Au Nanoparticles on 4-Thiophenol-Electrodeposited Carbon Surfaces for the Simultaneous Detection of 8-Hydroxyguanine and Guanine

: In this proof-of-concept study, gold nanoparticles (AuNPs) were immobilized on glassy carbon electrode (GCE) surfaces using a surface-anchored diazonium salt of 4-aminothiophenol (GCE-Ph-S-AuNPs). X-ray photoelectron spectroscopy (XPS) studies conﬁrmed the attachment of the AuNPs via 4-thiophenol onto the surface of the modiﬁed electrode. Differential pulse voltammetry (DPV) was performed for the simultaneous determination of guanine (G) and 8-hydroxyguanine (8-OH-G). The calibration curves were linear up to 140 µ M and 60 µ M with a limit of detection of 0.02 µ M and 0.021 µ M for G and 8-OH-G, respectively. Moreover, chronoamperometric studies were carried out for the determination of diffusion coefﬁcients of 8-OH-G and G. The GCE-Ph-S-AuNPs were also applied in genomic DNA-spiked samples for the determination of G and 8-OH-G with recovery rates between 98.5% and 103.3%. The novel electrochemical surface provided a potential platform for the sensitive detection of 8-OH-G related to oxidative stress-induced DNA damage in clinical studies.


Introduction
Deoxyribonucleic acid (DNA) is transcribed to ultimately form proteins that serve both structural and functional purposes for our cells to survive [1,2]. DNA is composed of a sugar-phosphate backbone, as well as nitrogenous bases [3]. The information which pertains to the proper transcription of proteins is stored in the form of nitrogenous bases, the four most recognized being adenine, thymine, cytosine, and guanine (G). In addition to these nucleobases functioning as the genetic code of organisms, they also have several metabolic functions in extracellular signaling pathways [4,5]. Any changes made to the arrangement of the four nitrogenous bases or harm caused to them can potentially modify the sequence of the protein it transcribes, and hence, it may change the structure of the organism as well as trigger the emergence of specific illnesses [6].
One of the ways in which these nucleobases may be damaged is due to oxidative damage resulting from reactive oxygen species (ROS), such as hydrogen peroxide, superoxide anions, and hydroxyl radicals, which may be generated from cellular oxygen metabolism, environmental carcinogens, or ionizing radiation [7,8]. One way in which ROS may damage DNA is by oxidizing G to 8-hydroxyguanine (8-OH-G), which has been implicated in several neurodegenerative diseases including Alzheimer's disease [9,10], Parkinson's disease [11,12], as well as some cancers [13,14]. Furthermore, a recent study suggested that 8-OH-G detected in saliva was a biomarker for oxidative stress in workers [15]. The salivary 8-OH-G levels were significantly elevated in older persons, as well as those who smoked, had hypertension, or excess visceral fat [15]. Therefore, the detection of 8-OH-G 4-aminothiophenol (SH-Ph-NH 2 , ≥97.0%) and acetonitrile (CH 3 CN, ≥99.9%), doublestranded and lyophilized fish sperm DNA (FS-DNA) with low molecular-weight were all purchased from Sigma-Aldrich Company (Oakville, ON, Canada). Sodium hydroxide (NaOH, ≥97.0%), and methanol (CH 3 OH, ≥99.9%) were purchased from ACP Chemicals Inc. (Montreal, QC, Canada). Hydrochloric acid (HCl, 36.5-38.0%) was purchased from Caledon Laboratory Chemicals (Georgetown, ON, Canada). 8-OH-G (≥90.0%) was purchased from Cayman Chemical Company (Ann Arbor, MI, USA). All alumina powder (1.0, 0.3, 0.05 µm) were purchased from Allied High Tech Products Inc. (Rancho Dominguez, CA, USA). Phosphate electrolytes and buffer solutions (PBS) were prepared at a concentration of 0.2 M using phosphoric acid (H 3 PO 4 , 85.0%) purchased from Fischer Scientific (Mississauga, ON, Canada), and the pH was adjusted using 10 M NaOH. The stock solutions for both G and 8-OH-G (0.01 M) were made by dissolving each compound in deionized water in the presence of 100 µL of 10 M NaOH and sonicating the solution for approximately 5 min.

Instrumentation
Transmission Electron Microscopy (TEM) was performed using a Hitachi H7500 Transmission Electron Microscope (Hitachi Ltd., Tokyo, Japan) equipped with an Olympus SIS MegaView II 1.35 MB digital camera. TEM images were processed using iTEM version 5.2 software. X-ray photoelectron spectroscopy (XPS) was performed using a Thermo Scientific K-Alpha spectrometer (Mississauga, ON, Canada) equipped with a monochromated Al Kα X-ray source (1486.6 eV), using an acquisition angle of 90 • with a 20-eV pass energy, and the acquisition chamber being at a pressure of 10 −8 mbar. The sonication of the DNA base solutions (G and 8-OH-G) was performed using a VWR B2500A-DTH ultra-sonicator (Thermo Scientific, Mississauga, ON, Canada). The pH of the solutions was measured using a VWR SB70P pH meter (Thermo Scientific, Mississauga, ON, Canada). Glassy carbon electrodes (GCEs) (3.0 mm diameter) were purchased from CHInstruments Inc. (Austin, TX, USA). For XPS studies, glassy carbon plates (GCP) (10 mm × 10 mm) were purchased from London Scientific Limited (London, ON, Canada). Electrochemical studies were performed at room temperature using Autolab Potentiostat/Galvanostat (PGSTAT 302N, Metrohm AG, Herisau, Switzerland) and the results were processed using the NOVA TM software (NOVA 2.1.2, Metrohm AG, Herisau, Switzerland). The three-electrode system was comprised of the modified GCE as a working electrode. The counter electrode was a Pt rod, and the reference electrode was a saturated Ag/AgCl electrode housed in a 3 M KCl solution. Differential pulse voltammetry (DPV) measurements were measured between −0.3-+1.0 V, at a step potential of 0.005 V, a modulation amplitude of 0.0025 V, and a modulation time of 0.05 s.

AuNP Synthesis and Electrode Modification
The synthesis of the AuNPs was adapted from the protocol reported by Liu and Lu [33]. Briefly, 100 mL of a 1 mM solution of HAuCl 4 was dissolved in deionized (DI) water and heated on a hotplate. Once the solution reached boiling, 10 mL of a 38.8 mM solution of sodium citrate was added, resulting in a color change from pale yellow to dark blue to eventually wine red. The solution was allowed to stir overheat for another 15 min, after which it was allowed to cool down to room temperature. The AuNP solution was stored at 4 • C until use.
The protocol for the modification of GCE and GCP was adapted from Liu et al. [32]. The GCEs were polished using mixtures of 1.0, 0.3, and 0.05 µm alumina powder mixed with DI water in a sequence of 10 min with each. The electrodes were then rinsed with copious amounts of DI water, followed by sonication for 5 min in both DI water and ethanol, respectively.
The electrodeposition of the 4-aminothiophenol diazonium analogue was achieved by generating the aryl diazonium in situ. A 25 mL solution of 1 mM 4-aminothiophenol in 0.5 M HCl was mixed with NaNO 2 (1 mM) under N 2 gas (99.999%) at 0 • C. The reaction was allowed for 15 min. The electrodeposition of the in situ generated diazonium salt was carried out under N 2 gas conditions by running cyclic voltammetry (CV) between the potentials of +0.6 V to −1.0 V at a scan rate of 100 mV/s for 2, 10, and 50 scans (Figure 1). The anodic peak current values remained the same for electrodeposition procedures beyond 50 cycles, thus, we determined the application of 50 cycles for the optimum condition for the diazonium electrodeposition. The modified GCE is denoted as GCE-Ph-S-AuNPs. copious amounts of DI water, followed by sonication for 5 min in both DI water and ethanol, respectively.
The electrodeposition of the 4-aminothiophenol diazonium analogue was achieved by generating the aryl diazonium in situ. A 25 mL solution of 1 mM 4-aminothiophenol in 0.5 M HCl was mixed with NaNO2 (1 mM) under N2 gas (99.999%) at 0 °C. The reaction was allowed for 15 min. The electrodeposition of the in situ generated diazonium salt was carried out under N2 gas conditions by running cyclic voltammetry (CV) between the potentials of +0.6 V to −1.0 V at a scan rate of 100 mV/s for 2, 10, and 50 scans (Figure 1). The anodic peak current values remained the same for electrodeposition procedures beyond 50 cycles, thus, we determined the application of 50 cycles for the optimum condition for the diazonium electrodeposition. The modified GCE is denoted as GCE-Ph-S-AuNPs. The electrodes were then rinsed with DI water, acetonitrile, and DI water in sequence, followed by drying the electrodes under a gentle stream of N2 gas. Once the electrodes were modified, they were incubated in the AuNPs solution overnight at room temperature. The electrodes were then rinsed with DI water and allowed to dry. The three-electrode system included the GCE-Ph-S-AuNPs as the working electrode, a platinum foil as the auxiliary electrode, and a saturated Ag/AgCl as the reference electrode. All potentials were reported versus the saturated Ag/AgCl reference electrode at room temperature.

Real Sample Preparation
To prepare the real DNA sample (salmon sperm), 1 mg of FS-DNA was dissolved in 25 mL of 0.2 M phosphate buffer at pH 7. This stock solution was used for the preparation of the spiked solutions of G and 8-OH-G. Artificial cerebrospinal fluid (ACSF) was prepared by a method described by Tamano et al. [34] This solution consisted of 119 mM NaCl, 36.2 mM NaHCO3, 2.5 mM KCl, 1.3 mM MgCl2, and 10 mM glucose (pH = 7.3). The reagents were added and stored at 4 °C until use. Figure 2 shows the TEM images taken of the AuNPs synthesized. The TEM images show that the AuNPs had a near-uniform size, which was approximately 10 nm i.d., well in agreement with previous reports [35]. The electrodes were then rinsed with DI water, acetonitrile, and DI water in sequence, followed by drying the electrodes under a gentle stream of N 2 gas. Once the electrodes were modified, they were incubated in the AuNPs solution overnight at room temperature. The electrodes were then rinsed with DI water and allowed to dry. The three-electrode system included the GCE-Ph-S-AuNPs as the working electrode, a platinum foil as the auxiliary electrode, and a saturated Ag/AgCl as the reference electrode. All potentials were reported versus the saturated Ag/AgCl reference electrode at room temperature.

Real Sample Preparation
To prepare the real DNA sample (salmon sperm), 1 mg of FS-DNA was dissolved in 25 mL of 0.2 M phosphate buffer at pH 7. This stock solution was used for the preparation of the spiked solutions of G and 8-OH-G. Artificial cerebrospinal fluid (ACSF) was prepared by a method described by Tamano et al. [34] This solution consisted of 119 mM NaCl, 36.2 mM NaHCO 3 , 2.5 mM KCl, 1.3 mM MgCl 2 , and 10 mM glucose (pH = 7.3). The reagents were added and stored at 4 • C until use. Figure 2 shows the TEM images taken of the AuNPs synthesized. The TEM images show that the AuNPs had a near-uniform size, which was approximately 10 nm i.d., well in agreement with previous reports [35]. Figure 3 shows the XPS of the surfaces deposited with either the thiophenol moiety (i) or the thiophenol moiety along with the AuNPs (ii). As shown in Figure 2A, the C 1s spectra were deconvoluted using a five-peak model. The spectra for both GCE with and without the AuNPs are nearly identical indicating the glassy carbon surface and the electrodeposited thiophenol moiety remained intact following the incubation with the AuNPs. Specifically, the peaks deconvoluted at 284.4 and 288.3 eV corresponded to graphitic carbon and a carbonyl group, respectively, originating from the glassy carbon surface, while the peak at 286.5 eV corresponded to the phenolic groups found on both the glassy carbon surface as well as the electrodeposited thiophenol group [36]. The deconvoluted S 2p spectra ( Figure 2B) for the sensor before incubation with the AuNPs and after incubation showed peaks at 164.5 and 164.7 eV which were both indicative of the thiophenol moiety as reported by Kwan et al. [36]. Lastly, the Au 4f spectra of the modified GCE before incubation with the AuNPs showed no deconvoluted peaks, however after incubation, two peaks were deconvoluted at 84.0 and 87.7 eV corresponding to the Au 4f 7/2 and Au 4f 5/2 spin orbit coupling characteristic of AuNPs with an oxidation state of 0 [37]. Table S1 shows the relative atomic percentage of each element before incubating with AuNPs and after incubation. According to this table, the modified GCE with thiophenol diazonium salt (GCE-Ph-SH) had C and S but no N. This confirmed that the diazonium group formed N 2(g) after electrografting. Because AuNPs had a higher affinity for the sulfur group, the sulfur was covered by AuNPs and thus, the sulfur percentage dropped as the percentage of Au increased.  Figure 3 shows the XPS of the surfaces deposited with either the thiophenol moiety (i) or the thiophenol moiety along with the AuNPs (ii). As shown in Figure 2A, the C 1s spectra were deconvoluted using a five-peak model. The spectra for both GCE with and without the AuNPs are nearly identical indicating the glassy carbon surface and the electrodeposited thiophenol moiety remained intact following the incubation with the AuNPs. Specifically, the peaks deconvoluted at 284.4 and 288.3 eV corresponded to graphitic carbon and a carbonyl group, respectively, originating from the glassy carbon surface, while the peak at 286.5 eV corresponded to the phenolic groups found on both the glassy carbon surface as well as the electrodeposited thiophenol group [36]. The deconvoluted S 2p spectra ( Figure 2B) for the sensor before incubation with the AuNPs and after incubation showed peaks at 164.5 and 164.7 eV which were both indicative of the thiophenol moiety as reported by Kwan et al. [36]. Lastly, the Au 4f spectra of the modified GCE before incubation with the AuNPs showed no deconvoluted peaks, however after incubation, two peaks were deconvoluted at 84.0 and 87.7 eV corresponding to the Au 4f7/2 and Au 4f5/2 spin orbit coupling characteristic of AuNPs with an oxidation state of 0 [37]. Table  S1 shows the relative atomic percentage of each element before incubating with AuNPs and after incubation. According to this table, the modified GCE with thiophenol diazonium salt (GCE-Ph-SH) had C and S but no N. This confirmed that the diazonium group formed N2(g) after electrografting. Because AuNPs had a higher affinity for the sulfur

Comparison between the Modified Surfaces
In an aim to ensure that the AuNPs-modified sensor performed better in terms of detecting 8-OH-G and G, a study was performed to compare the electrochemical activity of the AuNPs-modified sensor versus control electrodes. Figure 4 shows the DPV of the comparison study, which was performed in 0.2 M PBS (pH 7.4). The concentration of both nucleobases was kept constant at 2.0 µM for G, and 4.0 µM for 8-OH-G. As shown in Figure 3, there was a significant difference in terms of sensitivity between the modified sensor at different scan rates and the bare gold electrode (AuE) and the bare GCE. The average anodic peak current values (n = 3) for 8-OH-G were 0.17, 0.52 µA at the bare GCE and bare AuE, and 1.10, 1.23, and 1.56 µA at the GCE-Ph-S-AuNPs prepared with 2, 10 and 50 cycles of diazonium electrodeposition, respectively, which showed an approximately 50-fold increase in sensitivity in comparison with the bare electrode. The average anodic peak current values (n = 3) for G were 0.08, 0.21, 0.32, 0.54, 0.67 µA at the bare GCE, bare AuE, and the GCE-Ph-S-AuNPs prepared with 2, 10 and 50 cycles, respectively. The oxidation peaks of 8-OH-G and G were separated with a considerable enhancement by using the AuNPs-modified sensor. These results indicated that the well-defined voltammetric signals were promising for sensitive and simultaneous determination of 8-OH-G and G.

Comparison between the Modified Surfaces
In an aim to ensure that the AuNPs-modified sensor performed better in terms of detecting 8-OH-G and G, a study was performed to compare the electrochemical activity of the AuNPs-modified sensor versus control electrodes. Figure 4 shows the DPV of the comparison study, which was performed in 0.2 M PBS (pH 7.4). The concentration of both nucleobases was kept constant at 2.0 µM for G, and 4.0 µM for 8-OH-G. As shown in Figure 3, there was a significant difference in terms of sensitivity between the modified sensor at different scan rates and the bare gold electrode (AuE) and the bare GCE. The average anodic peak current values (n = 3) for 8-OH-G were 0.17, 0.52 µA at the bare GCE and bare AuE, and 1.10, 1.23, and 1.56 µA at the GCE-Ph-S-AuNPs prepared with 2, 10 and 50 cycles of diazonium electrodeposition, respectively, which showed an approximately 50-fold increase in sensitivity in comparison with the bare electrode. The average anodic peak current values (n = 3) for G were 0.08, 0.21, 0.32, 0.54, 0.67 µA at the bare GCE, bare AuE, and the GCE-Ph-S-AuNPs prepared with 2, 10 and 50 cycles, respectively. The oxidation peaks of 8-OH-G and G were separated with a considerable enhancement by using the AuNPsmodified sensor. These results indicated that the well-defined voltammetric signals were promising for sensitive and simultaneous determination of 8-OH-G and G.

pH Study
The acidity of the electrolyte had a significant influence on the 8-OH-G and G electrooxidation because protons took part in the redox reaction at the electrode surface. To determine the optimum pH at which the GCE-Ph-S-AuNPs operated, as well as to determine the electrochemical oxidation mechanism of both G and 8-OH-G, the sensor was tested in solutions of PBS at varying pH (pH 4.0, 5.0, 6.0, 7.0, 7.4, 8.4). Figure 5A shows the DPV of the GCE-Ph-S-AuNPs detecting G and 8-OH-G in the different pH conditions. The peak potentials of the analytes shifted to more negative values when the pH increased, as shown in Figure 5A, because protons participated in the oxidation of these analytes [38,39]. The optimum pH was determined to be 7.4, since at this pH, the GCE-Ph-S-AuNPs exhibited excellent sensitivity for both G and 8-OH-G. Furthermore, Figure 5B shows a plot of the anodic peak potentials for G and 8-OH-G with respect to the pH, as well as the equations representing a linear relationship. The slopes of the linear relationship between the anodic peak potential and pH for both G and 8-OH-G were found to be 55.8 mV/pH and 62.5 mV/pH, respectively. Given that both slopes were near the Nernstian theoretical value of 59.1 mV/pH, it was concluded that the electrochemical oxidation mechanism for both analytes involved an equal number of protons and electrons transferred in a similar fashion to those reported in prior literature [40][41][42]. Figure 6 shows the proposed electrochemical oxidation mechanisms for both 8-OH-G ( Figure 6a) and G (Figure 6b), based on the correlation of anodic peak potential as a function of pH. G was oxidized to 8-OH-G via the loss of two electrons and two protons, which could then undergo a ketone-enol tautomerization. The oxidation of 8-OH-G took place with the loss of two protons and two electrons. Once 8-OH-G was oxidized, the resulting compound was then broken down to 2,5-diamino-4-imidazolone and 5-guanidohydantoin, which was well documented in the literature [40][41][42].

pH Study
The acidity of the electrolyte had a significant influence on the 8-OH trooxidation because protons took part in the redox reaction at the electro determine the optimum pH at which the GCE-Ph-S-AuNPs operated, as w mine the electrochemical oxidation mechanism of both G and 8-OH-G, tested in solutions of PBS at varying pH (pH 4.0, 5.0, 6.0, 7.0, 7.4, 8.4). Figur DPV of the GCE-Ph-S-AuNPs detecting G and 8-OH-G in the different pH peak potentials of the analytes shifted to more negative values when the p shown in Figure 5A, because protons participated in the oxidation of [38,39]. The optimum pH was determined to be 7.4, since at this pH, the GC exhibited excellent sensitivity for both G and 8-OH-G. Furthermore, Figu plot of the anodic peak potentials for G and 8-OH-G with respect to the pH equations representing a linear relationship. The slopes of the linear relatio the anodic peak potential and pH for both G and 8-OH-G were found to and 62.5 mV/pH, respectively. Given that both slopes were near the Nerns value of 59.1 mV/pH, it was concluded that the electrochemical oxidation both analytes involved an equal number of protons and electrons transfer fashion to those reported in prior literature [40][41][42]. Figure 6 shows the pr chemical oxidation mechanisms for both 8-OH-G (Figure 6a) and G (Figur the correlation of anodic peak potential as a function of pH. G was oxidi via the loss of two electrons and two protons, which could then undergo

Calibration Study
In order to determine the limit of detection (LOD) for G and 8-OH-G, a calibration curve was constructed by simultaneously spiking G and 8-OH-G to 0.2 M PBS (pH 7.4), and the overlaid DPV curves are shown in Figure 7A. Calibration plots were constructed by plotting the respective current peaks of 8-OH-G and G with respect to the concentration of the nucleobases ( Figure 7B,C). For both G and 8-OH-G, two linear ranges were observed. For 8-OH-G, the linear ranges were determined to be between 1.3-12.0 µM and 12.0-125.0 µM, while for G, the linear ranges were determined to be between 0.3-12.0 µM and 12.0-60.0 µM. This phenomenon of two linear ranges was also observed in the literature [43][44][45][46][47], in which it was hypothesized that calibration plots displayed variable slopes in a wide concentration range of biomolecules. The analytes diffused to the surface easily at lower concentrations, whereas at higher concentrations, the adsorption of the analytes at the surface of the electrode hindered the diffusion as well as slowed down the kinetics of the oxidation processes [48,49]. At low concentrations, the redox-active surface area was available to the biomolecules. However, once the concentration of the analyte increased, the biomolecules saturated the redox-active surface areas leading to a decrease in the sensitivity of the electrode. As a result, we observed a decrease in the slope at higher concentrations. The LOD was calculated using the equation LOD = 3S bk m . In this equation, S bk is the standard deviation of the blank signals (n = 10), and m is the slope of the linear calibration curve. Based on this equation, the LOD was found to be 0.020 and 0.021 µM for 8-OH-G and G, respectively.   Table 1 shows a comparison of sensors reported in the literature for the detection of either 8-OH-G, G, or both. GCE-Ph-S-AuNPs detected both a lower LOD and a wider linear range for 8-OH-G than those reported in the literature [40,41]. Our sensor displayed relatively similar performance with other sensors that detected G [46,47]. Furthermore, GCE-Ph-S-AuNPs was one of the few reported sensors that detected 8-OH-G and G simultaneously. The application of GCE-Ph-S-AuNPs led to an increase in sensitivity and potential window range, making it well-suited for studying the redox processes of 8-OH-G and G. The high surface area of GCE allowed for a wide potential window range that enabled a broader range of electrochemical reactions to be studied. Additionally, the incorporation of AuNPs enhanced the stability of this sensor, making it a highly favorable choice for this study. In addition, the repeatability of the GCE-Ph-S-AuNPs was studied using the DPV measurements for 10 consecutive measurements (n = 10) by simultaneously detecting 8-OH-G and G, and the results are shown in Figure S1A. The relative standard deviation (RSD) of the results were 1.72 and 1.75% for 8-OH-G and G, respectively. These results confirmed the excellent repeatability of the sensor. The stability of the GCE-Ph-S-AuNPs was explored after storing the sensor for 5 months in 0.2 M phosphate electrolyte solution (pH 7.4). The same concentrations of the analytes were used to record the DPV signals. The anodic peak current values of 8-OH-G and G were constant with negligible shifts in the anodic peak potentials ( Figure S1B). The RSD for anodic peak current values for 8-OH-G and G were 1.32% and 2.45%, respectively, which indicated excellent long-term stability of the GCE-Ph-S-AuNPs.
Chemosensors 2023, 10, x FOR PEER REVIEW 9 Figure 6. The oxidation mechanism of (a) 8-OH-G and (b) G as deduced from the effect of va the pH with respect to the oxidation potential peaks using DPV.

Calibration Study
In order to determine the limit of detection (LOD) for G and 8-OH-G, a calibr curve was constructed by simultaneously spiking G and 8-OH-G to 0.2 M PBS (pH and the overlaid DPV curves are shown in Figure 7A. Calibration plots were constru by plotting the respective current peaks of 8-OH-G and G with respect to the concentr of the nucleobases (Figure 7B,C). For both G and 8-OH-G, two linear ranges wer served. For 8-OH-G, the linear ranges were determined to be between 1.3-12.0 µM 12.0-125.0 µM, while for G, the linear ranges were determined to be between 0.3-12.0 and 12.0-60.0 µM. This phenomenon of two linear ranges was also observed in the li ture [43][44][45][46][47], in which it was hypothesized that calibration plots displayed variable sl in a wide concentration range of biomolecules. The analytes diffused to the surface e at lower concentrations, whereas at higher concentrations, the adsorption of the ana at the surface of the electrode hindered the diffusion as well as slowed down the kin of the oxidation processes [48,49]. At low concentrations, the redox-active surface area available to the biomolecules. However, once the concentration of the analyte incre the biomolecules saturated the redox-active surface areas leading to a decrease in the sitivity of the electrode. As a result, we observed a decrease in the slope at higher con Figure 6. The oxidation mechanism of (a) 8-OH-G and (b) G as deduced from the effect of varying the pH with respect to the oxidation potential peaks using DPV.

Interference Study
The oxidative stress biomarker, 8-OH-G can coexist with G in real samples, thus, it is of great interest to study the interferences between them for the selective detection of each species. In all control experiments, the concentration of one species was changed, while the concentration of the other species was kept constant. An interference study was performed to determine whether the detection of 8-OH-G interfered with G or vice-versa ( Figure 8). This was conducted by gradually increasing the concentration of either 8-OH-G or G from 6.7-66.7, while keeping the concentration of the other nucleobase constant at 6.7 µM. As shown in both interference studies for 8-OH-G ( Figure 8A) and G ( Figure 8B), progressively increasing the concentration of one nucleobase did not interfere with the detection of the other nucleobase. Other DNA bases such as adenine, thymine, and cytosine showed oxidation peaks at higher anodic potentials beyond 1 V (vs. Ag/AgCl) under the same ionic strength and pH conditions [39,44], and thus, they did not show any interference on the simultaneous determination of 8-OH-G and G.   Table 1 shows a comparison of sensors reported in the literature for the detection either 8-OH-G, G, or both. GCE-Ph-S-AuNPs detected both a lower LOD and a wider ear range for 8-OH-G than those reported in the literature [40,41]. Our sensor display relatively similar performance with other sensors that detected G [46,47]. Furthermo GCE-Ph-S-AuNPs was one of the few reported sensors that detected 8-OH-G and G s ultaneously. The application of GCE-Ph-S-AuNPs led to an increase in sensitivity and tential window range, making it well-suited for studying the redox processes of 8-OH and G. The high surface area of GCE allowed for a wide potential window range t enabled a broader range of electrochemical reactions to be studied. Additionally, the corporation of AuNPs enhanced the stability of this sensor, making it a highly favora choice for this study. In addition, the repeatability of the GCE-Ph-S-AuNPs was stud using the DPV measurements for 10 consecutive measurements (n = 10) by simultaneou detecting 8-OH-G and G, and the results are shown in Figure S1A. The relative stand deviation (RSD) of the results were 1.72 and 1.75% for 8-OH-G and G, respectively. Th results confirmed the excellent repeatability of the sensor. The stability of the GCE-Ph AuNPs was explored after storing the sensor for 5 months in 0.2 M phosphate electrol solution (pH 7.4). The same concentrations of the analytes were used to record the D signals. The anodic peak current values of 8-OH-G and G were constant with neglig shifts in the anodic peak potentials ( Figure S1B). The RSD for anodic peak current val for 8-OH-G and G were 1.32% and 2.45%, respectively, which indicated excellent lo

Electrochemical Impedance Spectroscopy (EIS) and Scan Rate Study
EIS is a powerful technique for analyzing interfacial properties and the shape of the impedance arc (depressed or distinct) is dependent on the contact impedance between the substrate and the active material on the surface of the electrodes. For fitting and simulation of EIS data, the NOVA™ software (NOVA 2.1.2, Metrohm AG, Herisau, Switzerland) was used, and the modified Randles equivalent circuit is shown in the inset of Figure S2. In this circuit, R s , R ct , C dl , and Z W represent solution resistance, charge-transfer resistance, double-layer capacitance, and Warburg element, respectively (inset in Figure S2). For bare GCE, the Nyquist plots revealed a semicircle with an average R ct of 1719.3 Ω (n = 3). The R ct decreased to 879.1, 478.4, and 425.7 Ω for the GCE-Ph-S-AuNPs that were prepared with 2, 10, and 50 scans of diazonium electrodeposition, respectively, resulting in a significantly lower R ct for GCE-Ph-S-AuNPs-50 scans in comparison to the rest of the electrodes. Surface modifications with AuNPs increased the conductivity of the electrode by lowering the R ct . The remarkable decrease in the R ct of GCE-Ph-S-AuNPs-50 scans was attributed to the presence of AuNPs with good electrical conductivity at the electrode interface. GCE-Ph-S-AuNPs

Electrochemical Impedance Spectroscopy (EIS) and Scan Rate Study
EIS is a powerful technique for analyzing interfacial properties and the shape of the impedance arc (depressed or distinct) is dependent on the contact impedance between the substrate and the active material on the surface of the electrodes. For fitting and simulation of EIS data, the NOVA™ software (NOVA 2.1.2, Metrohm AG, Herisau, Switzerland) was A scan rate study was performed on both the GCE-Ph-S-AuNPs as well as a bare GCE to assess the electroactive surface area of both electrodes. This was conducted by performing successive CVs in a solution of 1.0 mM K 3 [Fe(CN) 6 ] with 0.1 M KCl as the supporting electrolyte while varying the scan rates from 5 mV/s to 500 mV/s. Figure S3 shows an overlay of the cyclic voltammograms with the varying scan rates for bare GCE ( Figure S3A) and GCE-Ph-S-AuNPs ( Figure S3B), respectively. To calculate the electroactive surface area, the Randles-Sevcik Equation (1) [48] was used as follows: where i p is the anodic peak current, n is the number of electrons (n = 1) transferred, A is the electroactive surface area. C 0 and D are the concentration and diffusion coefficient of K 3 Fe(CN) 6 , respectively and v is the scan rate. The D of K 3 Fe(CN) 6 is 7.6 × 10 −6 cm 2 ·s −1 [48,49]. The electroactive surface area calculated for the bare GCE and the GCE-Ph-S-AuNPs were 0.0494 cm 2 and 0.1170 cm 2 , respectively. This indicated that the GCE-Ph-S-AuNPs had an electroactive surface area 2.37 times larger compared to the bare GCE which facilitated an improved electrocatalytic performance due to the presence of AuNPs.

Chronoamperometry
The diffusion coefficient (D) is governed by the movement and reaction of analytes at the electrode surface. If the diffusion rate is too slow, the reaction may result in poor electrochemical performance. Controlling the diffusion rate and improving the electrochemical performance can be achieved by adjusting various experimental factors such as the electrode design and solution conditions. To calculate the D for 8-OH-G and G, chronoamperometry was performed using GCE-Ph-S-AuNPs. Figures S4A and S5A show the chronoamperograms for 8-OH-G and G, respectively. The Cottrell plots were drawn using the current values obtained from the chronoamperograms with respect to the inverse square root of time ( Figures S4B and S5B). Lastly, the slopes extrapolated from the Cottrell plot were plotted with respect to the concentration of 8-OH-G and G (Figures S4C and S5C). Using these plots, the D values of 8-OH-G and G were determined using the Cottrell Equation (2) as shown below: where i is the current in Amperes, n is the number of electrons involved in the redox process, F is the Faraday constant, A is the electroactive surface area as determined by the Randles-Sevcik equation in the previous section, C is the concentration of the analyte, D is the diffusion coefficient, and t is the time in seconds. Using the Cottrell equation, the D for 8-OH-G and G were determined to be 1.47 × 10 −4 and 4.64 × 10 −5 cm 2 ·s −1 , respectively.

Real Sample Study
To evaluate the performance of GCE-Ph-S-AuNPs in practical settings, the sensor was subjected to the simultaneous detection measurements of 8-OH-G and G in two challenging matrices, spiked FS-DNA and artificial cerebral spinal fluid (ACSF). The real samples were diluted in 0.2 M PBS (pH 7.4) and the background DPV signals were measured. Randomized concentrations of 8-OH-G and G were then added to the diluted samples using the standard addition method. The recovery signals were recorded in triplicates (n = 3) and displayed in Table 2. 'Detected' corresponds to any previous present analytes found in the diluted solutions prior to spiking, 'Spiked' refers to the analytes added via the standard addition method, and 'Detected' represents the concentration of analyte detected (n = 3) after spiking. The recovery percentage was within an acceptable range, and thus demonstrated that GCE-Ph-S-AuNPs provided a promising platform for real sample analyses in future diagnostic studies. The recovery values were between 96.0 and 106.7%, which were in the acceptable range reported in similar standard addition studies with biological samples.

Conclusions
A novel electrochemical sensor was developed with the electrodeposition of 4-thiophenol via diazonium chemistry for the immobilization of AuNPs using thiol-gold covalent bonding. This nanostructured surface of GCE-Ph-S-AuNPs was applied for the simultaneous detection of 8-OH-G and G. The optimal pH under which the sensor performed was pH 7.4, which was ideal for biological samples. The GCE-Ph-S-AuNPs showed a wide linear range of detection for both 8-OH-G and G using DPV. The sensor was also tested in a spiked DNA samples and the results showed a recovery between 98.5-103.3%. The GCE-Ph-S-AuNPs can potentially be a promising platform for the simultaneous detection of 8-OH-G and G for investigations of the oxidative stress-induced DNA damage in cancers.