MWCNTs-CTAB and HFs-Lac Nanocomposite-Modified Glassy Carbon Electrode for Rutin Determination

Rutin is a flavonoid glycoside compound, which is mainly transported via the blood circulation system in the human body. The monitoring of the blood concentration of rutin is of great significance in many fields such as pharmacology and pharmacokinetics. In this work, a biosensor based on multi-walled carbon nanotubes (MWCNTs), cetyltrimethylammonium bromide (CTAB), hydroxyl fullerenes (HFs), and laccase (Lac) nanocomposite-modified glassy carbon electrodes was constructed. The modified materials were characterized with a transmission electron microscope (TEM), cyclic voltammograms (CV), and electrochemical impedance spectroscopy (EIS). CTAB is used to disperse MWCNTs and improve hydrophilicity and biocompatibility of MWCNTs, while the use of Lac can enhance the oxidation of catechol structure in rutin, thus significantly improving the sensitivity and selectivity of the modified electrode. Linear sweep voltammetry (LSV) studies showed that the determination linear ranges of rutin were 0.1 µmol L−1 to 2 µmol L−1 and 2 µmol L−1 to 11 µmol L−1, with the determination limits of 30 nmol L−1 and 95.5 nmol L−1, respectively. The proposed biosensor can be used to detect rutin tablets and serum samples with high recovery, which indicates a good accuracy of this method, and the results are consistent with those measured by the traditional ultra-high performance liquid chromatography (UHPLC) method. Hence, this biosensor has potential practical application value in rutin drug quality testing and clinical blood drug concentration monitoring.

Nanomaterials applied in electrochemical devices offer properties that are important in relation to the sensor interface at the molecular level [22]. Multi-walled carbon (LSV) were carried out in 50 mmol L −1 pH 4 phosphate-buffered solution with a scan rate of 0.1 V s −1 . Electrochemical impedance spectroscopy (EIS) was carried out in 0.1 M KCl solution containing 5 mmol L −1 [Fe(CN) 6 ] 3−/4− in the frequency range from 10 2 to 10 6 Hz.
Currently, HPLC is generally used to determine rutin content in drugs and blood. Hence, HPLC was used as a comparative technique with the electrochemical method. Rutin tablets samples were detected using ultra-high-pressure liquid chromatograph (UHPLC) (1290 Infinity II, Agilent, Santa Clara, CA, USA) using the standard addition method. Separation was achieved using an Eclipse Plus C18 column (Agilent, Santa Clara, CA, USA) (2.7 µm, 4.6 × 100 mm); the mobile phase was methanol and water (65: 35, v/v) at a flow rate of 0.2 mL min −1 . The injection volume of the sample was 10 µL, and the UV detector wavelength was set at 360 nm [47].

Preparation of Modified Electrode
Ahead of modification, the surface of a GCE was mechanically polished with slurry of alumina powder with the particle sizes of 1, 0.3, and 0.05 µm, respectively. Then, the GCE was ultrasonicated in 75% ethanol and ultrapure water for 10 min. Afterwards, the GCE was placed in a drying tower.
Preparation of Chi solution: 50 mg of chitosan was added to 10 mL of ultra-pure water containing 1% acetic acid solution (v/v = 1:1), stirred until completely dissolved, and the pH was adjusted to 5-6 with the NaOH solution.
MWCNTs (2 mg mL −1 ) were ultrasonically dispersed in CTAB (2 mg mL −1 ), and 2 µL of the mixture of the MWCNTs-CTAB dispersive turbid liquid was dropped on the prepared GCE and dried at room temperature for 20 min. HFs (2 mg mL −1 ) and Lac (5 mg mL −1 ) were co-cultured at 4 • C for 1 h. Then, 2 µL of HFs-Lac co-culture solution was added on the electrode and dried at 4 • C overnight. At the end, in order to prevent the modified materials from falling off the electrode surface, 1.5 µL of Chi solution was dropped on the GCE, dried, and stored at 4 • C ( Figure 1).

Apparatus and Measurements
All electrochemical measurements were carried out on a CHI660E electrochemical workstation (CH Instrument, Austin, TX, USA) at 25 °C. A GCE 3 mm in diameter, a Ag/AgCl-saturated electrode, and a platinum wire were used as the working electrode, reference electrode, and counter electrode, respectively. CV and linear sweep voltammetry (LSV) were carried out in 50 mmol L −1 pH 4 phosphate-buffered solution with a scan rate of 0.1 V s −1 . Electrochemical impedance spectroscopy (EIS) was carried out in 0.1 M KCl solution containing 5 mmol L −1 [Fe(CN)6] 3−/4− in the frequency range from 10 2 to 10 6 Hz.
Currently, HPLC is generally used to determine rutin content in drugs and blood. Hence, HPLC was used as a comparative technique with the electrochemical method. Rutin tablets samples were detected using ultra-high-pressure liquid chromatograph (UHPLC) (1290 Infinity II, Agilent, Santa Clara, CA, USA) using the standard addition method. Separation was achieved using an Eclipse Plus C18 column (Agilent, Santa Clara, CA, USA) (2.7 µm, 4.6 × 100 mm); the mobile phase was methanol and water (65:35, v/v) at a flow rate of 0.2 mL min −1 . The injection volume of the sample was 10 µL, and the UV detector wavelength was set at 360 nm [47].

Preparation of Modified Electrode
Ahead of modification, the surface of a GCE was mechanically polished with slurry of alumina powder with the particle sizes of 1, 0.3, and 0.05 µm, respectively. Then, the GCE was ultrasonicated in 75% ethanol and ultrapure water for 10 min. Afterwards, the GCE was placed in a drying tower.
Preparation of Chi solution: 50 mg of chitosan was added to 10 mL of ultra-pure water containing 1% acetic acid solution (v/v = 1:1), stirred until completely dissolved, and the pH was adjusted to 5-6 with the NaOH solution.
MWCNTs (2 mg mL −1 ) were ultrasonically dispersed in CTAB (2 mg mL −1 ), and 2 µL of the mixture of the MWCNTs-CTAB dispersive turbid liquid was dropped on the prepared GCE and dried at room temperature for 20 min. HFs (2 mg mL −1 ) and Lac (5 mg mL −1 ) were co-cultured at 4 °C for 1 h. Then, 2 µL of HFs-Lac co-culture solution was added on the electrode and dried at 4 °C overnight. At the end, in order to prevent the modified materials from falling off the electrode surface, 1.5 µL of Chi solution was dropped on the GCE, dried, and stored at 4 °C ( Figure 1).

Preparation of Samples
Chi/HFs-Lac/MWCNTs-CTAB/GCE was used to determine rutin content in rutin tablets samples and serum samples. The preparation of rutin tablets samples was carried out as follows: rutin tablets were purchased from a pharmacy, the tablets were sonicated to dissolve in ethanol, and then the solution was centrifuged at 5000 r min −1 for 10 min after standing at room temperature for 1 h, and the supernatant was taken for electrochemical and UHPLC analyses. The preparation of rutin serum samples was performed as follows: Mouse blood was centrifuged at 3000 r min −1 for 10 min and the serum supernatant was obtained; then, the rutin solution was added to the serum and diluted for electrochemical analysis. Next, 1 mL of methanol was added to the serum sample of rutin (100 µL), then vortex-mixed for about 1 min, and left to stand at 4 • C for 10 min. After centrifugation at 12,000 r min −1 , the supernatant was taken and blow-dried with nitrogen in water at 40 • C. Finally, the samples were redissolved in methanol for UHPLC analysis [49].

Characteristics of Modified Materials
The TEM images of MWCNTs, MWCNTs-CTAB, HFs, and HFs-Lac are shown in Figure 2A-D, respectively. The MWCNTs mixed with CTAB in Figure 2B showed better dispersion than MWCNTs alone in Figure 2A. Figure 2C shows that the particle of HFs was about 15 nm, which might have been due to the dispersion of HFs in water and aggregation during dying. After HFs and Lac were cultured, Lac seemed to be wrapped by HFs, as shown in Figure 2D, which might have been conducive to the catalytic ability of Lac and the electrical conductivity of HFs.
as follows: rutin tablets were purchased from a pharmacy, the tablets were sonicated to dissolve in ethanol, and then the solution was centrifuged at 5000 r min −1 for 10 min after standing at room temperature for 1 h, and the supernatant was taken for electrochemical and UHPLC analyses. The preparation of rutin serum samples was performed as follows: Mouse blood was centrifuged at 3000 r min −1 for 10 min and the serum supernatant was obtained; then, the rutin solution was added to the serum and diluted for electrochemical analysis. Next, 1 mL of methanol was added to the serum sample of rutin (100 µL), then vortex-mixed for about 1 min, and left to stand at 4 °C for 10 min. After centrifugation at 12,000 r min −1 , the supernatant was taken and blow-dried with nitrogen in water at 40 °C. Finally, the samples were redissolved in methanol for UHPLC analysis [49].

Characteristics of Modified Materials
The TEM images of MWCNTs, MWCNTs-CTAB, HFs, and HFs-Lac are shown in Figure 2A-D, respectively. The MWCNTs mixed with CTAB in Figure 2B showed better dispersion than MWCNTs alone in Figure 2A. Figure 2C shows that the particle of HFs was about 15 nm, which might have been due to the dispersion of HFs in water and aggregation during dying. After HFs and Lac were cultured, Lac seemed to be wrapped by HFs, as shown in Figure 2D, which might have been conducive to the catalytic ability of Lac and the electrical conductivity of HFs. The impedances of the electrodes modified with different materials were studied using the EIS method. The EIS diagrams of GCE, Chi/MWCNTs/GCE, Chi/MWCNTs-CTAB/GCE, Chi/HFs/MWCNTs-CTAB/GCE, and Chi/HFs-Lac/MWCNTs-CTAB/GCE are shown in Figure 3, respectively. The EIS curve consists of a semicircular part representing electron transfer resistance (Rct) and a liner part representing the adsorption process. The smaller semicircle diameter balances with the smaller Rct value of the electrode, which means that there is a faster electron transfer rate [50]. The impedance data were fitted with Randle's equivalent circuit in Figure 3 (inset). The Rct of Chi/MWCNTs-CTAB/GCE (97 Ω) was lower than the Rct of Chi/MWCNTs/GCE (117 Ω) and GCE (132 Ω), indicating that the MWCNTs had good conductivity, and the dispersion of CTAB might have helped to improve its effective electron transmission area. The Rct of Chi/HFs/MWCNTs-CTAB/GCE The impedances of the electrodes modified with different materials were studied using the EIS method. The EIS diagrams of GCE, Chi/MWCNTs/GCE, Chi/MWCNTs-CTAB/GCE, Chi/HFs/MWCNTs-CTAB/GCE, and Chi/HFs-Lac/MWCNTs-CTAB/GCE are shown in Figure 3, respectively. The EIS curve consists of a semicircular part representing electron transfer resistance (R ct ) and a liner part representing the adsorption process. The smaller semicircle diameter balances with the smaller R ct value of the electrode, which means that there is a faster electron transfer rate [50]. The impedance data were fitted with Randle's equivalent circuit in Figure 3 (inset). The R ct of Chi/MWCNTs-CTAB/GCE (97 Ω) was lower than the R ct of Chi/MWCNTs/GCE (117 Ω) and GCE (132 Ω), indicating that the MWCNTs had good conductivity, and the dispersion of CTAB might have helped to improve its effective electron transmission area. The R ct of Chi/HFs/MWCNTs-CTAB/GCE decreased to 87 Ω; this is an indication that the HFs might have improved electron transfer on the electrode surface. However, the R ct of Chi/HFs-Lac/MWCNTs-CTAB/GCE was increased to 95 Ω, which might have been due to Lac hindering the electron transport.
Biosensors 2022, 12, x FOR PEER REVIEW 5 of 14 decreased to 87 Ω; this is an indication that the HFs might have improved electron transfer on the electrode surface. However, the Rct of Chi/HFs-Lac/MWCNTs-CTAB/GCE was increased to 95 Ω, which might have been due to Lac hindering the electron transport.  The Chi/HFs-Lac/MWCNTs-CTAB/GCE (curve e) showed the strongest oxidation peak and reduction peak at 0.478 V and 0.445 V, respectively, with a difference potential (ΔE) of 0.033 V (ΔE > 59/n mV, n = 2). The anodic peak current (Ipa) and cathodic peak current (Ipc) were 4.69 µA, 4.34 µA, respectively, with a peak current ratio (Ipa/Ipc) of about 1. Therefore, the Chi/HFs-Lac/MWCNTs-CTAB/GCE electrochemical process was quasi-reversible. Compared with Chi/MWCNTs/GCE, Chi/MWCNTs-CTAB/GCE had a larger electroactive surface area. At the same time, the HFs-Lac structure was more conducive to improving the catalytic performance of Lac, which made up for its low electrical conductivity to a certain extent. Figure 4B shows CVs of Chi/HFs-Lac/MWCNTs-CTAB/GCE in the absence (curve a) and presence (curve b) of 5 µmol L −1 rutin in phosphate-buffered solution. The results further indicate that the redox peaks were caused by rutin.  . The anodic peak current (I pa ) and cathodic peak current (I pc ) were 4.69 µA, 4.34 µA, respectively, with a peak current ratio (I pa /I pc ) of about 1. Therefore, the Chi/HFs-Lac/MWCNTs-CTAB/GCE electrochemical process was quasi-reversible. Compared with Chi/MWCNTs/GCE, Chi/MWCNTs-CTAB/GCE had a larger electroactive surface area. At the same time, the HFs-Lac structure was more conducive to improving the catalytic performance of Lac, which made up for its low electrical conductivity to a certain extent. Figure 4B shows CVs of Chi/HFs-Lac/MWCNTs-CTAB/GCE in the absence (curve a) and presence (curve b) of 5 µmol L −1 rutin in phosphate-buffered solution. The results further indicate that the redox peaks were caused by rutin.
Biosensors 2022, 12, x FOR PEER REVIEW 5 of 14 decreased to 87 Ω; this is an indication that the HFs might have improved electron transfer on the electrode surface. However, the Rct of Chi/HFs-Lac/MWCNTs-CTAB/GCE was increased to 95 Ω, which might have been due to Lac hindering the electron transport.  The Chi/HFs-Lac/MWCNTs-CTAB/GCE (curve e) showed the strongest oxidation peak and reduction peak at 0.478 V and 0.445 V, respectively, with a difference potential (ΔE) of 0.033 V (ΔE > 59/n mV, n = 2). The anodic peak current (Ipa) and cathodic peak current (Ipc) were 4.69 µA, 4.34 µA, respectively, with a peak current ratio (Ipa/Ipc) of about 1. Therefore, the Chi/HFs-Lac/MWCNTs-CTAB/GCE electrochemical process was quasi-reversible. Compared with Chi/MWCNTs/GCE, Chi/MWCNTs-CTAB/GCE had a larger electroactive surface area. At the same time, the HFs-Lac structure was more conducive to improving the catalytic performance of Lac, which made up for its low electrical conductivity to a certain extent. Figure 4B shows CVs of Chi/HFs-Lac/MWCNTs-CTAB/GCE in the absence (curve a) and presence (curve b) of 5 µmol L −1 rutin in phosphate-buffered solution. The results further indicate that the redox peaks were caused by rutin.

Effects of Scan Rates
CVs of Chi/HFs-Lac/MWCNTs-CTAB/GCE in 5 µmol L −1 rutin at different scan rates (ν) are shown in Figure 5A, and the relationship between scan rates (0.01 V s −1 -2 V s −1 ) and peak current is shown in Figure 5B. The regression equations were I pc = 0.036 ν + 1.243 (R 2 = 0.9965) and I pa = −0.036 ν − 1.668 (R 2 = 0.995). In Figure 5C, the absolute slopes value of log I vs. log ν were close to 1, so these indicate that the electrochemical behavior of rutin on the electrode surface is an adsorption-controlled process [16], which is the same as in other reports [51,52]. This phenomenon suggests that the determination of rutin should start from a low concentration, or the adsorbed rutin in the blank phosphate-buffered solution should be removed after detecting a high concentration of rutin. In addition, according to Equation (1) and the slope of the peak current (I p ) versus the scan rate (ν) (Figure 5B), the number (n) of electron transfer in the electrode reaction was 2, which is same result as the previous report [53,54].
Here, F, Q, R, and T represent the Faraday constant, the quantity of the electric charge, and the gas constant and temperature, respectively.

Effects of Scan Rates
CVs of Chi/HFs-Lac/MWCNTs-CTAB/GCE in 5 µmol L −1 rutin at different scan rates (ν) are shown in Figure 5A, and the relationship between scan rates (0.01 V s −1 -2 V s −1 ) and peak current is shown in Figure 5B. The regression equations were Ipc = 0.036 ν + 1.243 (R 2 = 0.9965) and Ipa = −0.036 ν − 1.668 (R 2 = 0.995). In Figure 5C, the absolute slopes value of log I vs. log ν were close to 1, so these indicate that the electrochemical behavior of rutin on the electrode surface is an adsorption-controlled process [16], which is the same as in other reports [51,52]. This phenomenon suggests that the determination of rutin should start from a low concentration, or the adsorbed rutin in the blank phosphate-buffered solution should be removed after detecting a high concentration of rutin. In addition, according to Equation (1) and the slope of the peak current (Ip) versus the scan rate (ν) (Figure 5B), the number (n) of electron transfer in the electrode reaction was 2, which is same result as the previous report [53,54].

= 4
(1) Here, F, Q, R, and T represent the Faraday constant, the quantity of the electric charge, and the gas constant and temperature, respectively.   Figure 6A shows the CVs of Chi/HFs-Lac/MWCNTs-CTAB/GCE in phosphatebuffered solution containing 5 µmol L −1 rutin at different pH values. Figure 6B shows that the cathodic peak current increased with a pH value from 2 to 4, but the current decreased with a pH value higher than 4. It was also reported that the optimum pH range of Lac catalytic activity was 3-4 [55]; therefore, the pH 4 phosphate-buffered solution was selected as the test solution. Furthermore, the formal potential was linearly related to pH with the equation of E • = −0.0607 pH + 0.645 (R 2 = 0.998). The slope value was −60.7 mV pH −1 , which was close to the Nernst's value of 59.2 mV pH −1 , indicating that the number of electrons and protons was the same in this reaction ( Figure 6C). The pKa value of rutin is 6.17 ± 0.4, while the pH values studied in this work were from 2 to 6, so rutin was still protonated. Figure 6A shows the CVs of Chi/HFs-Lac/MWCNTs-CTAB/GCE in phosphate-buffered solution containing 5 µmol L −1 rutin at different pH values. Figure 6B shows that the cathodic peak current increased with a pH value from 2 to 4, but the current decreased with a pH value higher than 4. It was also reported that the optimum pH range of Lac catalytic activity was 3-4 [55]; therefore, the pH 4 phosphate-buffered solution was selected as the test solution. Furthermore, the formal potential was linearly related to pH with the equation of E°′ = −0.0607 pH + 0.645 (R 2 = 0.998). The slope value was −60.7 mV pH −1 , which was close to the Nernst's value of 59.2 mV pH −1 , indicating that the number of electrons and protons was the same in this reaction ( Figure 6C). The pKa value of rutin is 6.17 ± 0.4, while the pH values studied in this work were from 2 to 6, so rutin was still protonated.

Optimization of Experimental Conditions
In order to optimize the performance of Chi/HFs-Lac/MWCNTs-CTAB/GCE, the determination time and ratio of modified materials were investigated. From 1 to 10 min, the cathodic peak current increased with time, and after 10 min, the current was basically stable ( Figure 7A). Therefore, rutin was tested at 10 min. The optimal ratios of MWCNTs to CTAB and HFs to Lac were studied. As seen in Figure 7B,C, the cathodic peak signal was strongest when the ratio of MWCNTs to CTAB was 15:1 (v/v), and that of HFs to Lac was 1:1 (v/v). Thus, these optimal ratios were used for electrode modification.

Optimization of Experimental Conditions
In order to optimize the performance of Chi/HFs-Lac/MWCNTs-CTAB/GCE, the determination time and ratio of modified materials were investigated. From 1 to 10 min, the cathodic peak current increased with time, and after 10 min, the current was basically stable ( Figure 7A). Therefore, rutin was tested at 10 min. The optimal ratios of MWCNTs to CTAB and HFs to Lac were studied. As seen in Figure 7B,C, the cathodic peak signal was strongest when the ratio of MWCNTs to CTAB was 15:1 (v/v), and that of HFs to Lac was 1:1 (v/v). Thus, these optimal ratios were used for electrode modification.

Calibration Plot of Rutin
The LSV method was used to determine rutin under optimal conditions ( Figure 8A). The plot of cathodic peak current versus rutin in the concentration range of 0.05 µmol L −1 -20 µmol L −1 is shown in Figure 8B. The linear range of Chi/HFs-Lac/MWCNTs-CTAB/GCE response to rutin was divided into two parts: 0.1 µmol L −1 -2 µmol L −1 and 2 µmol L −1 to 11 µmol L −1 , with regression equations of I (µA) = 6.888CRutin (µmol L −1 ) − 1.061 and I (µA) = 2.827CRutin (µmol L −1 ) + 7.643, respectively ( Figure 8C), and the limit of detection(LOD) was 30 nmol L −1 and 95.5 nmol L −1 (Equation (2), which is better than that previously reported [17,18]). The sensitivities of the biosensor were 6.88 µA µM −1 and 2.82 µA µM −1 , respectively, which were better than those without Lac modification, which demonstrated the catalytic activity of the Lac-modified electrode. Table 1 shows a comparison between this work and the performance of earlier reported rutin sensors. It can be seen that the improved modified biosensor in this work had good performance.
Here, S0 and S represent the standard deviation measured under blank solution and the slope of the linear range curve, respectively.

Calibration Plot of Rutin
The LSV method was used to determine rutin under optimal conditions ( Figure 8A). The plot of cathodic peak current versus rutin in the concentration range of 0.05 µmol L −1 -20 µmol L −1 is shown in Figure 8B. The linear range of Chi/HFs-Lac/MWCNTs-CTAB/GCE response to rutin was divided into two parts: 0.1 µmol L −1 -2 µmol L −1 and 2 µmol L −1 to 11 µmol L −1 , with regression equations of I (µA) = 6.888C Rutin (µmol L −1 ) − 1.061 and I (µA) = 2.827C Rutin (µmol L −1 ) + 7.643, respectively ( Figure 8C), and the limit of detection(LOD) was 30 nmol L −1 and 95.5 nmol L −1 (Equation (2), which is better than that previously reported [17,18]). The sensitivities of the biosensor were 6.88 µA µM −1 and 2.82 µA µM −1 , respectively, which were better than those without Lac modification, which demonstrated the catalytic activity of the Lac-modified electrode. Table 1 shows a comparison between this work and the performance of earlier reported rutin sensors. It can be seen that the improved modified biosensor in this work had good performance.
Here, S 0 and S represent the standard deviation measured under blank solution and the slope of the linear range curve, respectively.

Anti-Interference Ability and Stability of Biosensor
The storage stability of the biosensor was evaluated using the LSV method in phosphatebuffered solution containing 5 µmol L −1 rutin ( Figure 9A). The cathodic peak current showed little change when the biosensor was stored at 4 • C for 10 days. After 20 and 30 days of storage, the peak signal decreased by 15% and 19%, respectively. The antiinterference ability of the biosensor was studied by observing whether the electroactive species changed the cathodic peak current using the LSV method. In Figure 9B, interfering substances in 500-fold concentrations of rutin (250 µmol L −1 ), including ethanol, glucose (Glu), ascorbic acid (Vit C), glycine (Gly), and uric acid (UA), elicited only weak current responses. Both additions of 0.5 µmol L −1 rutin caused a strong current signal and a cathodic peak appeared, so this biosensor had a great anti-interference performance. Due to the different concentrations of interfering substances and rutin samples, the cathodic peak current could not accurately describe the anti-interference capability of this biosensor. Here, we introduced an interference factor (IF) (Equation (3)) to more intuitionally investigate the anti-interference ability of the biosensor. The lower the IF value, the stronger the anti-interference ability of the biosensor. The results are shown in Table 2.
Here, S interference and S rutin represent the sensitivity of interfering substances and sensitivity of rutin, respectively. The storage stability of the biosensor was evaluated using the LSV method in phosphate-buffered solution containing 5 µmol L −1 rutin ( Figure 9A). The cathodic peak current showed little change when the biosensor was stored at 4 °C for 10 days. After 20 and 30 days of storage, the peak signal decreased by 15% and 19%, respectively. The anti-interference ability of the biosensor was studied by observing whether the electroactive species changed the cathodic peak current using the LSV method. In Figure 9B, interfering substances in 500-fold concentrations of rutin (250 µmol L −1 ), including ethanol, glucose (Glu), ascorbic acid (Vit C), glycine (Gly), and uric acid (UA), elicited only weak current responses. Both additions of 0.5 µmol L −1 rutin caused a strong current signal and a cathodic peak appeared, so this biosensor had a great anti-interference performance. Due to the different concentrations of interfering substances and rutin samples, the cathodic peak current could not accurately describe the anti-interference capability of this biosensor. Here, we introduced an interference factor (IF) (Equation (3)) to more intuitionally investigate the anti-interference ability of the biosensor. The lower the IF value, the stronger the anti-interference ability of the biosensor. The results are shown in Table 2.
Here, Sinterference and Srutin represent the sensitivity of interfering substances and sensitivity of rutin, respectively.  To further verify the specificity of this biosensor, spectra of laccase-catalyzed rutin were studied. Continuous spectral scans from 200 nm to 700 nm of Lac and the rutin reaction system using UV-vis spectrophotometer are shown in Figure 10. The absorption bands at 354 nm and 254 nm are cinnamoyl chromophore and benzoyl chromophore of rutin, respectively [58]. It can be seen that the absorption peak of the cinnamoyl structure at 354 nm decreased with the reaction time, indicating that the catechol of this structure was oxidized. Lac catalyzed the oxidation of rutin to the corresponding o-quinone with the concomitant reduction of molecular oxygen to water; the corresponding o-quinone served as the two electrons' acceptor, and was reduced back to rutin on the surface of the biosensor (Figure 11).  To further verify the specificity of this biosensor, spectra of laccase-catalyzed rutin were studied. Continuous spectral scans from 200 nm to 700 nm of Lac and the rutin reaction system using UV-vis spectrophotometer are shown in Figure 10. The absorption bands at 354 nm and 254 nm are cinnamoyl chromophore and benzoyl chromophore of rutin, respectively [58]. It can be seen that the absorption peak of the cinnamoyl structure at 354 nm decreased with the reaction time, indicating that the catechol of this structure was oxidized. Lac catalyzed the oxidation of rutin to the corresponding o-quinone with the concomitant reduction of molecular oxygen to water; the corresponding o-quinone served as the two electrons' acceptor, and was reduced back to rutin on the surface of the biosensor ( Figure 11).

Determination of Rutin in Tablets and Serum
Rutin tablets and serum samples were determined using the proposed biosensor. The results of electrochemistry and UHPLC methods used to determine rutin tablets samples with different concentrations are shown in Table 3. In addition, the electrochemistry method was used to determine different concentrations of rutin in serum samples. The results show that the biosensor had a good recovery (Table 4). Statistical analysis showed that the f-test value was lower than the critical value of F (Fcritical = 5.05), while the t-test value was lower than the critical value of t (tcritical = 2.57), indicating that the biosensor is comparable to the traditional UHPLC method. Therefore, the improved modified biosensor has potential practical application value in rutin drug quality determination and clinical blood drug concentration monitoring.

Determination of Rutin in Tablets and Serum
Rutin tablets and serum samples were determined using the proposed biosensor. The results of electrochemistry and UHPLC methods used to determine rutin tablets samples with different concentrations are shown in Table 3. In addition, the electrochemistry method was used to determine different concentrations of rutin in serum samples. The results show that the biosensor had a good recovery (Table 4). Statistical analysis showed that the f-test value was lower than the critical value of F (Fcritical = 5.05), while the t-test value was lower than the critical value of t (tcritical = 2.57), indicating that the biosensor is comparable to the traditional UHPLC method. Therefore, the improved modified biosensor has potential practical application value in rutin drug quality determination and clinical blood drug concentration monitoring.

Determination of Rutin in Tablets and Serum
Rutin tablets and serum samples were determined using the proposed biosensor. The results of electrochemistry and UHPLC methods used to determine rutin tablets samples with different concentrations are shown in Table 3. In addition, the electrochemistry method was used to determine different concentrations of rutin in serum samples. The results show that the biosensor had a good recovery (Table 4). Statistical analysis showed that the f-test value was lower than the critical value of F (F critical = 5.05), while the t-test value was lower than the critical value of t (t critical = 2.57), indicating that the biosensor is comparable to the traditional UHPLC method. Therefore, the improved modified biosensor has potential practical application value in rutin drug quality determination and clinical blood drug concentration monitoring.

Conclusions
The Chi/HFs-Lac/MWCNTs-CTAB/GCE biosensor was prepared for the determination of rutin with a wide determination linear range, low detection limit, high accuracy, good anti-interference ability, and strong stability. The biosensor had the same accuracy as traditional methods such as UHPLC in determining rutin and can be applied to the field of drug quality determination. Moreover, the biosensor had a good recovery for detecting rutin serum samples using the standard addition method; so the biosensor has potential practical application value in blood drug concentration monitoring.