Nanocomposite of Ellagic Acid with Multi-Walled Carbon Nanotubes for the Simultaneous Voltammetric Detection of Six Biomolecules

In this proof-of-concept study, a highly sensitive electrochemical sensor using a graphite paste electrode modified with ellagic acid and multi-walled carbon nanotubes (MGPE/MWCNTsEA) was developed for the simultaneous determination of six biomolecules: ascorbic acid (AA), dopamine (DA), uric acid (UA), tryptophan (Trp), xanthine (XA), and caffeine (CA). Differential pulse voltammetry (DPV) was performed at a potential range from 0.1–1.2 V vs. Ag/AgCl in phosphate electrolyte (pH 2.0). The modified GPE enabled the simultaneous determination of biomolecules under investigation in human urine and blood serum samples with detection limits ranging from 11–91 nM with recoveries of 94.0–106.0%. The electrochemical performance of the modified GPE for the analytes was stable and reproducible and checked with standard high performance liquid chromatography technique. The data suggested that the modified GPE provided a promising platform for routine quantitative determination of the biomolecules under investigation in quality control studies of real samples collected from food and pharmaceutical products.


Introduction
The quantitative determination of small molecules such as ascorbic acid (AA), dopamine (DA), uric acid (UA), tryptophan (Trp), xanthine (XA) and caffeine (CA) is generally performed using separation methods such high performance liquid chromatography (HPLC), ultra-performance liquid chromatography, and capillary electrophoresis [1][2][3][4][5][6][7][8][9][10]. These sophisticated methods have high costs with time-consuming procedures and require trained technicians for operation and data interpretation [11,12]. Indeed, all sample solutions should be clean without any suspension materials and color because of the limitations of expensive columns in HPLC and UPLC with optical detectors. In addition, in order to detect these biomolecules individually and simultaneously, several other methods such as spectrofluorimetry [13,14], mass spectrometry [15,16], and electrochemistry [17,18] have been reported. Among them, spectroscopic methods have been sensitive with a low detection limit in individual determinations but not for simultaneous determination of two or three analytes. In comparison with separation and spectroscopic techniques, electrochemical methods offer advantages such as low cost, rapid procedures, and simple systems that can be operated for individual and simultaneous determinations without any limitations on the color of sample solutions [19,20]; however, for better results, the analytes should be redox-active on the surface of the working electrodes (WEs). For this purpose, WEs were modified using different organic/inorganic compounds, nanomaterials, and nanocomposite structures to develop sensors with higher selectivity, sensitivity, and precision for simultaneous determination of more than three biomolecules without extensive use of sample preparation and separation techniques [19,20]. Without a modified WE, the oxidation peak current Finally, a nanocomposite-modified electrode was used for the determination of analytes in real samples. The electrochemical results were compared with ones obtained from high performance liquid chromatography (HPLC) as the gold standard method. MGPE/MWCNTs-EA as a surface nanocomposite for the simultaneous electrochemical determination of biomolecules under investigation. Finally, a nanocomposite-modified electrode was used for the determination of analytes in real samples. The electrochemical results were compared with ones obtained from high performance liquid chromatography (HPLC) as the gold standard method.

Reagents and solutions.
Ascorbic acid (AA), dopamine (DA), uric acid (UA), tryptophan (Trp), xanthine (XA) and caffeine (CA) were obtained from Merck Company (Kenilworth, NJ, USA). Ellagic acid (EA) was obtained from Sigma-Aldrich Company (Oakville, ON, Canada). The multiwall carbon nanotubes (MWCNTs), with nanotube diameters, OD = 20-30 nm, wall thickness = 1-2 nm, length = 0.5-2 μm and purity 95% from were purchased from Aldrich. Phosphate electrolyte and buffer and solutions (PBS) were prepared from H3PO4 (0.1 M), and pH range was adjusted to 2.0-8.0 with 0.1 M H3PO4 and NaOH. All solutions were prepared with ultra-pure Milli-Q water at a resistivity of 18.2 MΩ cm. All measurements were performed in compliance with relevant laws and institutional guidelines.

Instrumentation
Electrochemical measurements were performed with an Autolab electrochemical analyzer (Metrohm-Autolab B.V., Utrecht, The Netherlands). All electrochemical experiments were carried out on a conventional three-electrode cell at room temperature. A platinum electrode and a silver/silver chloride electrode (Ag/AgCl) were used as the counter and reference electrodes, and MGPE/EA as the working electrode. A Metrohm pH meter, model 744 (Metrohm-Autolab B.V., Utrecht, The Netherlands), was used for pH measurements.

Reagents and Solutions
Ascorbic acid (AA), dopamine (DA), uric acid (UA), tryptophan (Trp), xanthine (XA) and caffeine (CA) were obtained from Merck Company (Kenilworth, NJ, USA). Ellagic acid (EA) was obtained from Sigma-Aldrich Company (Oakville, ON, Canada). The multi-wall carbon nanotubes (MWCNTs), with nanotube diameters, OD = 20-30 nm, wall thickness = 1-2 nm, length = 0.5-2 µm and purity 95% from were purchased from Aldrich. Phosphate electrolyte and buffer and solutions (PBS) were prepared from H 3 PO 4 (0.1 M), and pH range was adjusted to 2.0-8.0 with 0.1 M H 3 PO 4 and NaOH. All solutions were prepared with ultra-pure Milli-Q water at a resistivity of 18.2 MΩ cm. All measurements were performed in compliance with relevant laws and institutional guidelines.

Instrumentation
Electrochemical measurements were performed with an Autolab electrochemical analyzer (Metrohm-Autolab B.V., Utrecht, The Netherlands). All electrochemical experiments were carried out on a conventional three-electrode cell at room temperature. A platinum electrode and a silver/silver chloride electrode (Ag/AgCl) were used as the counter and reference electrodes, and MGPE/EA as the working electrode. A Metrohm pH meter, model 744 (Metrohm-Autolab B.V., Utrecht, The Netherlands), was used for pH measurements.

Preparation of MGPE/MWCNTs-EA
The modified graphite paste electrode (MGPE) was prepared by carefully hand mixing 3 mg of EA, 3 mg of MWCNTs, and 3 drops of paraffin oil using a mortar and pestle for 5 min and then 194 mg graphite powder was added to the mixture step by step and C 2021, 7, 43 4 of 20 mixed for 15 min to obtain a uniform paste. This paste was packed into the end of the glass cylinder tube with 2 mm diameter and a copper wire was used as electrical contact. Finally, the MGPEs were polished on a weighing paper. Before each measurement, for the elimination of any cross-contamination effects, the electrode surface was easily and rapidly renewed, and a few potential cycles in a cyclic voltammetric regime within the range of −0.5-1.5 V (vs. Ag/AgCl) at a scan rate of 50 mV/s were applied in the PBS (pH 2.0) until a stable voltammogram was obtained. The RSD% of this procedure was smaller than 5%, which was acceptable. The electrodes were stored in ultra-pure water at RT when not in use. The MGPE/EA and MGPE/MWCNTs were prepared with the same route for comparison studies.

Characterization
The TEM image of MWCNTs-EA in paraffin oil and SEM of MGPE/MWCNTs-EA are shown in Figure 1. Based on Figure 1A, the EA in paraffin oil was attached to the MWCNTs. Figure 1B shows the SEM of the prepared paste for graphite and MWCNTs-EA, which indicated higher homogeneity of the paste. Figure S1A shows CVs for the MGPE/MWCNTs-EA, MGPE/MWCNTs, MGPE/EA, and GPE. At the MGPE/MWCNTs-EA, the peak current was greater than the one recorded with MGPE/EA, indicating more efficient charge transfer property of the MWCNTs in the modified graphite paste electrode surface (syngeneic effect). The oxidation peaks of EA can be assigned to oxidation of catechol moieties, 3 ,4 -dihydroxyl electron-donating groups, at rings 2 and 4 (see Scheme 1) that undergo a two-electron procession. Electrochemical measurements of the MGPE/MWCNTs-EA, MGPE/MWCNTs, MGPE/EA, and GPE were analyzed for the anodic peak current (I pa ) of the respective cyclic voltammograms obtained in the presence of 1.0 mM of [Fe(CN) 6 ] 3− in 0.1 M KCl of supporting electrolyte (Figure S1B-E). As shown in Figure S1, all assays were performed using cyclic voltammetry between potentials of −1.0 to 1.1 V for MGPE/MWCNTs-EA, MGPE/MWCNTs, MGPE/EA, and GPE at various scan rates. For a reversible process, the Randles-Sevcik equation can be used:

Electrochemical Behavior of the Modified Electrodes
I pa = 2.69 × 10 5 n 3/2 AC 0 D R 1/2 v 1/2 (1) where I pa refers to the anodic peak current, D R is diffusion coefficient, n is the number of electrons exchanged, A is the surface area of the electrode, C 0 is the concentration of [Fe(CN) 6 ] 3− , v and is the scan rate. For [Fe(CN) 6 ] 3− , n = 1, D = 7.6 × 10 −6 cm s −1 , the effective area could be calculated from the slope of the I pa vs. v 1/2 relation (see Figure S1F) [21].

Characterization
The TEM image of MWCNTs-EA in paraffin oil and SEM of MGPE/MWCNTs-EA shown in Figure 1. Based on Figure 1A, the EA in paraffin oil was attached to MWCNTs. Figure 1B shows the SEM of the prepared paste for graphite and MWC EA, which indicated higher homogeneity of the paste.   Figure S1(B-E). As shown in Figure S1, all assays were performed using c voltammetry between potentials of -1.0 to 1.1 V for MGPE/MWCNTs MGPE/MWCNTs, MGPE/EA, and GPE at various scan rates. For a reversible proces  Figure 2 shows the Nyquist plots for these electrodes and the Randles equivalent circuit that was used for fitting the EIS data using NOVA software (Metrohm, Utrecht, The Netherlands). that was performed for the above four GPE, MGPE/EA, MGPE/MWCNTs, and MGPE/MWCNTs-EA electrodes in the presence of ferro/ferricynaide redox couple. Figure 2 shows the Nyquist plots for these electrodes and the Randles equivalent circuit that was used for fitting the EIS data using NOVA software (Metrohm, Utrecht, The Netherlands). The Randles equivalent circuit consisted of a solution resistance (Rs) (21.3 Ω in this work), a charge transfer resistance (Rct), a double-layer capacitance (Cdl), and Warburg impedance (ZW) (inset in Figure S3). Nyquist plots showed a semicircle with Rct of 1832 Ω for GPE. After modification with EA, MWCNTs, and MWCNTs-EA, the Rct decreased to 1187, 689, and 499 Ω, respectively, thus, a significantly lower Rct was obtained for MGPE/MWCNTs-EA in comparison with GPE, MGPE/EA, and MGPE/MWCNTs.  Figure  3A(b), the MGPE/MWCNTs show four oxidation peaks for AA, DA, UA, and CA and a broad oxidation peak for Trp and XN. The MGPE/EA shows three broad oxidation peaks: the first broad peak is for AA-DA-UA, the second for Trp-XA, and the third for CA (see Figure 3A(c)). For the bare GPE in Figure 3A(d), the first broad peak is for the mixture AA-DA-UA, the second board peak is for Trp-XA, and the third peak is for CA. The oxidation peaks of AA, DA, UA, Trp, XN, and CA on the surface of GPE/MWCNT were significantly low and no distinguishable current peaks were detected. In addition, bare GPE and MGPE/EA was unable to show six well-defined oxidation peaks for AA, DA, UA, Trp, XN, and CA. The Randles equivalent circuit consisted of a solution resistance (R s ) (21.3 Ω in this work), a charge transfer resistance (R ct ), a double-layer capacitance (C dl ), and Warburg impedance (Z W ) (inset in Figure S3). Nyquist plots showed a semicircle with R ct of 1832 Ω for GPE. After modification with EA, MWCNTs, and MWCNTs-EA, the R ct decreased to 1187, 689, and 499 Ω, respectively, thus, a significantly lower R ct was obtained for MGPE/MWCNTs-EA in comparison with GPE, MGPE/EA, and MGPE/MWCNTs.  Figure 3A(b), the MGPE/MWCNTs show four oxidation peaks for AA, DA, UA, and CA and a broad oxidation peak for Trp and XN. The MGPE/EA shows three broad oxidation peaks: the first broad peak is for AA-DA-UA, the second for Trp-XA, and the third for CA (see Figure 3A(c)). For the bare GPE in Figure 3A(d), the first broad peak is for the mixture AA-DA-UA, the second board peak is for Trp-XA, and the third peak is for CA. The oxidation peaks of AA, DA, UA, Trp, XN, and CA on the surface of GPE/MWCNT were significantly low and no distinguishable current peaks were detected. In addition, bare GPE and MGPE/EA was unable to show six well-defined oxidation peaks for AA, DA, UA, Trp, XN, and CA. Thus, these results indicated that MGPE/MWCNTs-EA significantly improved the separation of oxidation peak potentials and enhanced peak currents with a considerable enhancement in the anodic peak current of AA, DA, UA, Trp, XN, and CA in comparison with GPE and GPE-MWCNTs. These results clearly show the synergic effect of MWCNTs and EA for electrooxidation of the biomolecules under investigation. Thus, these results indicated that MGPE/MWCNTs-EA significantly improved the separation of oxidation peak potentials and enhanced peak currents with a considerable enhancement in the anodic peak current of AA, DA, UA, Trp, XN, and CA in comparison with GPE and GPE-MWCNTs. These results clearly show the synergic effect of MWCNTs and EA for electrooxidation of the biomolecules under investigation. Moreover, the selective senary detection of the biomolecules under investigation on the MGPE/MWCNTs-EA electrode was studied by comparing the DPV responses of the simultaneous and individual measurements of these six compounds. The results are displayed in Figure 3B, indicating that the oxidation processes of these six compounds were independent and therefore, simultaneous determination of AA, DA, UA, Trp, XN, and CA on the modified graphite paste electrode was possible without any significant interference and crosstalk between the biomolecules. The resulting selectivity and sensitivity of the six anodic peak currents was sufficient to achieve accurate simultaneous determination of these biomolecules in mixed samples.

3.4.
Influence of pH on the Simultaneous Electrooxidation of AA, DA, UA, Trp, XN, and CA Based on literature, the acidity of the electrolyte had a significant influence on the electro-oxidation of biomolecules, such as the biomolecules under investigation. This is because protons take part in the electrode reaction of biomolecules. Thus, the effect of pH on the nanocomposite-modified electrode, modified graphite paste electrode, and signal must be carefully investigated at pH levels ranging from 2.0-8.0. The results are shown in Figure 4A. Moreover, the selective senary detection of the biomolecules under investigation on the MGPE/MWCNTs-EA electrode was studied by comparing the DPV responses of the simultaneous and individual measurements of these six compounds. The results are displayed in Figure 3B, indicating that the oxidation processes of these six compounds were independent and therefore, simultaneous determination of AA, DA, UA, Trp, XN, and CA on the modified graphite paste electrode was possible without any significant interference and crosstalk between the biomolecules. The resulting selectivity and sensitivity of the six anodic peak currents was sufficient to achieve accurate simultaneous determination of these biomolecules in mixed samples.

Influence of pH on the Simultaneous Electrooxidation of AA, DA, UA, Trp, XN, and CA
Based on literature, the acidity of the electrolyte had a significant influence on the electro-oxidation of biomolecules, such as the biomolecules under investigation. This is because protons take part in the electrode reaction of biomolecules. Thus, the effect of pH on the nanocomposite-modified electrode, modified graphite paste electrode, and signal must be carefully investigated at pH levels ranging from 2.0-8.0. The results are shown in Figure 4A.   The effect of pH on the peak potentials of the biomolecules under investigation is shown in Figure 4B. Based on this figure, the peak potential for all of the biomolecules decreased linearly with increase in pH in the electrolyte solution from 2.0-8.0. This result indicated the proton transfer in the electrode processes and agreed with previously reported results for the oxidation of several biomolecules in modified electrodes [17,[38][39][40]. Generally, the anodic peak potentials were moved to lower potentials for analytes such as AA, DA, UA, Trp, XN, and CA, which agree with the theoretical slope − 2.303mRT nF of 0.059 m n V/pH (where m and n are the numbers of protons and electrons, respectively; the other parameters have the usual meaning). According to these results, the slopes for the oxidation process of AA, DA, UA, Trp, and XN are as 0.061, 0.0561, 0.0539, 0.0539, and 0.06 V/pH, respectively. These results show that a redox reaction for AA, DA, UA, Trp, and XN contain an equal number of protons and electrons (m = n). Therefore, the following electrooxidation reaction can be suggested for them, as shown in Equations (2)- (6).
The effect of pH on the peak potentials of the biomolecules under investigation is shown in Figure 4B. Based on this figure, the peak potential for all of the biomolecules decreased linearly with increase in pH in the electrolyte solution from 2.0-8.0. This result indicated the proton transfer in the electrode processes and agreed with previously reported results for the oxidation of several biomolecules in modified electrodes [17,[38][39][40]. Generally, the anodic peak potentials were moved to lower potentials for analytes such as AA, DA, UA, Trp, XN, and CA, which agree with the theoretical slope (− . ) of 0.059 ( ) V/pH (where m and n are the numbers of protons and electrons, respectively; the other parameters have the usual meaning). According to these results, the slopes for the oxidation process of AA, DA, UA, Trp, and XN are as 0.061, 0.0561, 0.0539, 0.0539, and 0.06 V/pH, respectively. These results show that a redox reaction for AA, DA, UA, Trp, and XN contain an equal number of protons and electrons (m = n). Therefore, the following electrooxidation reaction can be suggested for them, as shown in Equations (2)-(6).
Unlike the mentioned analytes, the CA is an exception, because the anodic peak potential of CA did not significantly shift with increasing pH and the slope amount, 0.0296 V/pH, is lower and approximately half of the theoretical Nernstian value than AA, DA, UA, Trp, and XN. This amount can be assigned a redox process involving m = 1 and n = 2. The same result was reported in literature [50][51][52] with a similar amount, but as mentioned before, the CA is an exception among biomolecules due to its redox reaction. In addition, Wang et al. [53] recently reported that there was no change in the Ep of CA with increasing pH conditions, which revealed that no proton was involved in the oxidation The effect of pH on the peak potentials of the biomolecules under investigation is shown in Figure 4B. Based on this figure, the peak potential for all of the biomolecules decreased linearly with increase in pH in the electrolyte solution from 2.0-8.0. This result indicated the proton transfer in the electrode processes and agreed with previously reported results for the oxidation of several biomolecules in modified electrodes [17,[38][39][40]. Generally, the anodic peak potentials were moved to lower potentials for analytes such as AA, DA, UA, Trp, XN, and CA, which agree with the theoretical slope (− . ) of 0.059 ( ) V/pH (where m and n are the numbers of protons and electrons, respectively; the other parameters have the usual meaning). According to these results, the slopes for the oxidation process of AA, DA, UA, Trp, and XN are as 0.061, 0.0561, 0.0539, 0.0539, and 0.06 V/pH, respectively. These results show that a redox reaction for AA, DA, UA, Trp, and XN contain an equal number of protons and electrons (m = n). Therefore, the following electrooxidation reaction can be suggested for them, as shown in Equations (2)-(6).
Unlike the mentioned analytes, the CA is an exception, because the anodic peak potential of CA did not significantly shift with increasing pH and the slope amount, 0.0296 V/pH, is lower and approximately half of the theoretical Nernstian value than AA, DA, UA, Trp, and XN. This amount can be assigned a redox process involving m = 1 and n = 2. The same result was reported in literature [50][51][52] with a similar amount, but as mentioned before, the CA is an exception among biomolecules due to its redox reaction. In addition, Wang et al. [53] recently reported that there was no change in the Ep of CA with increasing pH conditions, which revealed that no proton was involved in the oxidation The effect of pH on the peak potentials of the biomolecules under investigation is shown in Figure 4B. Based on this figure, the peak potential for all of the biomolecules decreased linearly with increase in pH in the electrolyte solution from 2.0-8.0. This result indicated the proton transfer in the electrode processes and agreed with previously reported results for the oxidation of several biomolecules in modified electrodes [17,[38][39][40]. Generally, the anodic peak potentials were moved to lower potentials for analytes such as AA, DA, UA, Trp, XN, and CA, which agree with the theoretical slope (− . ) of 0.059 ( ) V/pH (where m and n are the numbers of protons and electrons, respectively; the other parameters have the usual meaning). According to these results, the slopes for the oxidation process of AA, DA, UA, Trp, and XN are as 0.061, 0.0561, 0.0539, 0.0539, and 0.06 V/pH, respectively. These results show that a redox reaction for AA, DA, UA, Trp, and XN contain an equal number of protons and electrons (m = n). Therefore, the following electrooxidation reaction can be suggested for them, as shown in Equations (2)-(6).
Unlike the mentioned analytes, the CA is an exception, because the anodic peak potential of CA did not significantly shift with increasing pH and the slope amount, 0.0296 V/pH, is lower and approximately half of the theoretical Nernstian value than AA, DA, UA, Trp, and XN. This amount can be assigned a redox process involving m = 1 and n = 2. The same result was reported in literature [50][51][52] with a similar amount, but as mentioned before, the CA is an exception among biomolecules due to its redox reaction. In addition, Wang et al. [53] recently reported that there was no change in the Ep of CA with increasing pH conditions, which revealed that no proton was involved in the oxidation The effect of pH on the peak potentials of the biomolecules under investigation is shown in Figure 4B. Based on this figure, the peak potential for all of the biomolecules decreased linearly with increase in pH in the electrolyte solution from 2.0-8.0. This result indicated the proton transfer in the electrode processes and agreed with previously reported results for the oxidation of several biomolecules in modified electrodes [17,[38][39][40]. Generally, the anodic peak potentials were moved to lower potentials for analytes such as AA, DA, UA, Trp, XN, and CA, which agree with the theoretical slope (− . ) of 0.059 ( ) V/pH (where m and n are the numbers of protons and electrons, respectively; the other parameters have the usual meaning). According to these results, the slopes for the oxidation process of AA, DA, UA, Trp, and XN are as 0.061, 0.0561, 0.0539, 0.0539, and 0.06 V/pH, respectively. These results show that a redox reaction for AA, DA, UA, Trp, and XN contain an equal number of protons and electrons (m = n). Therefore, the following electrooxidation reaction can be suggested for them, as shown in Equations (2)-(6).
Unlike the mentioned analytes, the CA is an exception, because the anodic peak potential of CA did not significantly shift with increasing pH and the slope amount, 0.0296 V/pH, is lower and approximately half of the theoretical Nernstian value than AA, DA, UA, Trp, and XN. This amount can be assigned a redox process involving m = 1 and n = 2. The same result was reported in literature [50][51][52] with a similar amount, but as mentioned before, the CA is an exception among biomolecules due to its redox reaction. In addition, Wang et al. [53] recently reported that there was no change in the Ep of CA with increasing pH conditions, which revealed that no proton was involved in the oxidation The effect of pH on the peak potentials of the biomolecules under investigation is shown in Figure 4B. Based on this figure, the peak potential for all of the biomolecules decreased linearly with increase in pH in the electrolyte solution from 2.0-8.0. This result indicated the proton transfer in the electrode processes and agreed with previously reported results for the oxidation of several biomolecules in modified electrodes [17,[38][39][40]. Generally, the anodic peak potentials were moved to lower potentials for analytes such as AA, DA, UA, Trp, XN, and CA, which agree with the theoretical slope (− . ) of 0.059 ( ) V/pH (where m and n are the numbers of protons and electrons, respectively; the other parameters have the usual meaning). According to these results, the slopes for the oxidation process of AA, DA, UA, Trp, and XN are as 0.061, 0.0561, 0.0539, 0.0539, and 0.06 V/pH, respectively. These results show that a redox reaction for AA, DA, UA, Trp, and XN contain an equal number of protons and electrons (m = n). Therefore, the following electrooxidation reaction can be suggested for them, as shown in Equations (2)-(6).
Unlike the mentioned analytes, the CA is an exception, because the anodic peak potential of CA did not significantly shift with increasing pH and the slope amount, 0.0296 V/pH, is lower and approximately half of the theoretical Nernstian value than AA, DA, UA, Trp, and XN. This amount can be assigned a redox process involving m = 1 and n = 2. The same result was reported in literature [50][51][52] with a similar amount, but as mentioned before, the CA is an exception among biomolecules due to its redox reaction. In addition, Wang et al. [53] recently reported that there was no change in the Ep of CA with increasing pH conditions, which revealed that no proton was involved in the oxidation (6) Unlike the mentioned analytes, the CA is an exception, because the anodic peak potential of CA did not significantly shift with increasing pH and the slope amount, 0.0296 V/pH, is lower and approximately half of the theoretical Nernstian value than AA, DA, UA, Trp, and XN. This amount can be assigned a redox process involving m = 1 and n = 2. The same result was reported in literature [50][51][52] with a similar amount, but as mentioned before, the CA is an exception among biomolecules due to its redox reaction. In addition, Wang et al. [53] recently reported that there was no change in the E p of CA with increasing C 2021, 7, 43 10 of 20 pH conditions, which revealed that no proton was involved in the oxidation reaction of CA. Based on the evidence above and previous studies [50][51][52][53], we hypothesized that an equal number of protons and electrons were involved in the oxidation of CA. According to equation 7, the oxidation of CA corresponded to an overall 4H + and 4e − oxidation, following a two-step mechanism. The first mechanism was the slow 2H + and 2e − oxidation of the C-8-N-9 bonds of caffeine, resulting in 8-oxy-caffeine; followed by the fast 2e − and 2H + oxidation to 4,5-dihydroxy-8-oxy-caffeine [53]. Therefore, the fact that CA oxidation involved two electrons and two protons at MGPE/EA could be due to the first step of this mechanism being a rate determining step.

EVIEW
10 of 19 reaction of CA. Based on the evidence above and previous studies [50][51][52][53], we hypothesized that an equal number of protons and electrons were involved in the oxidation of CA. According to equation 7, the oxidation of CA corresponded to an overall 4H + and 4e − oxidation, following a two-step mechanism. The first mechanism was the slow 2H + and 2e − oxidation of the C-8-N-9 bonds of caffeine, resulting in 8-oxy-caffeine; followed by the fast 2e − and 2H + oxidation to 4, 5-dihydroxy-8-oxy-caffeine [53]. Therefore, the fact that CA oxidation involved two electrons and two protons at MGPE/EA could be due to the first step of this mechanism being a rate determining step.
It was observed in Figure 4C that Ip for biomolecules under investigation decreased from pH 2 to 3 and was stable in pH 3-8 for AA, UA, Trp, XN, and CA. For DA, the peak current was stable in pH 3-5 and then increased with increasing pH from 5 to 7 and then decreased until pH 8.0. Considering all of the results thus far, it is clear that for simultaneous determination of the biomolecules under investigation in the modified graphite paste electrode, the acidic media at pH 2.0 was the most suitable one for senary detection purposes. At low pH, there were two main advantages: well-defined high peak currents and clear peak separations.

Interference Studies
One of the important problems in simultaneous determination is the interference caused by redox-active molecules, which can be oxidized under the same conditions, in biological samples. Among all biomolecules, significant interference was observed between the biomolecules under investigation, as their oxidation peak potentials are near each other and they mainly coexist in real biological samples. Therefore, to study these interfering biomolecules and their effects, the spiking technique was used on the simultaneous quantification of these biomolecules. In all control assays, the concentration of only one species was changed, while the concentrations of the other species were kept constant. The results are shown in Figure S2(a-f). Figure S2(a) shows that the peak current of AA increased by increasing AA concentration when the concentrations of DA, UA, Trp, XN, and CA were kept constant. Although the charge current was enhanced after AA was oxidized, the peak currents of DA, UA, Trp, XN, and CA did not change (± 5%). Similarly, as shown in Figure S2(b-e), when keeping the concentrations of the other five compounds constant, the oxidation peak currents of DP or UA or Trp or XN or CA were almost linearly proportional to the five spiked concentrations, while those of the other five compounds did not change. These results indicated that the sensor was selective for the simultaneous determination of biomolecules under investigation in the presence of other biomolecules that co-exist in the real biological samples.

Calibration Curves
DPV was performed to investigate the relationship between the peak current and concentration of AA, DA, UA, Trp, XN, and CA. The dependence of peak currents on the concentrations of biomolecules under investigation is shown in Figure 5. The results for linear ranges and detection limits of the analytes are summarized in Table S1. Based on (7) It was observed in Figure 4C that I p for biomolecules under investigation decreased from pH 2 to 3 and was stable in pH 3-8 for AA, UA, Trp, XN, and CA. For DA, the peak current was stable in pH 3-5 and then increased with increasing pH from 5 to 7 and then decreased until pH 8.0. Considering all of the results thus far, it is clear that for simultaneous determination of the biomolecules under investigation in the modified graphite paste electrode, the acidic media at pH 2.0 was the most suitable one for senary detection purposes. At low pH, there were two main advantages: well-defined high peak currents and clear peak separations.

Interference Studies
One of the important problems in simultaneous determination is the interference caused by redox-active molecules, which can be oxidized under the same conditions, in biological samples. Among all biomolecules, significant interference was observed between the biomolecules under investigation, as their oxidation peak potentials are near each other and they mainly coexist in real biological samples. Therefore, to study these interfering biomolecules and their effects, the spiking technique was used on the simultaneous quantification of these biomolecules. In all control assays, the concentration of only one species was changed, while the concentrations of the other species were kept constant. The results are shown in Figure S2a-f. Figure S2a shows that the peak current of AA increased by increasing AA concentration when the concentrations of DA, UA, Trp, XN, and CA were kept constant. Although the charge current was enhanced after AA was oxidized, the peak currents of DA, UA, Trp, XN, and CA did not change (± 5%). Similarly, as shown in Figure S2b-e, when keeping the concentrations of the other five compounds constant, the oxidation peak currents of DP or UA or Trp or XN or CA were almost linearly proportional to the five spiked concentrations, while those of the other five compounds did not change. These results indicated that the sensor was selective for the simultaneous determination of biomolecules under investigation in the presence of other biomolecules that co-exist in the real biological samples.

Calibration Curves
DPV was performed to investigate the relationship between the peak current and concentration of AA, DA, UA, Trp, XN, and CA. The dependence of peak currents on the concentrations of biomolecules under investigation is shown in Figure 5. The results for linear ranges and detection limits of the analytes are summarized in Table S1. Based on Table S1, the calibration curves were linear in two segmented ranges of substrate concentration. LOD values were assessed using the equation Y LOD = X B + 3S B , where Y LOD is the signal for the detection limit; X B represents the blank signal, and SB represents the standard deviation of the blank signal. The theoretical LODs of the sensor for AA, DA, UA, Trp, XN, and CA were 91, 9.9, 13, 11, 14, and 78 nM, respectively. Based on Figure 5a-f, the calibration curves for AA, DA, UA, Trp, XN, and CA clearly had two linear segments with different slopes, corresponding to two different ranges of substrate concentrations. The difference in the slopes of the calibration curves was due to the different activity of the electrode surface with low and high concentrations of the analytes. In the low concentrations of the analytes, due to a high number of active sites on the surface of electrode, the slope of the first calibration curve was high. While in the higher concentration range of the biomolecules, due to decreasing active sites on the surface of electrode, the slope of the second segment of calibration curve decreased, too. The decreasing of sensitivity (slope) in the second linear range can be attributed to possible kinetic limitations.
A comparison study of the MGPE/MWCNTs-EA with other modified electrodes reported in literature for the simultaneous determination of biomolecules under investigation is shown in Table 1. Based on Table 1, the modified graphite paste electrode seems to provide a favorable alternative for the simultaneous determination of biomolecules under investigation with better or satisfactory results than those described in the literature [49,[54][55][56][57][58][59].  Table S1, the calibration curves were linear in two segmented ranges of substrate concentration. LOD values were assessed using the equation YLOD = XB + 3SB, where YLOD is the signal for the detection limit; XB represents the blank signal, and SB represents the standard deviation of the blank signal. The theoretical LODs of the sensor for AA, DA, UA, Trp, XN, and CA were 91, 9.9, 13, 11, 14, and 78 nM, respectively. Based on Figure 5a-f, the calibration curves for AA, DA, UA, Trp, XN, and CA clearly had two linear segments with different slopes, corresponding to two different ranges of substrate concentrations. The difference in the slopes of the calibration curves was due to the different activity of the electrode surface with low and high concentrations of the analytes. In the low concentrations of the analytes, due to a high number of active sites on the surface of electrode, the slope of the first calibration curve was high. While in the higher concentration range of the biomolecules, due to decreasing active sites on the surface of electrode, the slope of the second segment of calibration curve decreased, too. The decreasing of sensitivity (slope) in the second linear range can be attributed to possible kinetic limitations.  Table 1. Based on Table 1, the modified graphite paste electrode seems to provide a favorable alternative for the simultaneous determination of biomolecules under investigation with better or satisfactory results than those described in the literature [49,[54][55][56][57][58][59].  The analytical performance of our modified graphite paste electrode was also compared with reported standard method (high performance liquid chromatography); the results are shown in Table 2. Based on the results in Table 2, we conclude that there was good agreement between the results obtained with our electrochemical method and the standard one.

Stability and Reproducibility
The stability and reproducibility of the modified graphite paste electrode were investigated for ten successive determinations (n = 10). Differential pulse voltammograms of a mixture of AA (1.  Figure S3. The relative standard deviation of results was calculated as 0.07%, 0.06%, 0.08%, 0.16%, 0.06%, and 0.61% for AA, DA, UA, Trp, XN, and CA, respectively. Thus, the electrochemical signals of biomolecules on the MGPE/MWCNTs-EA had excellent stability and reproducibility. The long-term stability of MGPE/MWCNTs-EA was also tested by recording DPVs of the modified graphite paste electrode in a period of 30 days for simultaneous determination of AA, DA, UA, Trp, XN, and CA. Based on the results period, the MGPE/MWCNTs-EA was used extensively and its response was recorded twice a week for the same concentrations of biomolecules under investigation. The ∆I pa values AA, DA, UA, Trp, XN, and CA were retained at 94.3%, 96.6%, 97.2%, 94.6%, 95.1%, and 94.7%, respectively. These results can be attributed to excellent repeatability and stability of the modified graphite paste electrode response for simultaneous determination of these biomolecules.

Real Sample Analysis
To evaluate the practical applicability of our electrochemical approach, the modified graphite paste electrode was utilized for the simultaneous determination of AA, DA, UA, Trp, XN, and CA in human serum and blood plasma. The samples were diluted with 0.1 M PBS (pH 2.0) and DPVs were recorded using the standard addition method; the results were confirmed with the literature results obtained using the HPLC standard method [59][60][61][62][63]. As depicted in Table 2, acceptable recovery values were obtained, which indicated the applicability of MGPE/MWCNTs-EA as a promising tool for individual and simultaneous trace amount determination of AA, DA, UA, Trp, and XN in real samples.

Conclusions
The proof-of-concept study demonstrates a simple and affordable MGPE/MWCNTs-EA system for sensitive and selective simultaneous determination of AA, DA, UA, Trp, XN, and CA. The modified graphite paste electrode not only improved electrochemical catalytic activities toward the oxidation of the biomolecules under investigation, but also decreased the overpotentials of the AA, DA, UA, Trp, XN, and CA. The nanocompositemodified electrode was compared to the previously reported modified electrodes for the simultaneous determination of similar biomolecules and the results indicated that the nanocomposite-modified electrode provided a promising platform for the simultaneous determination of biomolecules under investigation with highly sensitive results in real samples.
Supplementary Materials: The following figures and tables are available online at www.mdpi.com/xxx/s1. Table S1. Analytical parameters for the simultaneous determination of AA,

Real Sample Analysis
To evaluate the practical applicability of our electrochemical approach, the modified graphite paste electrode was utilized for the simultaneous determination of AA, DA, UA, Trp, XN, and CA in human serum and blood plasma. The samples were diluted with 0.1 M PBS (pH 2.0) and DPVs were recorded using the standard addition method; the results were confirmed with the literature results obtained using the HPLC standard method [59][60][61][62][63]. As depicted in Table 2, acceptable recovery values were obtained, which indicated the applicability of MGPE/MWCNTs-EA as a promising tool for individual and simultaneous trace amount determination of AA, DA, UA, Trp, and XN in real samples.

Conclusions
The proof-of-concept study demonstrates a simple and affordable MGPE/MWCNTs-EA system for sensitive and selective simultaneous determination of AA, DA, UA, Trp, XN, and CA. The modified graphite paste electrode not only improved electrochemical catalytic activities toward the oxidation of the biomolecules under investigation, but also decreased the overpotentials of the AA, DA, UA, Trp, XN, and CA. The nanocomposite-modified electrode was compared to the previously reported modified electrodes for the simultaneous determination of similar biomolecules and the results indicated that the nanocompositemodified electrode provided a promising platform for the simultaneous determination of biomolecules under investigation with highly sensitive results in real samples.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to pending patent applications.