Bromhexine screening using molecularly imprinted polymers for electrochemiluminescence detection and molecular recognition

Abstract

Electrochemiluminescence (ECL) is a highly sensitive detection technique; however, its inherently low selectivity is a significant challenge in analytical and sensing applications. In this study, we developed a system for the specific determination of the pharmaceutical bromhexine (BH). In this system, ECL was combined with a molecularly imprinted polymer (MIP) capable of molecular recognition. A BH-template-based MIP-modified electrode, synthesized by electrolytic polymerization, was prepared. Interestingly, the MIP-modified electrode robustly immobilized the template molecule via peroxidation during electrolytic synthesis. The sensitivity was approximately three times that without further oxidation. X-ray absorption fine structure measurements were performed using synchrotron radiation to characterize interactions between the analyte and the polymer. From the results, a complex mechanism involving hydrogen bonding, π-π interactions, and halogen-π interactions was inferred between the analytical chemical species and the polymer. The fabricated MIP-modified electrode was applied to determine BH in commercially available pharmaceuticals, prescription drugs, and biological urine samples. The results were compared with those from liquid chromatography-mass spectrometry. This system shows potential for development into a selective sensor.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1007/s00216-026-06544-9.

Introduction

Electrochemiluminescence (ECL) detection uses the chemiluminescence generated by electrode reactions. Because ECL is triggered by the electrode reaction, location- and time-control of the measurement is relatively easy [1–3]. Furthermore, excitation in ECL measurements does not require a source like photoluminescence requires excitation light; hence, this method can be applied to semi-opaque samples for the determination of target chemical species [4, 5]. Because ECL can be induced by electrochemical reactions, it is of interest in analytical chemistry for determining pharmaceuticals and drugs in biological samples [6–9].

Tris(2,2′-bipyridine)ruthenium(II) ([Ru(bpy)₃]²⁺) complexes are widely used as probes and have high ECL quantum yields. Numerous chemical species have been applied as co-reactants in [Ru(bpy)₃]²⁺ ECL systems, with aliphatic amines exhibiting particularly strong ECL signals [10, 11]. The aliphatic amine framework is potentially physiologically active in humans [12], and it has been successfully applied to clinical and pharmaceutical analysis using ECL measurements. Paradoxically, selective or specific determination by ECL is difficult. The relatively complex treatments, such as chemical modification of the electrode surface or chromatographic procedures, are needed for analytical applications by ECL detection. The immunoassay technique is frequently used to improve the selectivity of ECL measurements with [Ru(bpy)₃]²⁺, and some analytical kits are commercially available [13]. Unfortunately, the preparation of antibodies for specific target chemical species is relatively complicated and time-consuming. Furthermore, immunoassay kits are expensive, need to be stored under specific conditions, and require relatively complex procedures for analysis [14, 15]. Therefore, an analytical method that recognizes specific analytes is required as an alternative to enzyme and antibody reactions for ECL detection.

Molecularly imprinted polymers (MIPs) are polymers that contain complementary binding sites for functional groups in target chemicals [16–19]. To prepare an MIP, the target analyte is used as a template molecule and is dissolved in a solution with a functional monomer. Polymerization of the polymer forms a membrane containing the template molecule. Next, the template molecules are removed, which leaves blank spaces that are shaped like the template. These blank spaces can specifically adsorb and desorb the target chemical during its separation from a mixture. The adsorption/desorption occurs via the formation of reversible covalent or non-covalent bonds between the functional groups in the MIP and the template molecule. Unlike antibodies, MIPs are robust against conditions like temperature and organic solvent composition. Furthermore, MIPs allow for the tailoring of selective or specific separation and concentration of target analytes.

Free radical polymerization, photopolymerization, molecular self-assembly, and electrolytic polymerization have been reported for the preparation of MIP as selective separation materials [20]. MIP synthesis via electrochemical synthesis has attracted attention because of its simplicity and ability to extend its functionality for the production of sensors. Polypyrrole, polytetrafluoroethylene, poly(3,4-ethylenedioxythiophene), and polyethylene oxide have been used as base substrates for the preparation of MIP membranes [21–24]. However, MIP-modified membranes synthesized by electrolytic polymerization are relatively thin. Consequently, the absolute amount of target chemical species that can be re-adsorbed within an MIP-modified membrane is limited, which restricts the screening capabilities of ECL analysis systems combined with MIP.

For clinical and pharmaceutical analytical applications, such as diagnosis, bioimaging, precise cancer diagnosis, and cancer therapy, electrochemical sensing is conducted using cyclic voltammetry [25], differential pulse voltammetry [26], or square-wave voltammetry [27] with an MIP that provides specific or selective detection for the template chemical [20, 28–32]. Polypyrrole-based MIP systems have been widely used in biomedical diagnostics because they are compatible with living systems [33] and analytical methods for drugs in biological samples.

Bromhexine (BH, Fig. 1) is widely used worldwide as an antitussive medicine, but its prolonged use or abuse can potentially cause fatigue and weakness [34]. Furthermore, nausea, vomiting, dizziness, and anaphylactic symptoms may occur during administration. Consequently, a rapid and simple sensing method is required for BH. This compound has an amine framework and contains aromatic rings and halogen groups, which are valuable for evaluating its behavior when incorporated into MIPs. The aim of this study was to develop and evaluate a detection method using an MIP to address the fundamental selectivity issue in analytical methods using ECL. We selected BH as a representative chemical species for MIP–ECL measurements. A detailed procedure for preparing a modified electrode was developed, and the electrode was applied to ECL measurements. In addition to evaluating basic electrochemical and ECL behavior, detailed chemical speciation was elucidated by analyzing the behavior of chemical species within the polymer using X-ray-absorption fine-structure spectroscopy (XAFS) by synchrotron radiation. A method for quantitative analysis of BH in actual pharmaceutical tablets was developed. Because knowledge of MIP–ECL systems prepared by electropolymerization is limited, we characterized the drugs incorporated into the MIP and developed a technique for simple drug determination.

Fig. 1.

Chemical structure of bromhexine

Experimental

Reagents and materials

Tris(2,2′-bipyridine)ruthenium(II) chloride ([Ru(bpy)₃]Cl₂) hexahydrate, pyrrole, and BH hydrochloride were purchased from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan) and were of the highest grade available. Phosphate-buffered solution (PBS) was prepared by mixing disodium hydrogen phosphate and potassium dihydrogen phosphate solutions with known volumes and concentrations (Wako Pure Chemical Industries Ltd., Osaka, Japan). The pH was adjusted with phosphoric acid or an aqueous sodium hydroxide solution. All other analytical-grade reagents were purchased from Fujifilm Wako Pure Chemical Corp. Distilled water was prepared using an Auto Still WG250 (Yamato Scientific Co. Ltd., Tokyo, Japan).

Preparation of a modified MIP on a glassy carbon electrode

A glassy carbon electrode (ø 3 mm) polished to a mirror finish was used as the working electrode, a silver chloride electrode was used as the reference electrode, and a platinum wire was used as the counter electrode. BH (0.10 mM) was used as the template molecule, pyrrole (5.0 mM) was used as the functional monomer, and PBS (0.10 M, pH 6) containing 0.10 M potassium chloride was used as the polymerization solution. A polypyrrole film was formed on the electrode surface by performing a potential sweep from − 0.6 to 1.0 V at a scan rate of 50 mV s⁻¹ in the polymerization solution. To increase the number of interaction sites between the template BH molecules and the polymer, oxidation was performed (Fig. 2). The electrode was scanned from 0 to 1.2 V for ten cycles at a scan rate of 50 mV s⁻¹ in pH 4.0 PBS. The electrode was then immersed in methanol for 10 min to remove the template molecule and used as the MIP-modified electrode. To quantitatively evaluate the molecular recognition space generated within the MIP-modified electrode, a non-molecularly imprinted polymer (NIP), which lacked such a molecular recognition space, was also prepared as a control. The NIP-modified electrode was prepared under the same conditions but without BH.

Fig. 2.

Preparation of molecularly imprinted polymer-modified electrode

ECL measurements using the MIP- and NIP-modified electrodes

The prepared MIP-modified electrode was immersed in a 0.10 mM BH aqueous solution for 5 min to re-adsorb BH into the polypyrrole film on the electrode surface. After the re-adsorption process, the electrode was washed with water for 30 s under stirring to remove non-specifically adsorbed BH. The electrode was then transferred to 0.10 M PBS (pH 7) containing 0.50 mM [Ru(bpy)₃]²⁺, and ECL measurements were performed using the potential sweep mode.

A schematic diagram of the ECL measurement system is shown in Fig. S1. ECL measurements were performed using a three-electrode system. An MIP-modified glassy carbon electrode (GCE) was used as the working electrode, a silver/silver chloride (Ag/AgCl) electrode was used as the reference electrode, and a platinum wire was used as the counter electrode. The electrode potential was controlled by a potentiostat/galvanostat (200C, Toho Technical Research, Kanagawa, Japan) and a potential sweep unit (HECS 980, Huso, Kanagawa, Japan). The ECL from the electrode surface was detected by a photomultiplier tube (H11901-01, Hamamatsu Photonics, Shizuoka, Japan) positioned opposite the working electrode. The ECL signal was amplified by an amplifier unit (C7319, Hamamatsu Photonics) and recorded using C-LOGGER software through a converter (DI-16TY-USB AD/DA, Contec Co., Ltd., Osaka, Japan). All ECL measurements were performed in a light-proof box.

Sample preparation

A prescription drug (sample A) and over-the-counter combination cold remedies (samples B and C) containing BH were purchased at a pharmacy (December 2025, Matsumoto, Nagano, Japan). To extract BH, each sample was homogenized in a mortar with 10 mL of methanol. The extracted methanol solution was diluted with water, and the diluted solution was used as a test solution. The extract was collected from the test tube, and 100 µL was transferred to the ECL measurement cell. The MIP- or NIP-modified electrode was immersed in the extract for 5 min, and the electrode was washed with water for 30 s under stirring. Next, the electrode was transferred to a solution of 0.50 mM [Ru(bpy)₃]²⁺ in PBS (900 µL, 0.1 M), and the ECL measurement was conducted.

Blank human urine samples were collected from healthy volunteers and stored in a refrigerator at 4 °C without any pretreatment. None of the volunteers had taken BH. An arbitrary amount of BH was added to each urine sample. Next, the MIP- or NIP-modified electrode was directly immersed in the urine sample for 5 min to re-adsorb BH. The ECL measurement was then performed using the procedure detailed above three times (n = 3).

Results and discussion

Electrochemical and ECL behavior of the [Ru(bpy)₃]²⁺/BH system

The ECL mechanism of the [Ru(bpy)₃]²⁺/BH system was characterized using cyclic voltammetry (Fig. 3A) and measurement of the ECL signals (Fig. 3B) for 0.50 mM [Ru(bpy)₃]²⁺ containing 0.50 mM BH or without BH in 0.1 M PBS (pH 7.0) under bare GCE. A redox peak pair of [Ru(bpy)₃]²⁺/[Ru(bpy)₃]³⁺ was observed at approximately 1.10 V (vs. Ag/AgCl) (Fig. 3A). When 0.50 mM BH was present in the solution, an oxidation peak appeared at 0.80 V (Fig. 3A). It has been proposed that electrochemical oxidation of BH proceeds via deprotonation [35]. To evaluate this point, we investigated the pH dependence of the oxidation peak potential of BH. The results revealed that under neutral to weakly alkaline conditions, the reaction mechanism involved two electrons and two protons (Fig. S2). The electrode potential shifted in the positive direction, and an ECL peak was observed close to the oxidation potential of [Ru(bpy)₃]²⁺ (Fig. 3B). Production of ECL by chemicals with an amine framework using [Ru(bpy)₃]²⁺ as the probe involves the generation of radical intermediates by electrochemical oxidation processes, including deprotonation of the co-reactant. In the present study, the formation of radical intermediates was affected by the solution pH (Fig. 3B inset). Although a weak ECL response was observed at weakly acidic pH, the ECL intensity increased when the pH was neutral. BH has a tertiary amine framework and should undergo an ECL mechanism similar to that of the typical ECL for compounds with an amine framework (Eqs. 1, 2, 3 and 4).

Fig. 3.

A Cyclic voltammograms of 50 μM bromhexine (BH) in the presence of 500 μM [Ru(bpy)₃]²⁺ in 0.1 M Britton–Robinson (BR) buffer solution (pH 7.0, red line) and only 500 μM [Ru(bpy)₃]²⁺ (black line) under a bare glassy carbon electrode. B The corresponding electrochemiluminescence (ECL)–potential curves of 50 μM BH in the presence of 500 μM [Ru(bpy)₃]²⁺ in 0.1 M BR buffer solution (pH 7.0, red line) and only [Ru(bpy)₃]²⁺ (black line). The relationship between the ECL intensity and pH (inset). Conditions: scan rate, 50 mV/s; and photomultiplier tube voltage, 850 mV

$$\mathtt{BH}\to {\mathtt{BH}}^{+}+{\mathtt{H}}^{+}+{\mathtt{2}\mathtt{e}}^{-}$$ 1

$${\mathtt{BH}}^{+}\to \mathtt{BH}·+{\mathtt{H}}^{+}$$ 1’

$${[\mathtt{Ru}{(\mathtt{bpy})}_{\mathtt{3}}]}^{\mathtt{2}+}\to {[\mathtt{Ru}{(\mathtt{bpy})}_{\mathtt{3}}]}^{\mathtt{3}+}+{\mathtt{e}}^{-}$$ 2

$${[\mathtt{Ru}{(\mathtt{bpy})}_{\mathtt{3}}]}^{\mathtt{3}+}+\mathtt{BH}·\to {[\mathtt{Ru}{(\mathtt{bpy})}_{\mathtt{3}}]}^{\mathtt{2}+*}+\mathtt{products}$$ 3

$${[\mathtt{Ru}{(\mathtt{bpy})}_{\mathtt{3}}]}^{\mathtt{2}+*}\to {[\mathtt{Ru}{(\mathtt{bpy})}_{\mathtt{3}}]}^{\mathtt{2}+}+\mathtt{h}\nu$$ 4

The pK_a of BH is approximately 7 [35], and it is in a free-base form under neutral-weak base conditions. The electrochemically and subsequently deprotonated radical species of BH (BH·) reduces [Ru(bpy)₃]³⁺ at the electrode surface, and the excited state of [Ru(bpy)₃]²⁺ ([Ru(bpy)₃]^2+*) is generated through electron transfer (Eqs. 1, 2 and 3). The proposed ECL reaction mechanism involves [Ru(bpy)₃]^2+* deactivating to its ground state by emitting light. Because the formation and stability of radical intermediates are affected by deprotonation, the ECL intensity of the [Ru(bpy)₃]²⁺/BH system was higher at high pH values. However, because the solubility of the free-base form of BH is low under high pH conditions, the ECL intensity decreased under alkaline conditions. Besides, it is reported that the electrochemically generated [Ru(bpy)₃]³⁺ can directly oxidize aliphatic amine-type co-reactants [36]. BH may be oxidized to BH⁺ or BH· by [Ru(bpy)₃]³⁺ with a homogeneous reaction. The catalytic ECL reaction occurs when the [Ru(bpy)₃]²⁺ concentration is greater than approximately 0.1 mM. Therefore, the homogeneous catalytic ECL route occurred simultaneously in the BH/[Ru(bpy)₃]²⁺ system.

In the ECL reaction, which occurs via the proposed mechanism for the [Ru(bpy)₃]²⁺/BH system, the ECL intensity is proportional to the concentration of BH. Therefore, the dependence of the ECL intensity on the concentration of BH was investigated (Fig. S3). The ECL intensity increased linearly with the BH concentration up to 100 µM, and the detection limit was approximately 3.7 µM (1.5 µg/mL).

Next, we investigated the effect of interfering species on the ECL reaction. The ECL reaction was also induced by several co-reactants, including aliphatic amines, vitamins, and components of over-the-counter drugs. Ambroxol, diphenhydramine, methylephedrine, and pseudoephedrine, which are frequently used in combination with BH in over-the-counter medications, exhibited ECL activity as co-reactants with the [Ru(bpy)₃]²⁺ system (Fig. S4). Therefore, we next investigated improving the selectivity for the determination of BH.

BH determination using ECL and an MIP-modified electrode

For the selective determination of BH, an MIP-modified electrode was fabricated by electrochemical synthesis. A cyclic voltammogram was obtained by scanning the potential from − 0.6 to 1.0 V in a phosphate buffer solution (PBS, pH 6) containing 0.1 M KCl as dopant, and co-existing 5.0 mM pyrrole (Fig. 4A). In a solution containing 0.10 mM BH, an oxidation peak for BH was observed at approximately 0.8 V during the first potential sweep. This result showed that BH was incorporated into and immobilized within the polypyrrole film formed on the electrode surface during the initial potential sweep. An oxidation peak for BH was not observed during the subsequent potential sweeps. The current response near 0.4 V increased with the number of potential sweeps, which suggested that a pyrrole film formed during the electrolytic synthesis of the MIP-modified electrode. As a control experiment, electrolytic synthesis was performed in the absence of the template molecule BH to prepare a NIP-modified electrode. The current response at 0.4 V increased with the number of scan cycles, confirming the growth of the polypyrrole film (Fig. 4B). The thickness of the polypyrrole film (x cm) can be calculated using the following Eq. (5) based on the amount of electricity consumed during the electropolymerization in the cyclic voltammogram [37].

$$x=\frac{\mathit{qM}}{\rho AzF}$$ 5

Fig. 4.

Cyclic voltammograms obtained under electrochemical synthesis for A the molecularly imprinted polymer (MIP)-modified electrode and B the non-molecularly imprinted polymer (NIP)-modified electrode. The corresponding scanning electron microscopy images of the modified electrode surface at C MIP-modified electrode and D NIP-modified electrode, respectively

M is the molar mass of the pyrrole (67.09 g/mol), F is the Faraday constant (96,500 C/mol), A is the surface area of the working GCE (0.196 cm²), and ρ is the polymer density (1.5 g/cm³). The number of electrons under polymerization (z (mol)) of pyrrole was estimated at 2.25 in previous reports, and the nominal density for polypyrrole films (ρ) was 1.5 g/cm³, respectively [37]. q (C) is the electrical charge in the cyclic voltammogram. Because BH has electrochemical activity, it is difficult to accurately calculate the amount of charge consumed during polymer formation in electropolymerization under the MIP-modified electrode. In this study, the value of q was estimated from cyclic voltammograms obtained under NIP-modified electrode conditions to determine the charge at Fig. 4B. As a result, the thickness of the synthesized polymer film was estimated at approximately 220 nm, consistent with previous studies [37]. The morphologies of the MIP and NIP at the electrode surface were evaluated by SEM (Fig. 4C and D). The MIP-modified electrode was relatively flat, whereas the NIP-modified electrode surface contained cracks. These results suggested that incorporation of template molecules within the membrane filled the voids in the NIP.

Next, the prepared MIP-modified electrode was applied to BH detection. The electrode was washed with methanol to remove the template BH molecules. Subsequently, the MIP-modified electrode was immersed in 0.10 mM BH solution for 5 min. After re-adsorption of BH, ECL measurements were performed on the MIP-modified electrode in a solution containing 0.50 mM [Ru(bpy)₃]²⁺ and showed a weak ECL signal (Fig. 5A). Pyrrole functional groups in the polypyrrole membrane will have weak chemical interactions with BH in the MIP. Because the polypyrrole membrane is not robust, the molecular recognition space is destroyed when template molecules are removed by the washing process [38]. Consequently, the absolute amount of BH re-adsorbed on the simple MIP-modified polypyrrole electrode was relatively low, which resulted in a weak ECL signal. This result indicated that additional functionalization of the MIP-modified electrode was necessary.

Fig. 5.

Electrochemiluminescence (ECL)–potential curves of 0.50 mM [Ru(bpy)₃]²⁺ in 0.1 M Britton–Robinson (BR) buffer solution (pH 7.0) for the molecularly imprinted polymer-modified electrode A before and B after further oxidation processes. Conditions: scan rate, 50 mV/s; and photomultiplier tube voltage, 850 mV

When a potential more positive than 1.0 V is applied to a polypyrrole membrane in aqueous solution, oxidation occurs and introduces carbonyl or carboxyl groups into some of the pyrrole units [39]. These introduced functional groups will enhance the interactions between template molecules and the polymer. Consequently, we investigated the effect of oxidation of the polymer membrane on BH re-adsorption. The MIP-modified electrode containing the template molecule was subjected to ten scans from 0 to 1.2 V to oxidize the polymer membrane. After washing the electrode with water, the electrode was immersed in an aqueous solution containing 1.0 mM BH for 5 min. To remove non-specifically adsorbed BH on the electrode surface, the electrode was washed with water and then transferred to 0.50 mM [Ru(bpy)₃]²⁺ in PBS (pH 7.0) for ECL measurement. The ECL signal at 1.1 V was approximately three times that obtained without the additional oxidation process (Fig. 5B). These results showed that oxidation of the polymer improved its robustness and physical stability, which preserved the molecular recognition space in the MIP. Furthermore, oxidation likely increased BH re-adsorption by enhancing interactions between the functional groups and the MIP. By contrast, when using the NIP-modified electrode, the oxidation current at 0.8 V for BH re-adsorption was weak (Fig. S5A). The corresponding ECL response was similar in intensity to the background signal (Fig. S5B). Overall, these results suggested that there was a specific molecular recognition space in the MIP membrane.

X-ray absorption fine structure (XAFS) measurements were performed to evaluate the behavior of BH in detail. The Br K-edge XAFS spectrum of the bromine from BH incorporated into the MIP was similar to that of the BH standard (Fig. S6-1). The bromine signal in the MIP-modified electrode was homogeneously distributed at the working electrode (Fig. S6-2). BH incorporated in the MIP was stabilized by p-p stacking, p-halogen interactions, and hydrogen bonding with the pyrrole groups and oxide framework of the polymer (Fig. 6). Because BH was increased by performing a further oxidation of the MIP-modified electrode, the hydrogen bonding between carbonyl-functionalized groups in the polypyrrole membrane plays a particularly important role in the re-adsorption of BH.

Fig. 6.

Re-adsorption of BH to the molecularly imprinted polymers (MIPs)

Analytical performance of the MIP-ECL measurement for BH determination and application to real samples

To evaluate the characteristics of the prepared MIP-modified electrode, the ECL behavior of BH re-adsorption was investigated. The MIP-modified electrode was immersed in a solution containing 0.10 mM BH for 5 min, washed, transferred to a solution containing 0.50 mM [Ru(bpy)₃]²⁺, and then subjected to ECL measurements. The ECL response was weak for the NIP-modified electrode under all pH conditions, which indicated that non-selective re-adsorption of BH had little effect. For the MIP-modified electrode, the ECL response was weak under acidic conditions. The amount-of-substance fraction of BH, which has a pK_a of approximately 7, is shown in Fig. 7A. At acidic pH values, the protonated form of BH was predominant. This predominance suggested that BH was not captured within the molecular recognition space of the MIP at these pH values. Increasing the pH of the immersion solution resulted in a large increase in the ECL intensity under near-neutral conditions. The increase in the ECL intensity was attributed to the generation of the free-base form of BH, which would be more readily re-adsorbed onto the MIP. However, because the free-base form of BH has low solubility in solution, the ECL response decreased at pH 10.

Fig. 7.

A Mole fraction of bromhexine (BH) and B the relationship between (ECL) intensity and pH of the immersion solution for the molecularly imprinted polymer (MIP)-modified electrode (red bars) and the non-molecularly imprinted polymer (NIP)-modified electrode (blue bars)

After establishing that a neutral pH was optimal for the immersion solution, the effect of the immersion period was investigated. The ECL response was measured after immersing the modified electrode in a 0.10 mM BH solution at pH 7 for a fixed period, followed by transfer to a 0.50 mM [Ru(bpy)₃]²⁺ solution (Fig. 8). For the NIP-modified electrode, the ECL response did not greatly increase with increases in the immersion period (Fig. 8A). For the MIP-modified electrode, the ECL intensity increased with increases in the immersion period. However, at immersion times exceeding 5 min, the ECL intensity remained nearly constant. It was inferred that the increase in ECL decreased due to saturation of the absolute amount of BH molecular recognition space in MIP. The saturation of the molecular recognition space on the MIP-modified electrode depended on the BH concentration used for impregnation (Fig. S7). Therefore, a shorter immersion time provides a wider detection range for quantitative BH analysis. However, while improved quantifiability was expected under short immersion times, the sensitivity was insufficient. Therefore, we set the immersion time to 5 min as an optimal condition that allows for semi-quantitative detection.

Fig. 8.

A Dependence of the electrochemiluminescence (ECL) intensity on the immersion time in bromhexine (BH) solution using electrodes modified with the molecularly imprinted polymer (MIP, red data points) and non-molecularly imprinted polymer (NIP, blue data points). B The calibration curve of BH

The dependence of the ECL response obtained with the MIP on the BH concentration in the immersion solution was measured at pH 8 with immersion for 5 min (Fig. 8B). A linear relationship was observed between the BH concentration and the ECL response for concentrations up to approximately 50 mM. The detection limit was approximately 3.1 mM (1.0 µg/mL). At high concentrations outside the linear range, saturation of the MIP occurred due to limited availability of the MIP recognition site. Next, the repeatability of the fabricated electrodes was investigated. After the ECL measurement, the MIP-modified electrode was washed and re-adsorbed with BH. ECL measurements were performed again on the electrode, and the ECL intensity increased. The ECL intensity increased with increasing the number of repetitions (Fig. S8). [Ru(bpy)₃]²⁺ was adsorbed and accumulated in the MIP. However, when modified electrodes were prepared independently for each measurement, the initial ECL profile showed intra-day reproducibility within 10%. Furthermore, the MIP-modified electrode remained stable for more than 1 week at room temperature. These results suggest that the proposed MIP-modified electrode is applicable as a disposable on-site analytical sensor.

Because the determination of BH by the MIP-modified electrode was feasible, we investigated the selectivity using different chemical species in the immersion solution (Fig. 9). Almost no response was observed for ECL measurements using the MIP-modified electrode on ambroxol, which is structurally similar to BH and one of its metabolites. Furthermore, the ECL signals for diphenhydramine, methylephedrine, and pseudoephedrine, which are often included as active ingredients in medicines, were weak. These chemical species, which contain secondary and tertiary amine structures, could not be distinguished using ECL measurements with the [Ru(bpy)₃]²⁺ system. By contrast, the proposed method, which uses ECL in combination with an MIP, provided high sensitivity for BH detection in the presence of interfering compounds. Consequently, we investigated applying this method to the measurement of BH in actual pharmaceuticals.

Fig. 9.

Comparison of the re-adsorbed amounts of bromhexine (BH), ambroxol, diphenhydramine, dl-methylephedrine, and pseudoephedrine on the BH-templated molecularly imprinted polymer (MIP) and non-molecularly imprinted polymer (NIP)-modified glassy carbon electrodes (GCE)

An MIP-modified electrode was used to determine BH in prescription and over-the-counter medicines dissolved in methanol and diluted with distilled water (Table 1). The ECL obtained using MIP exhibited a response corresponding to the BH concentration contained in sample A. For this sample, the ECL results were consistent with the indicated BH concentration on the product packaging and showed good correspondence with the liquid chromatography-mass spectrometry (LC–MS) results. The LC–MS conditions were provided in the supplementary information. Samples B and C contained several compounds that interfered with the ECL measurements of BH. In particular, methylephedrine, dextromethorphan, dihydrocodeine, and chlorpheniramine provide positive interference to ECL responses due to their aliphatic amine structures. Therefore, the selective detection of BH by conventional ECL measurement under bulk conditions was essentially difficult. However, the BH concentrations measured by the proposed MIP-modified electrode were close to the indicated concentrations. The determined concentrations of BH corresponded to the LC–MS results. Upon immersion of the MIP-modified electrode in the sample solution, only BH was re-adsorbed onto the polymer membrane. Therefore, the proposed method was expected to enable selective detection of BH in medicines containing ECL-active chemical species.

Table 1.

Results for the determination of bromhexine (BH) in pharmaceutical formulations using the molecularly imprinted polymer-modified electrode

	ECL		RSD (%)‡	LC–MS detection rate (%)†	Indicated value
	Detected	Detection rate (%)†
Sample Aa (prescription)	4.5 ± 0.4 (mg/tablet)	112	10	104 ± 2.4	4 (mg/tablet)
Sample Bb (OTC medicine)	4.4 ± 0.4 (mg/tablet)	109	11	107 ± 1.0	4 (mg/3 tablets)
Sample Cc (OTC medicine)	11 ± 0.6 (mg/tablet)	91	5.6	102 ± 3.9	4 (mg/9 tablets)

Components:

^aLactose hydrate, hydroxypropylcellulose, and magnesium stearate

^bLactose hydrate; dextromethorphan hydrobromide, methylephedrine, acetaminophene, chlorpheniramine maleate, calcium ascorbate, and riboflavine

^cClemastine fumarate, acetaminophene, dihydrocodeine phosphate, methyephedrine phosphate, anhydrous caffeine, thiamine disulfide, riboflavine, lactose hydrate, hydroxypropylcellulose, magnesium stearate, and titanium oxide

^†The detection ratio indicates the percentage of Found value relative to the amount added (Detection ratio (%) = Indicated value/Detected value × 100)

^‡RSD, relative standard deviation (n = 3)

Next, we applied the proposed MIP-ECL method to the analysis of human urine samples (Table 2). Reasonable detection rates were obtained in all three urine samples, and the results corresponded to LC–MS results. Urine contains numerous impurities, including a biological matrix. In our previous study, ascorbic acid, which co-exists at millimolar concentrations, strongly affects the ECL response [40–43]. Therefore, pretreatment methods such as solvent extraction or solid-phase extraction were required in the analytical protocol for chromatography. In contrast, the method proposed in this study allows selective capture of BH by simply immersing the urine sample. Table S1 compares the conventional BH analytical method with the MIP-modified electrode proposed in this study. LC–MS, GC–MS, and capillary electrophoresis, which are conventionally used for BH analysis, have sensitivity and selectivity. However, these methods were often unsuitable for on-site analysis because of their size or weight. Additionally, the analytical cost of these instruments was relatively high, and the operating process required specialized technical skills. By contrast, the MIP-modified electrode proposed in this study provided a simple extraction procedure by immersion in the sample solution. In addition, ECL measurements can be completed in a few minutes and miniaturized for on-site analysis at a low cost.

Table 2.

Bromhexine determination in urine samples using the developed electrochemiluminescence ECL sensor with a molecularly imprinted polymer-modified glassy carbon electrode, and comparison with the liquid chromatography-mass spectrometry (LC–MS) method

	Added (µM)	ECL			LC–MS
		Found (µM)	Detection rate (%)a	RSD (%)b	Detection rate (%)a
Urine A (male)	0	N.D.c	-	-	N.D.c
	20	23.2 ± 1.8	116	8.1	106.7 ± 3.8
	50	55.8 ± 1.8	111	3.8	99.9 ± 2.5
Urine B (female)	0	N.D.c	-	-	N.D.c
	20	22.4 ± 3.1	112	14	111.8 ± 9.1
	50	52.3 ± 4.3	105	8.3	91.2 ± 1.2
Urine C (male)	0	N.D.c	-	-	N.D.c
	20	21.1 ± 0.9	99.8	4.7	115.0 ± 3.1
	50	50.5 ± 1.0	101	1.9	92.2 ± 0.5

^aThe detection ratio indicates the percentage of Found value relative to the amount added (Detection ratio (%) = Added concentration/Found concentration × 100)

^bRSD  relative standard deviation (n = 3)

^cN.D not detected

Conclusions

A selective measurement system that coupled ECL with an MIP was developed for BH determination. First, an MIP-modified electrode was prepared using a simple potential sweep method for the electrochemical synthesis of biocompatible polypyrrole. Further oxidation of the prepared polypyrrole enabled the immobilization of the template molecule, and incorporation of this step increased the amount of BH re-adsorbed to three times that obtained without oxidation. Behavior of BH within the MIP membrane was characterized by electrochemical analysis and XAFS measurements. The MIP-modified electrode was applied to the analysis of pharmaceutical and biological samples and enabled selective detection of BH. The prepared modified electrode shows potential as a simple screening technology because it can selectively capture analytes when simply immersed in the test solution. In the future, the template molecule can be changed to prepare MIP-modified electrodes for different analytes.

Supplementary Information

Below is the link to the electronic supplementary material.

Biographies

Fumiki Takahashi

is an Associate Professor of Faculty of Science, Shinshu University, Japan. He received a PhD degree in Applied Chemistry, Graduate School, Division of Science from Shinshu University in 2010. His main research interests are the development of on-site sensing systems for forensic analytical chemistry. Research on detection technology by electrochemiluminescence and the a rapid and simple separation method using ultrasound were reported in Analytical and Bioanalytical Chemistry.

Shota Ohno

is an MS candidate in the Shinshu University, Japan. His main research interests are the development of the electroanalytical sensor, and chromatographic selective determination.

Masachika Yoshida

is an MS candidate in the Shinshu University, Japan. His main research interests are the development of the selective determination sensor based on a molecularly imprinted polymer.

Yuta Harayama

is an Assistant Professor in the Department of Legal Medicine, Faculty of Medicine, Shinshu university, Japan from 2022. His research focuses on DNA detection and body fluid identification using capillary sequencing, massively parallel sequencing, and immunochemical approaches. His main research interests are the develop of a rapid and sensitive analytical methods for forensic applications.

Kanya Kobayashi

(M.S.) has been an Assistant Professor in the Department of Legal Medicine at Shinshu University School of Medicine since 2019. His specialty is the application of mass spectrometry to forensic toxicology. He has experience conducting toxicological analyses related to the Aum Shinrikyo sarin gas attacks, which shocked the world. He is currently focusing his research on postmortem drug redistribution.

Katsuya Nakamura

is an Associate Professor of Clinical Neurology at Shinshu University, Japan. He received his Ph.D. in Medicine from Shinshu University. His expertise lies in clinical genetics of neurological diseases. His research focuses on the application of analytical chemistry to develop novel diagnostic biomarkers for neurological disorders.

Takuma Kaneko

is a program-specific researcher at Kyoto Univ., Japan. His research interests include material analysis in fuel cells and water electrolysis. He also became a devices at SPring-8. researcher of Japan Synchrotron Radiation Research Institute (JASRI) in October 2025. He is engaged in constructing new beamline for the material analysis of hydrogen energy devices at SPring-8.

Tomoya Uruga

became researcher of Japan Synchrotron Radiation Research Institute (JASRI) in 1996. He constructed beamlines (experimental facilities of a synchrotron radiation laboratory) concerning X-ray absorption fine structure (XAFS) and developed XAFS measurement systems at SPring-8. He is engaged in constructing new beamline dedicated to the analysis of the hydrogen energy devices.

Yasuo Seto

became Group Director of Forensic Science Group at RIKEN Sprig-8 Center in July 2018 from the former Director position of National Research Institute of Police Science (Kashiwa, Japan). In NRIPS, he engaged in R&D of laboratory analysis and on site detection for chemical warfare agents and biological toxins and forensic examination on Sarin gas attack incidents. Now, his research mission is R&D of forensic science technologies by adopting synchrotron radiation analysis such as fingerprint detection, synthetic fiber discrimination and trace component distribution profiling of drug powders.

Hirosuke Tatsumi

is a Professor at Shinshu University, Japan. He studies fundamental and practical electroanalytical chemistry. He is especially interested in automatic renewal of working electrodes. His recent research includes determination of potential of zero charge and real potentials of ions. He is currently a board member of the Japan Society for Analytical Chemistry (JSAC) and the Editor-in-Chief of Review of Polarography, published by the Polarographic Society of Japan (PSJ).

Jiye Jin

is Professor of Analytical Chemistry in the Department of Chemistry, Faculty of Science at Shinshu University, Matsumoto, Nagano, Japan. His research focuses on the development of electrochemical methods integrated with luminescent techniques for chemical and bioanalytical applications. His work spans diverse areas, including the development of analytical methodologies for evaluating antioxidant activity and studies in sonochemistry and photocatalytic reactions.

Author contribution

Fumiki Takahashi: conceptualization, methodology, formal analysis, writing—original draft. Shota Ohno: formal analysis, writing—original draft. Masachika Yoshida: formal analysis. Yuta Harayama: validation, writing—review and editing. Kanya Kobayashi: validation, writing—review and editing. Katsuya Nakamura: writing—review and editing. Takuma Kaneko: methodology, formal analysis, writing—review and editing. Tomoya Uruga: methodology, formal analysis, writing—review and editing. Yasuo Seto: formal analysis, writing—review and editing. Hirosuke Tatsumi: writing—review and editing. Jiye Jin: writing—review and editing.

Funding

Open Access funding provided by Shinshu University. This work was supported by a Grant-in-Aid for Scientific Research (B) and (C) from the Japan Society for the Promotion of Sciences (JSPS) (Grant Numbers 20K10553, 22H01732, 23H03177, 23K04787, and 24K10637). This work used research equipment shared in the MEXT Project for Promoting Public Utilization of Advanced Research Infrastructure (Program for Supporting Construction of Core Facilities) (Grant Numbers JPMX0440100023, JPMX0440100024, and JPMX0440100025). Part of this work was supported by Shinshu University Gender Equality Promotion Centre, Researcher Assistance Program. We thank Gabrielle David, PhD, from Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.

Data availability

Data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval

All experiments were conducted under the Ethics Regulations for Research with Human Subjects at Shinshu University (H28-No. 160 and R4-No. 328).

Conflict of interest

The authors declare no competing interests.