interact with the SYN amine group. To confirm this assumption, we performed DFT calculations at the B3LYP/6-31g(d) basis set level for a prepolymerization complex, at room temperature, in acetonitrile (Scheme S2). The optimized structure of this complex in the SYN:FM molar ratio of 1:3 is shown in Scheme 1. The calculated Gibbs free energy change (ΔG) of complex formation was sufficiently negative (−227.4 kJ mol−1) to indicate that this complex may be formed in the solution, and then transfer to the polymer indeed. Preparation of the MIP-SYN Film. The MIP-SYN film selective with respect to SYN was simultaneously synthesized and deposited on the electrode surface by electropolymerization under potentiodynamic conditions according to the procedure described above. For that purpose, five consecutive potential cycles were executed (Figure 1a). In the first cycle, an anodic peak appeared at 1.20 V vs Ag/Ag+ pseudoreference electrode. This peak can be assigned to electro-oxidation of the 2,2′- and 2,3′-bithiophene moieties of the functional, FM, and cross-linking monomer, CM, leading to radical cation formation.18 Then, in the second and subsequent cycles, current increased in the potential of 1.0 to 1.10 V vs Ag/Ag+, thus demonstrating that more conjugated, easily oxidizable oligo-/polythiophene chains were synthesized. Moreover, this current increased in consecutive cycles, thus indicating that the deposited polymer was conductive. After MIP-SYN film deposition, SYN template molecules were removed by extraction with 0.1 M NaOH, thus vacating imprinted molecular cavities in the MIP matrix. This extraction was monitored by DPV in the presence of the K4[Fe(CN)6]/ K3[Fe(CN)6] redox probe (Figure 1b). After extraction exceeding 60 min, there was no further change in the DPV peak height. We characterized NIP and MIP-SYN films both before and after template extraction using polarization-modulation infrared reflection−absorption spectroscopy (PM-IRRAS, Figure
Figure 1. (a) Current−potential curves for deposition of an MIP-SYN film by potentiodynamic electropolymerization on a 1 mm diameter Pt disk electrode with five consecutive potential cycles at a scan rate of 50 mV s−1 in the acetonitrile solution of 10 μM SYN template, 30 μM FM, 300 μM CM, and 100 mM (TBA)ClO4 as the supporting electrolyte. (b) DPV curves for the MIP-(p-synephrine) film (1) before and after SYN extraction with 0.1 M NaOH for (2) 20, (3) 40, (4) 60, and (5) 80 min.
S1) and X-ray photoelectron spectroscopy (XPS, Table S1) to unravel the chemical compositions of these films. Both measurements confirmed successful deposition of the films. Additionally, a decrease in the nitrogen and the increase in the sulfur atoms relative content in the MIP-SYN film after template extraction (Table S1) confirmed removal of SYN template molecules from the cavities. Characterization of Analytical Performance of the MIP-SYN Chemosensors. We used the template-extracted MIP-SYN film-coated electrodes for SYN determination. In the DPV determination using the K4[Fe(CN)6]/K3[Fe(CN)6] redox probe, the DPV peak linearly decreased (Figure 2a) with the increase of the SYN concentration in the range of 0.1 to 0.99 μM (Figure 2b), thus obeying the linear regression equation of (IDPV,0 − IDPV,s)/IDPV,0 = 11.83 × 10−2(±1.44 × 10−2) + 3.21 × 10−4(±0.24 × 10−4) cSYN [nM] with the correlation coefficient, R2 = 0.979. The SYN limit of detection (LOD) was 12.2 nM at a signal-to-noise ratio, S/N = 3. Moreover, the chemosensor was quite selective. That is, selectivity coefficients for the creatinine, adrenalin, urea, and glucose interferences were 2.08, 2.63, 2.05, and 2.46, 7548
DOI: 10.1021/acs.analchem.8b05512 Anal. Chem. 2019, 91, 7546−7553
Article
Analytical Chemistry
Figure 2. (a) The DPV curves recorded in the presence of the redox probe, 0.1 M K4[Fe(CN)6] and 0.1 M K3[Fe(CN)6], in 0.1 M PBS (pH = 7.4), for the MIP-SYN film coated Pt disk electrode after SYN template extraction with 0.1 M NaOH, and then in the presence of (1) 0, (2) 0.10, (3) 0.29, (4) 0.48, (5) 0.74, and (6) 0.99 μM SYN. (b) Calibration plots (1 and 3−6) for the MIP-SYN and (2) the NIP film coated electrodes for (1 and 2) SYN, (3) urea, (4) creatinine, (5) adrenalin, and (6) glucose.
Figure 3. (a) The EIS Nyquist and (insert) Bode plots for the MIPSYN film coated Pt disk electrode after SYN template extraction with 0.1 M NaOH for 60 min, and then in the presence of SYN at concentrations indicated with numbers at curves. (b) Calibration plots for (1 and 3−6) the MIP-SYN and (2) the NIP film coated electrodes for (1 and 2) SYN, (3) urea, (4) creatinine, (5) adrenalin, and (6) glucose. All experiments were performed in the presence of the redox probe, 0.1 M K4[Fe(CN)6] and 0.1 M K3[Fe(CN)6], in 0.1 M PBS (pH = 7.4) at OCP, in the frequency range of 1 MHz to 0.1 Hz.
respectively (Figure 2b), and the calculated apparent imprinting factor was IF = 2.7 (Figures 2b and S4). Determined from EIS measurements (Figure 3a), the linear dynamic concentration range of 0.1 to 0.99 μM SYN satisfied the linear regression equation of ΔRct [Ω] = 75.38(±11.48) [Ω] + 0.47(±0.02)cSYN (nM) with the correlation coefficient, R2 = 0.9926. The LOD for SYN was 5.69 nM at S/N = 3. The selectivity coefficients for the creatinine, adrenalin, urea, and glucose interferences were 9.97, 9.65, 10.90, and 11.27, respectively, and the apparent imprinting factor was, IF = 6.31 (Figures 3b and S5). Investigation of the “Gate Effect” Mechanism. Herein, we thoroughly discuss origins of the DPV signal change of a redox probe at the MIP film coated electrode with the change of the target analyte concentration in solution (Figure 2). There are few conceivable mechanisms to account for this change. These mechanisms can be divided into two groups. One involves the change of the rate of diffusion of the redox probe through the MIP-SYN film while the other is related to the change of the electrochemical properties of the film itself. Diffusion Related Mechanisms. Swelling and Shrinking of the MIP Film. This mechanism was first proposed to explain
the gating encountered in the CV determination of theophylline at the theophylline imprinted nonconductive poly(methacrylic acid-co-ethylene glycol dimethacrylate) film.5 In this determination, binding of the target analyte molecules by molecular cavities caused swelling of the MIP film. This swelling resulted in pore widening in the MIP film, thus enhancing diffusion of the K4[Fe(CN)6] redox probe through the film to the electrode surface. This enhanced diffusion significantly increased faradaic current corresponding to oxidation of the redox probe at the electrode−(MIP film) interface. Therefore, this effect was called the “gate effect”. Later, it occurred that the faradaic current also decreased because of analyte binding in the MIP film.6 This redox probe hindered diffusion should be well manifested in the low frequency region of the EIS spectrum. Therefore, we recorded a series of EIS curves for different SYN analyte concentrations in order to verify this mechanism (Figure 3a). In a low frequency region of Nyquist plots recorded for the MIP-SYN film coated electrodes, only straight lines of Wartburg impedance were observed. Slopes of these lines, equal to 42(±1)°, were independent of the SYN concentration in solution. However, an increase in the charge transfer resistance 7549
DOI: 10.1021/acs.analchem.8b05512 Anal. Chem. 2019, 91, 7546−7553
Article
Analytical Chemistry with the SYN concentration increase was quite well manifested. Moreover, Bode plots (Figure 3a insert) demonstrated a pronounced increase of the phase angle with the increase of the SYN concentration, thus indicating a change of the double-layer capacitance. Apparently, the diffusion related mechanism of the gating is inadequate to account for operation of the conductive MIP-SYN film coated electrodes. We have already encountered a quite similar behavior of other conductive MIP film coated electrodes.19 To support the above conclusion, we coated SPR chips with MIP-SYN and NIP films (Figures S6 and S7). The SPR signal changed with the refractive index changes of the insulating phase, remaining in a close contact with the gold surface. Therefore, film swelling caused by analyte insertion into molecular cavities should push out water from the close vicinity of the gold surface. This would manifest itself with a pronounced refractive index change and, hence, the SPR signal change. Importantly, this signal was independent of the SYN concentration up to 330 μM (Figure 4), still much higher than
Figure 5. In situ AFM determination of MIP-SYN film thickness in air, then in 0.1 M PBS (pH = 7.4), and then in 0.5 μM SYN in 0.1 M PBS (pH = 7.4).
AFM determined thickness of 15.7(±0.3) nm quite well agreed with that of 17.6 nm estimated from changes of the SPR angle recorded during MIP-SYN film deposition (Figure S6).20 Physical Blocking of the Redox Probe Diffusion Through the MIP Imprinted Cavities. This mechanism of blocking of the redox probe diffusion is postulated for surface imprinting of macromolecules in thin (
