Development of a spiramycin sensor based on adsorptive stripping linear sweep voltammetry and its application for the determination of spiramycin in chicken egg samples

Herein, an adsorptive stripping linear sweep voltammetric technique was described to determine spiramycin, a macrolide antibiotic, using a carboxylic multiwalled glassy carbon electrode modified with carbon nanotubes. The main principle of the analytical methodology proposed was based on the preconcentration of spiramycin by open-circuit accumulation of the macrolide onto the modified electrode surface. As a result of the adsorption affinity of spiramycin to the modified surface, the sensitivity of the glassy carbon electrode was significantly increased for the determination of spiramycin. The electrochemical behavior of spiramycin was evaluated by cyclic voltammetry and the irreversible anodic peak observed was measured as an analytical signal in the methodology. The proposed electrochemical sensing platform was quite linear in the range of 0.100–40.0 µM of spiramycin concentration with a correlation coefficient of 0.9993. The limit of detection and the limit of quantification were 0.028 and 0.094 µM, respectively. The intra- and interday repeatability of the proposed sensor was within acceptable limits. Finally, the applicability of the electrochemical methodology was examined by determining the drug content of chicken egg samples spiked with spiramycin standard. A rapid and easy extraction technique was performed to extract spiked spiramycin from the egg samples. The extraction technique followed had good recovery values between 85.3 ± 4.0% and 93.4 ± 1.9%.

Millipore Elix 5 Water Purification System (Burlington, MA, USA) was used to obtain ultrapure water. A stock solution of spiramycin at 800 mg L -1 was prepared in methanol. Standard solutions were prepared by diluting the stock solution with ultrapure water.

Instrumentation
Voltammetric measurements were performed by using an Ivium Compactstat analyzer (Eindhoven, the Netherlands) with triple electrode system involves an Ag/AgCl (sat. KCl) reference electrode, a bare GCE (3 mm in diameter), and a platinum wire. Measurement with cyclic voltammetry (CV) and linear sweep voltammetry (LSV) were carried out between 500 and 1500 mV. The scan rate was 100 mV s -1 .
The ultrasound bath was Daihan (Seoul, Korea), WUC-D10H. A Heidolph Hei-Vap (Schwabach, Germany) rotary evaporator was used in the extraction step. Millipore Elix 5 Water Purification System was used to produce ultrapure water.

Pretreatment of MWCNTs-COOH
Initially, an adequate amount of commercial MWCNTs were boiled in nitric acid to produce carboxylic groups in the chemical structure by oxidation. Subsequently, the oxidized MWCNTs (MWCNTs-COOH) were rinsed multiple times by ultrapure water and were dried at 40 °C overnight. Finally, ready-to-use suspension was prepared by dispersing 5 mg of the MWCNTs-COOH in 5 mL of dimethylformamide [40].

Fabrication of MWCNTs-COOH/GCE and the other electrodes decorated with extra modifications on MWCNTs-COOH/GCE
Initially, GCE was polished with alumina slurry on an alumina polishing pad. Ten microliters of the prepared carboxylic MWCNTs suspension was dropped on the GCE surface and dried under infrared lamp to obtain MWCNTs-COOH/GCE. The poly Eriochrome Black T (EBT)/MWCNTs-COOH/GCE was fabricated by electropolymerization of EBT monomers (5.0 mM in 0.1 M NaOH) on the MWCNTs-COOH/GCE by using CV in a potential range of -0.4 V and 1.5 V at 100 mV s -1 for 10 cycles [41][42][43]. After the electropolymerization process, the fabricated electrode was rinsed by ultrapure water to remove excess EBT that could be physically adsorbed on the electrode surface. Au and Pt nanoparticles (NPs) were doped on the MWCNTs-COOH/GCE by CV at 100 mVs -1 for 10 cycles between 1.0 V and -1.5 V in 1 mM HAuCl 4 solution and 0.2 V to -1.0 V in 1 mM H 2 PtCl 6 solution to fabricate AuNPs/MWCNTs-COOH/GCE and PtNPs/ MWCNTs-COOH/ GCE, respectively [44].

Surface area of the electrode
The electro-active surface areas of the electrodes were calculated via the Randles-Sevcik Eq. (1) by CV technique at different scan rates using 10 -3 M K 3 Fe(CN) 6 in 0.1 M KCl solution. Peak current (Ip) is as follows for a reversible process at 298 K [40]: where n is the number of electrons transferred, D 0 is the diffusion coefficient (7.6 × 10 -6 cm 2 s -1 ), υ is the scan rate, A is the surface area of the electrode, and C 0 * is the concentration of K 3 Fe(CN) 6 .

Extraction method
The method previously proposed by Wang et al. [9] was slightly modified and was followed to extract spiramycin from chicken eggs. Spiramycin blank chicken egg samples were homogenized to combine yolk and albumen at room temperature. Two grams of homogenized egg sample was spiked with proper quantities of spiramycin standard. The spiked sample was kept at room temperature for 2 h. Afterwards, 0.5 mL of 0.1 M EDTA and 7.5 mL of ACN-H 2 O (90:10, v/v) solutions were added to the sample and it was extracted in an ultrasonic bath for 15 min, and centrifuged at 3000 rpm at 4 °C for 10 min. Five milliliters of the supernatant was evaporated by using a rotary evaporator at 45 °C. The residue was redissolved in 1 mL of methanol and was filtered through a 0.45 μm PTFE filters before analysis.

Electrode mechanism of spiramycin
The electrochemical behavior of spiramycin on MWCNTs-COOH/GCE was examined by CV. The CVs of 5 × 10 -5 M spiramycin in 0.2 M HNO 3 obtained after completion of the open-circuit accumulation process of spiramycin onto the MWCNTs-COOH/GCE were displayed in Figure 1. A well-defined oxidation peak was detected at 1230 mV, but no corresponding cathodic peak was observed on the reverse scan. Therefore, the electrode reaction of spiramycin was determined to be irreversible. Subsequent sweeps, without stirring, resulted in significant decreases in the oxidation peak current of spiramycin, showing that spiramycin was strongly adsorbed onto the modified electrode surface. After the third scan, the subsequent peak currents remained almost unchanged.
The electrode process of midecamycin, another macrolide with similar molecular structure, was discussed previously [45][46][47] and it was attributed to the oxidation of aldehyde group on the macrocycle in the molecule. Similarly, it was estimated that the anodic peak observed for spiramycin was obtained by the oxidation of the aldehyde group on the molecule to a carboxylic acid by losing two electrons.
The CVs of 5 × 10 -5 M spiramycin in different electrolyte solutions and a chart representing the influence of different electrolyte solutions on the anodic peak current of spiramycin were shown in Figures 2 and 3, respectively. In cases where the pH was greater than 1, spiramycin showed irreversible two-step one electron oxidation reaction in CVs (see Figure  2, pH = 8 and 10 data were not shown). As the pH decreased, the second oxidation peak began to disappear. At pH = 1, the second anodic wave appeared as a shoulder peak just next to the main oxidation peak. Finally, the oxidation reaction of spiramycin occurred rapidly in one step in 0. HNO 3 solution provided higher anodic peak current (see Figure 3). Therefore, the electrolyte solution was 0.2 M HNO 3 in subsequent studies.
The influence of pH on the anodic peak potential of spiramycin was also investigated. As pH decreased from 10 to 1, the anodic peak potential shifted linearly to more positive potential, implying that proton transfer should be accompanied in the oxidation process. Linear chart was inset in Figure 2. The equation was E (mV) = -60.712 pH (mV/pH) + 1117. The Nernst equation slope of -60.7 mV/pH indicates that the ratio of the participated protons to the transferred electrons was 1:1. Consequently, the possible numbers of electron and proton involved in the oxidation of spiramycin were estimated to be two.

The voltammetric behavior of spiramycin and influence of stirring on accumulation
The effects of the accumulation time and stirring the electrolyte on the oxidation peak current of spiramycin are indicated in Figure 4A. Firstly, the variation in the oxidation peak current of 5 × 10 -5 M spiramycin versus accumulation time was examined on both the bare and the MWCNTs-COOH/GCE by using CV in 0.2 M HNO 3 . The anodic current of spiramycin increased as a function of accumulation time on the modified electrode. However, no important variation was observed in the anodic current obtained on the bare GCE. This result was attributed to the preconcentration of the spiramycin on the fabricated MWCNTs-COOH/GCE surface by open-circuit accumulation. Furthermore, the influence of introducing stirring the electrolyte solution on the oxidation peak current of spiramycin was also observed. It was clear from the results that stirring the electrolyte solution during preconcentration of spiramycin on the fabricated electrode surface provided both an important increase in the oxidation peak current and a shorter period to reach the maximum anodic current of spiramycin. Stirring times greater than 60 s were not preferred as they resulted in the desorption of spiramycin gradually from the surface. Therefore, the accumulation time with stirring was selected as 60 s.
The CVs of 5 × 10 -5 M spiramycin on bare GCE and the modified electrode were presented comparatively in Figure 4B. As can be seen from the figure, the proposed modification of the electrode surface significantly increased the sensitivity of GCE for spiramycin determination. Compared with the bare GCE, the anodic peak current of spiramycin increased approximately 2 × 10 4 times on MWCNTs-COOH/GCE after completion of the accumulation process of spiramycin onto the modified surface. The adsorption affinity of spiramycin on the fabricated electrode surface allowed the described methodology to be used as a quite sensitive spiramycin sensor.
On the other hand, the effect of stirring rate between 100 and 500 rpm on the anodic peak current of spiramycin was also investigated. The results were depicted in Figure 5. The analytical signal obtained increased up to a stirring rate of 300 rpm. However, stirring rates over 300 rpm reduced the measured anodic peak current of spiramycin. According to the results, the optimum stirring rate was found to be 300 rpm.
As stated in the Randles-Sevcik equation, the surface areas of the bare GCE and MWCNTs-COOH/GCE were calculated to be 0.0334 ± 0.0025 and 0.2281 ± 0.0107 cm 2 from the slope of the plot of Ip versus υ 1/2 , respectively. The results presented that the modification of the surface increased the electro-active area of the electrode up to approximately seven times.

Effect of scan rate
The relationship between sweep rate (v) and peak current (Ip) was examined on MWCNTs-COOH/GCE and the oxidation peak currents of spiramycin were determined to be proportional to sweep rates between 25 and 500 mV s -1 (Fig 6A). In addition, slopes of log Ip vs. log v was found to be approximately 0.90 which was quite greater than the expected value of 0.53 for a purely diffusion-controlled process ( Figure 6B). Both results obtained were in agreement and supported that the oxidation reaction of spiramycin on the modified electrode was adsorption-controlled process.

The influence of additional modifications on MWCNTs-COOH/GCE
It has been shown in many studies that metal nanoparticles and conductive thin film polymer modifications on GCE provide improvement in sensitivity by increasing the catalytic effect for numerous analytes. To this end, several additional   and PtNPs/MWCNTs-COOH/GCE was given in Figure 7. The best signal obtained on MWCNTs-COOH/GCE proved that the adsorption interaction between spiramycin and MWCNTs-COOH was reduced in the presence of the additional modifiers on MWCNTs-COOH/GCE.

Linearity and sensitivity
Quantitative analysis of spiramycin on fabricated MWCNTs-COOH/GCE was carried out using the LSV technique. The anodic peak currents of a series of standard spiramycin solutions obtained with the technique described were plotted versus concentration to evaluate the linearity of the described electrochemical sensing platform. The oxidation peaks were displayed in Figure 8A. The proposed sensor showed a good linearity between 0.100 and 40.0 µM of spiramycin concentration with a correlation coefficient of 0.9993 ( Figure 8B). The limit of detection (LOD) and the limit of quantification (LOQ) of the sensing platform were calculated to be 0.028 and 0.094 µM, respectively, by using the formulas LOD = 3.3SD/b and LOQ = 10SD/b, where SD is the standard deviation of ten reagent blank determinations and b is the slope of the calibration curve [48].
The linear working ranges and the sensitivity properties of the proposed technique and the existing electrochemical methods for the determination of spiramycin were comparatively given in Table 1. The results clearly indicated that the described sensing platform would be the most sensitive electrochemical technique ever proposed for spiramycin detection.

Repeatability
The calibration standards of three different concentrations of 0.500, 5.00, and 40.0 µM were used to examine the intraday and interday repeatability of the described methodology. In all cases RSD values were less than 8% (Table 2). Thus, the proposed platform was proven to be highly repeatable in spiramycin analysis. In addition, the stability of the proposed sensing platform was evaluated by measuring the anodic response of 5 × 10 -5 M spiramycin on MWCNTs-COOH/GCE after the fabricated electrode was stored at room temperature for 7 days. The sensor retained 96% and 79% of the initial response of spiramycin on days 3 and 7, respectively.

Real sample analysis and extraction recovery
The applicability of the proposed methodology was examined by determining the amount of spiramycin spiked in chicken egg samples at two different levels. For this purpose, 2 g of spiramycin blank chicken egg samples (albumen and yolk combined) were spiked with 13.57 (low level) and 271.40 µg (high level) of spiramycin standards and then the described extraction procedure was followed (n = 3). The spiked samples were analyzed (the representative LSVs and standard addition calibration were given in Figure 9) and the amounts of analyte recovered were calculated. According to the results, the technique followed had good recovery values with 85.3 ± 4.0% and 93.4 ± 1.9% in spiramycin extraction for low and high concentration levels. Finally, the analytical features of the described methodology are summarized in Table 3.

Conclusion
Spiramycin is a macrolide antibiotic frequently used to treat several infections in poultry. Therefore, it is important to analyze this drug in poultry products before human consumption. In this study, electrochemical behavior of spiramycin were investigated by CV and an accurate, sensitive, simple, and rapid analysis method was proposed for the drug on a MWCNTs-COOH/GCE by Ad-SLSV. Initially, the electrode mechanism of spiramycin was predicted and the possible irreversible oxidation reaction was indicated using the results obtained from the CV analysis. Then, the validation of the Ad-SLSV technique proposed as a result of the open-circuit accumulation of spiramycin onto the modified surface was performed and the optimum conditions for the analysis were determined. Consequently, the validated electrochemical sensing platform was proven to be an effective method for quantitative spiramycin determination by improving the sensitivity of GCE. The technique would be the most sensitive electrochemical method ever proposed for spiramycin analysis.