Introduction

Isometamidium (ISM), as a long-acting antitrypanosomal, can block the nucleic acids synthesis of trypanosome by inhibiting the activities of RNA and DNA polymerases (Shweta 2013). Up to now, ISM has been used for chemotherapy and chemoprophylaxis of the disease in cattle, sheep, and goats under conditions of natural tsetse challenge for more than 50 years (Kazunari 1999; Wesongah et al. 2004). However, the ISM can last long and cause teratogenesis and embryotoxicity when it was used in livestock. As a result, the animal-derived foods can be easily contaminated by the overuse of ISM (Geerts S 1999). For this reason, many countries and organizations have regulated the maximum residual limits (MRLs) of ISM, especially in milk which is an important food of humans for all ages. JECFA and Japan have regulated the MRL of ISM in milk is 1.0 mg/L. (Kazunari 1999). China has regulated MRL of ISM in milk at 100 μg/L. (Liu 2011) As a result, reliable and robust analytical techniques are required for food contaminant surveillance. So far, a number of laboratory based analytical techniques have been developed for determination of the ISM, such as ELISA (Ardelli 2000; Delespaux 2002; Whitelaw 1991) and RIA methods (Kinbo 1988). HPLC (Schad et al. 2012) and LC-MS/MS methods (Li et al. 2011). Compared with these methods, the electrochemical sensors (EC) provide an accurate and rapid analytical tool for chemical contaminant detection in the environmental and food matrix samples (Diculescu et al. 2006; Jain et al. 2009; Yin et al. 2012). More recently, various nanomaterials have been widely employed to improve the performance of electrode interface by improving sensitivity, selectivity, and multiplexing capacity. Among these materials, gold nanoparticles (AuNPs) and carbon nanotube (CNT) are the most commonly used types. AuNPs have been used as excellent scaffolds for the fabrication of novel electrochemical sensors (Cao et al. 2011; Chen et al. 2015; Choi et al. 2011; Zhou et al. 2013). due to its distinct physical and chemical attributes, such as a favorable microenvironment, high surface-to-volume ratio, good biocompatibility, and high conductivities between biomolecules and the electrode surface (Li et al. 2014; Marie-Christine 2004; Saha et al. 2012). CNTs have a superconductivity property when its diameter is less than 6 nm (Kociak 2001). and has been widely used as a material to amplify the electrochemical signals (Agui et al. 2008; Gomez et al. 2015; Herrasti et al. 2014; Jacobs et al. 2010; Silva et al. 2014). The hybrid of CNT with AuNPs was also reported to potentially improve the performance of electrochemical sensors. The SWCNT/AuNP-coated electrode was used directly to improve the sensitivity of chloramphenicol analysis in milk samples (Xiao et al. 2007). Li et al. used a CNT drop method and electrochemically deposited AuNPs to improve the sensitivity of pesticide triazophos analysis in vegetable samples (Li et al. 2012). The bottleneck of these methods is that the homogeneity of CNTs dropped directly on the GCE surface is hard to be controlled, and the CNT-film can be easily detached. Therefore, the approach to stabilize the CNT-film on a GCE surface is also of concern.

To the best of our knowledge, the electrochemical properties of ISM and the mechanisms underlying its voltammetric behavior have not been studied. An EC method for ISM analysis in milk has not yet been developed as well. In this study, we investigated the electrochemical behavior of ISM over a wide pH range and put forward the possible reaction mechanism using CV and DPV. We also pursued the modification of GCE by SWCNT/AuNPs. In order to improve the SWCNT/AuNP stability and better performance on GCE surface, a diazotization and electropolymerization method was used. Moreover, the AuNPs were electrochemically deposited by optimizing the deposition time to obtain the best performance for ISM analysis. Finally, an EC sensor method for ISM analysis in milk samples were developed, which can be potentially used for commercial market surveillance. The EC sensor method was validated by comparison with a LC-MS/MS method.

Experimental

Materials and reagents

Isometamidium chloride standard was purchased from Wako (Wako Pure Chemical Industries, Ltd, Japan). The molecular structure of ISM was shown in Fig. 1. Methanol, acetonitrile (MeCN), and formic acid (HPLC-grade) were purchased from Tedia Company (Fairfield, USA). Ammonium formate, phenylamine, p-phenylenediamine, sodium nitrite, hydrochloric acid, potassium chloride, disodium, trisodium, sodium o-phosphate and o-phosphoric acid, hydrogen tetra-chloroaurate (HAuCl4 · 4H2O) was guaranteed reagent from Sinopharm Chemical Reagent Co., Ltd, China. Single-walled carbon nanotube (SWCNT, O.D. 1–2 nm) was obtained from Sigma-Aldrich (MO, USA). A 2.5 g/L ISM stock solution was prepared by dissolving the appropriate mass in ultrapure water. Working standards were then prepared by dilution of the primary stock solution with ultrapure water. Ultrapure water was obtained from a Milli-Q Academic System, (Millipore, USA). Solutions of disodium, trisodium, sodium o-phosphate and o-phosphoric acid were made at a concentration of 0.2 M by dissolving the appropriate mass in ultrapure water. These were then titrated together to give the desired pH plus 0.1 M KCl. An appropriate volume was then added directly to the voltammetric cell and diluted with sufficient ultrapure water to give an overall phosphate concentration of 0.1 M.

Fig. 1
figure 1

The molecular structure of the isometamidium

Full size image

Instrumentation

Cyclic voltammetry was performed using an Autolab potentiostat interfaced to a PC for data acquisition and instrument control via the Nova operating system (Autolab, The Netherlands). The voltammetric cell contained a graphite rod counter electrode, a silver/silver chloride reference electrode (Ag/AgCl) (Chen Hua Instruments, Shanghai, China) and a 3 mm diameter glassy carbon electrode as the working electrode (Chen Hua Instruments, Shanghai, China). Field emission scanning electron microscope (FE-SEM, Helios Nanolab 600i, FEI Company, USA) was used to characterize the morphology of the material. ISM was analyzed by a Waters AcquityTM UPLC with a BEH C18 column (100 mm × 2.1 mm, 1.7 μm, Waters, USA) and the MS/MS instrument (AB SCIEX QTRAP, AB SCIEX, USA).

Voltammetric behavior of ISM with bare electrode

The glassy carbon working electrode was polished to a mirror by using 1, 0.3, and 0.05 μm alumina pastes, respectively, and was then thoroughly ultra-sonicated in distilled water and ethanol before used. The electrode was dried under nitrogen flow and was placed into phosphate buffer solutions at pH values between 2.0 and 12.0 containing 0.25 g/L ISM. Cyclic voltammograms were initially recorded in plain solutions of phosphate buffer, then in the same solution containing ISM. For initial studies a starting potential of 0.0 V with a switching potential of +1.2 V was employed. Differential pulse voltammetry was taken using a starting potential of 0.0 V and a final potential of +1.2 V; using a step height of 10 mV, pulse repetition time 0.2 s, pulse amplitude of 100 mV, and pulse duration of 50 ms.

Preparation of SWCNT/AuNP-functionalized electrodes

In this study, we used a diazotization method to immobilize the CNT which was easier to control and could be more stable. Modification of GCE includes two steps. First, the SWCNT was covalently linked to the surface of GCE by using diazo-reaction. The SWCNT-modified electrode was further modified by HAuCl4 to form SWCNT/AuNP-functionalized electrodes. The p-phenylenediamine modified GC electrode GC–Ph–NH2 was prepared as reported previously (Anjum S 2015; Sun and Ma 2012; Zhang et al. 2011) with the following modification. A clean GCE was dipped in HCl solution (0.5 mol/L) containing p-phenylenediamine (6 × 10−3 mol/L) and aniline (4 × 10−3 mol/L). NaNO2 (10 mg) was added to the reaction mixture and left to react overnight. The modified GCE was sonicated in ultrapure water, acetonitrile and ultrapure water to eliminate all adsorbed molecules and finally dried under a stream of nitrogen gas.

The carefully rinsed electrode (GC–Ph–NH2) was immersed overnight in 1 mL of an SWCNT dispersion (1 mg/mL) in HCl (0.5 mol/L) that contained NaNO2 (10 mg). Then the modified GCE was sonicated in ultrapure water, acetonitrile and ultrapure water and finally dried under a stream of nitrogen gas as previously described.

For the formation of the SWCNT/AuNP-modified electrode, the GCE was immersed into 1 mM HAuCl4 containing 0.1 M KCl solution which was deaerated by nitrogen gas for about 10 min before use and the gold nanoparticles were electrodeposited via potentiostatic deposition method with optimization of deposition time and voltage (vs. Ag/AgCl) (Atta et al. 2012). The resulting electrode (SWCNT/AuNPs-GCE) was removed and rinsed with ultrapure water. The process was shown in Fig. 2.

Fig. 2
figure 2

Schematic representation of SWCNT and AuNP assembly on GCE

Full size image

Methods development for ISM analysis in bovine milk samples by EC and LC-MS/MS

An EC method for analyzing ISM residues in bovine milk samples was developed. The skimmed milk was used for sample extraction and was obtained by centrifugation at 6000 rpm for 30 min for three times to remove the fat layer. The sample extraction procedure was described as follows: a total of 5 g of skimmed milk was extracted in a 50 mL polypropylene centrifuge tube by adding 20 mL of a mixed solvent containing ACN and methanol (1:1). The methanol contains 0.25 mol/L of ammonium formate. The mixture was shaken vigorously for 30 min and the supernatant was collected after centrifugation at 10,000 rpm for 10 min. The extraction procedure was repeated twice. All extracts were collected, and transferred into a 50-mL polypropylene centrifuge tube and dried under the nitrogen gas at 45 °C. The residue was further reconstituted in 2 mL of water, and was used for EC analysis using a DPV method after dilution four times by PBS buffer (pH = 7.0).

For the development of a LC-MS/MS method, an extraction procedure similar to EC method was used, except that the final reconstituted solvent was passed through 0.22 mm filter paper (Jinteng company, Tianjin, China), and then transferred to inserts for analysis. Mobile phases A and B were methanol and 0.2 % of formic acid in water, respectively. The gradient elution program was shown in Table S1 (Supplementary Information). The flow rate was 0.30 mL/min and the injection volume was 5 μL. The MS/MS instrument was operated using an electrospray ion source in positive mode (ESI+), and monitored using multiple reaction monitoring (MRM). The product Ion spectrum and the MRM chromatogram of ISM were shown in Fig. S1 (Supplementary Information) and Fig. S2 (Supplementary Information), respectively. In addition, the spectrometry parameters were listed in Table S2 (Supplementary Information).

Results and discussion

Electrochemical behavior of isometamidium at a glassy carbon electrode

For finding the adsorptive character of the ISM at GCE, a cyclic voltammogram of ISM at 0.25 g/L in 0.1 M phosphate buffer (pH = 7.0) was recorded after 60 s preconcentration at 0 V with different scan rate as reported previously (Atta et al. 2011). As the sweep rate increased from 20 to 200 mV/s; the peak current increased steadily. Plots of peak current (i p) vs. the square root of scan rate (v ½) showed a linear dependence, indicating the electrochemical process to be diffusion controlled. However, when the pH was greater than 10, the linearity was not obvious. To investigate the influence of pH for the ISM electrochemical behavior, the voltammograms of the solution containing 2.5 mg/L of the isometamidium were taken in 0.1 M phosphate buffer solutions at pH values between 2.0 and 11.0. Oxidation peaks were observed on voltammogram, which corresponding to the amine group of ISM. The peak currents and potentials were dependent on pH, and the peak potential shifted towards more negative values as the pH increased gradually, indicating involvement of proton in the electrode process. Similar results were also found in previous study (Diculescu et al. 2006; Menek 1998).

Characterization of the SWCNT/AuNP composite

The FE-SEM was used to characterize the morphology of the hybrid nanocomposite. As illustrated in Fig. 3 (I), a long and tortuous SWCNT was covalently bound by the diazonium ions onto the surface of the GCE. Figure 3 (II) displays the SEM image of the SWCNT/AuNP-modified GCE and the top left inset exhibits the magnified view of the electrodepositied gold-nanoparticals. The size and shape of each AuNP were observed clearly from the inset image. Large amounts of AuNPs were embedded or wrapped on the SWCNT. This structure could effectively increase the surface-to-volume ratio, so it could reduce the electron transfer and result in the excellent electrical conductivity (Huang et al. 2015).

Fig. 3
figure 3

SEM image of SWCNT (I) and SWCNT/AuNPs (II) (Inset is the magnified view)

Full size image

Optimization of AuNPs electrodeposition time

To show the effect of the deposition time of the AuNPs on the capability of electron transfer, the performance of the electrode fabricated in HAuCl4 solution with 100, 200, 300, 400, 500, 600, and 700 s for deposition of AuNPs was studied with DPV in 0.1 M phosphate-buffer saline (pH 7.0) containing 250 μg/L ISM (Fig. 4). The data indicated that the electrochemical sensor fabricated by the AuNPs for 600 s has a better capability for the electron transfer (Fig. 4, curve a ~ f).

Fig. 4
figure 4

The DPVs of the ISM (250 μg/L, in 0.1 M pH 7.0 phosphate buffer) at SWCNT-GCEs modified with AuNPs for different deposition time (a 100 s, b 200 s, c 300 s, d 400 s, e 500 s, f 600 s, g 700 s)

Full size image

Optimization of accumulation potential and accumulation time

Figure 5 (I) shows the effect of accumulation potential on the resulting i p magnitude of the oxidation peak using an accumulation time of 300 s in a 250 μg/L ISM solution (0.1 M pH 7.0 phosphate buffer). Peak current was found to increase from −0.5 V to a maximum value between −0.2 V and +0.2 V (vs. Ag/AgCl) which was then found to decline at potentials more positive than this. Consequently, further studies were carried out using an accumulation potential of 0.0 V (vs. Ag/AgCl). Figure 5 (II) indicates the effect of increasing accumulation time at an applied potential of 0.0 V using a 250 μg/L ISM solution. The oxidation stripping peak (at about 0.5 V) was found to increase with increased accumulation time and reached a maximum value at ca. 300 s.

Fig. 5
figure 5

(I) Effect of accumulation potential for a 250 μg/L ISM solution in pH 7.0, 0.1 M phosphate buffer. Accumulation time: 300 s; accumulation potential: −0.5, −0.4, −0.3, −0.2, −0.1, 0.0, 0.1 0.2 0.3, 0.4, and 0.5 V. (II) Effect of accumulation time for a 250 μg/L ISM solution pH 7.0, 0.1 M phosphate buffer. Accumulation potential: 0.0 V; accumulation time: 60, 120, 180, 240, 300, 360, and 600 s

Full size image

Detection of ISM with the SWCNT/AuNP-modified electrochemical sensor

The SWCNT/AuNPs-GCE exhibits good electrocatalytic activity towards the ISM. Figure 6 (I) shows the DPVs of the ISM (250 μg/L), in 0.1 M pH 7.0 phosphate buffer at the bare GCE (curve b) and the SWCNT/AuNPs-GCE (curve a), respectively. It is obvious that the electrochemical signal of the peak becomes stronger and a little wider because of the effect of the nano materials modified on the surface of the GCE. The current increases swiftly after +0.7 V, this might due to the occurrence of Au peak after +1.1 V. For further analysis of ISM with the SWCNT/AuNP-modified GCE, the potential between 0.0–+0.7 V was used.

Fig. 6
figure 6

(I) The DPVs of the ISM (250 μg/L, in pH 7.0, 0.1 M phosphate buffer) at the SWCNT/AuNPs-GCE (curve a) and bare GCE (curve b) from 0.0 V ~ +0.7 V. (II) The calibration curve of the linear dependence of cathodic peak current on the ISM concentration at the SWCNT/AuNPs-GCE (curve a) and bare GCE (curve b) ( Inset was the linearity range from 10 ~ 400 μg/L for curve a)

Full size image

A calibration study was carried out using DPV in phosphate buffer (0.1 M, pH 7.0), over a concentration range 10 μg/L ~ 1000 μg/L at the SWCNT/AuNPs-GCE and 200 μg/L ~ 2000 μg/L at the bare GCE. The linear range for ISM at the SWCNT/AuNPs-GCE is 10 ~ 400 μg/L, with the plot slope of 0.011 μA/ (μg/L) and a correlation coefficient R 2 = 0.998 (Fig. 6 (II) curve a). The limit of detection (LOD) is estimated to be 3 μg/L (signal-to-noise ratio at 3). In comparison, the linear range for ISM at the bare GCE is 200 ~ 1000 μg/L, with the plot slope of 0.001 μA/ (μg/L) and a correlation coefficient R 2 = 0.994. Thus, the limit of detection for ISM using the modified GCE is far higher than that using the bare GCE.

Interferences, reproducibility, and stability

To validate the performance of the modified GCE, the interferences, reproducibility, and stability of the EC method were evaluated. Some interferences may exist in raw milk or be generated, derived from the preparation of pure milk, such as K+, Na+, Ca2+, CO3 2−, PO4 3−, and so on. These impurities can possibly interfere with the quantitative detection of ISM in milk. Therefore, the interferences of these species were conducted with the modified GCE at the optimized condition by DPV for detection of 250 μg/L ISM. Thirteen types of metal and acidic ions, i.e., K+, Na+, Ca2+, Fe2+, Mg2+, Zn2+, Cl−, I−, SO4 2−, NO3 −, CO3 2−, PO4 3−, and CH3COO− which were reported previously to be the common interfering substances in buffer solution were used for interference test (Lin et al. 2013; Sun et al. 2015; Wang et al. 2014). No interference was observed to the developed EC method, until the substances reached a rather high concentration (≥100-fold of ISM). In addition, to evaluate the reproducibility and stability of the same electrode and between electrodes, the AuNPs-GCE was prepared separately five times with use of the same GCE, as well as fabrication of five different electrodes for measurement. The oxidation peak current from a 250 μg/L ISM solution in N2 saturated PBS (0.1 M, pH 7.0) was analyzed. The relative standard deviation (RSD) of the response was 2.89 % and 5.4 %, respectively, indicating that the reproducibility of the modified electrode was satisfactory.

Analytical application: determination of the ISM in commercial milk samples

Milk samples were spiked with ISM at three concentrations (50, 100, and 200 μg/L). The results were shown in Table 1. It can be seen that the average recoveries are in the range of 79.4–94.7 % using the established EC method (n = 10). We also prove that the EC method is comparable with LC-MS/MS method (Table 1). The results demonstrate that the electrochemical sensor is adequate to practical application for detecting ISM residue in milk.

Table 1 Comparison of EC and LC-MS/MS method for ISM analysis spiked in milk samples. Data were presented as mean ± SD
Full size table

Conclusions

This is the first report to investigate the voltammetric behavior of isometamidium, a molecule with a broad range of applications in clinical activities. We also developed a facile and effective method based on the diazo-reaction and electropolymerization to fabricate SWCNT/AuNP-modified GCE. This fabrication exhibited more stable and sensitive character. The prepared GCE was applied for the analysis of isometamidium in pure milk samples. The method was proved to be accurate when comparing with the reference HPLC-MS/MS method. The satisfactory performance of the electrochemical biosensor made it a qualified alternative for ISM determination in practical and routine analyses with lower cost and quicker detection rate.