Highly Sensitive Electrochemical Sensor for Anticancer Drug by a Zirconia Nanoparticle-Decorated Reduced Graphene Oxide Nanocomposite.

Because of their large surface area and conductivity, some inorganic materials have emerged as good candidates for the trace-level detection of pharmaceutical drugs. In the present work, we demonstrate the detection of an anticancer drug (regorafenib, REG) by using an electrochemical sensor based on a nanocomposite material. We synthesized a zirconia-nanoparticle-decorated reduced graphene oxide composite (ZrO2/rGO) using a one-pot hydrothermal method. Reduction of the graphene oxide supports of the Zr2+ ions with hydrazine hydrate helped in preventing the agglomeration of the zirconia nanoparticles and in obtaining an excellent electrocatalytic response of the nanostructure ZrO2/rGO-based electrochemical sensor. Structural and morphological characterization of the nanostructure ZrO2/rGO was performed using various analytical methods. A novel regorafenib (REG) electrochemical sensor was fabricated by immobilizing the as-prepared nanostructure ZrO2/rGO on to a glassy carbon electrode (GCE). The resulting ZrO2/rGO/GCE could be used for the rapid and selective determination of REG in the presence of ascorbic acid and uric acid. The ZrO2/rGO/GCE showed a linear response for the REG analysis in the dynamic range 11-343 nM, with a remarkable lower detection limit and limit of quantifications of 17 and 59 nM, respectively. The newly developed sensor was used for the accurate determination of REG in both serum samples and pharmaceutical formulations, with satisfactory results.

Introduction

Regorafenib [4-(4-(3-(4-chloro-3-(trifluromethyl) phenyl)-3-flurophenoxy)-N-ethylpicolinamide] (BAY 73-4506) is an orally bioavailable multikinase inhibitor (MKI), which also obstructs multiple tumor pathways, inhibiting targets in the receptors of the vascular endothelial growth factor 1–3 (VEGF 1–3), fibroblast growth factor, and platelet-derived growth factor, including the mutant oncogenic kinesis c-KIT, RET, and B-RAF.[1,2] This MKI generates dynamic metabolites, which could become agglomerated, particularly in elderly, malnourished patients or in patients treated for hepatocellular carcinoma, as in the case of other MKIs.[3] Oral drugs present high protein binding and poor bioavailability and are effectively metabolized by CYP3A4 and UGT1A9 in the liver.[4,5] Regorafenib (REG), being orally administrated, may lead to drug interactions and major toxicities that may lead to early termination of the treatment and thus diminish its chances of success. It is important to maintain the benefits of these treatments, particularly in the elderly or in patients treated for metastatic colon cancer and gastrointestinal tumors, which is approved by FDA.[6−11] The most serious adverse reaction was drug-induced hepatotoxicity, and a black box warning has been indicated by the US-FDA.[12] Thus, the detection of this anticancer active drug is extensively important and a universal challenge. Some of the sophisticated analytical methods such as high-performance liquid chromatography (HPLC),[13] liquid chromatography-mass spectrometry (LC-MS),[14−19] and spectrophotometry[20] are used for the detection of REG in urine, plasma, and other biological samples. However, the aforementioned methods are highly expensive, time-consuming, difficult procedures and require skilled personnel for the specimen, which restricts their particle application. To mitigate these issues, as revealed earlier, much effort has been made to develop novel substituted methods. In this concern, the electrochemical technique is one of the best methods due to its easy operation, spontaneous detection, excellent sensitivity, inexpensiveness, simple pretreatment procedure, and short analysis time for monitoring of bioelectroactive molecules and pharmaceutical drugs. However, for the detection of bioelectroactive molecules, these electrochemical methods have some analytical complications like high overpotential requirement, the reversible process at the bare and carbon paste electrode, GCE, and by-products that may be deposited on the electrode surface, which decrease its activity. Nevertheless, a familiar approach to triumph over these issues is electrodes’ surface modification with various materials, like polymers,[21−23] carbon materials, and metal-oxide nanoparticles.[24−27]

For the last few decades, graphene oxide (GO) and reduced graphene oxide (rGO) have received significant interest owing to their excellent properties in electrochemical applications, such as good electric conductivity, large surface area, high chemical activity, and wide electrochemical window.[28,29] Moreover, chemically, rGO is established as a promising supporting material for the uniform distribution of metal-oxide NPs.[30−34]

In recent years, various metal-oxide-doped graphene oxide composites have been widely used in electrochemical devices and electrocatalysis. Metal oxides, particularly, transition-metal oxides have various physicochemical properties, such as morphological structure, oxygen stoichiometry, good electrochemical conductivity, and interfacial microenvironment of the reaction. Among the transition-metal oxides, zirconium oxide nanoparticles (ZrO2 NPs) show excellent electrochemical properties, including nontoxicity, thermal stability, wide band gap, and good electrical and surface properties and are one of the most abundant metals.[35,36] A critical issue in utilizing bare ZrO2 nanoparticles is that they tend to aggregate and form large clusters during their synthesis.[37,38] In this connection, rGO is an excellent material to mitigate the agglomeration of ZrO2 NPs and subsequently enhance the electrochemical properties. Therefore, researchers have been giving dedicated extensive efforts to synthesize and explore ZrO2 decorated on rGO sheets, for example, Pt/ZrO2-RGO/GCE for significant enhancement of the catechol and hydroquinone oxidation,[39] ZrO2/rGO-based biosensor for detection of the oral cancer drug,[40] and Meth/ZrO2/rGO-based immunosensor.[41]

To the best of our knowledge, this is the first example of electrochemical REG sensing in human blood serum and pharmaceutical formulations using ZrO2/rGO/GCE. In this work, we tried to validate such a voltammetric sensor for the detection of REG. The prepared ZrO2/rGO/GCE can resolve overlapping signals from REG, uric acid (UA), and ascorbic acid (AA). In addition, the present work showed that this sensor possesses an excellent linear dynamic range (LDR) and limit of detection (LOD) for the novel REG determination (Scheme ).

Scheme 1

Synthesis of the ZrO2/rGO Nanocomposite for the Electrochemical Sensing of REG

Results and Discussion

Characterization of the ZrO2/rGO Nanocomposite

Transmission electron microscopy (TEM), high-resolution TEM (HRTEM), and selected area electron diffraction (SAED) analyses were conducted to examine the morphology and structure of the synthesized nanocomposite. The TEM images of the pristine ZrO2, GO, and ZrO2/rGO nanocomposite are shown in Figure . Pristine ZrO2 nanoparticles (Figure a) are nearly spherical with a uniform size of size of 6–9 nm, which is in good agreement with the calculated values based on the powder X-ray diffraction (PXRD) result. Moreover, the displaced-lattice spacing of 0.291 nm, determined from the HRTEM images (blue circles in Figure b) is consistent with the (111) plane of ZrO2. The pristine GO nanosheets are highly wrinkled, and the ZrO2 nanoparticles (blue circles) are well decorated and uniformly distributed on the surface of the wrinkled rGO (Figure d–f). The SAED patterns for ZrO2 (Figure c) and the ZrO2/rGO nanocomposite shown in Figure f (inset) illustrate the crystalline dots instead of amorphous rings, which indicate the polycrystalline nature of the ZrO2 nanoparticles and nanocomposite. To further confirm the formation of the nanocomposition, energy dispersive X-ray spectroscopy (EDX) analysis was employed. The presence of carbon, zirconium, and oxygen elements confirms the presence of ZrO2 on to the GO surface (Figure S3).

Figure 1

TEM, HRTEM, and SAED images of (a–c) pristine ZrO2 and (d–f) ZrO2/rGO nanocomposite.

The phase purities of the as-synthesized GO and the ZrO2/rGO nanocomposite were examined using PXRD. Figure S4a shows a diffraction peak at 2θ = 10.7°, corresponding to the (001) planes of GO. The peaks of the ZrO2/rGO nanocomposite, in Figure S4b, show the existence of both tetragonal and monoclinic mixed phases, which coincide with the standard cards (JCPDS card nos. 49-1642 and 37-1484, respectively) and also show that the peak at 2θ = 10.7°, in Figure S4a, has shifted to 22.4° (002), indicating that GO has been reduced after treating it with the ZrO2 nanoparticles. In addition, the particle size of the ZrO2 nanoparticles was calculated using the Debye–Scherrer equation (D = 0.89λ/β cos θ), where D, λ, β, and θ are the average particle size, wavelength of the Cu Kα irradiation, intensity at the full width at half-maximum of the diffraction peak, and diffraction angle of the (111). The average particle size of pristine ZrO2 is about 7 nm, in good agreement with the TEM result. These results clearly confirm the formation of the ZrO2/rGO nanocomposite.

The Fourier transform infrared (FT-IR) spectra of the GO and ZrO2/rGO samples were recorded at wave numbers 500–4000 cm–1. Pristine GO has a large number of surface functional groups, as shown in Figure S5a, such as O–H, C=O, C=C, and C–O, which are confirmed by the IR bands at 3320, 1700, 1604, and 1009 cm–1, respectively. In the ZrO2/rGO sample shown in Figure S5b, the hydroxyl, carbonyl and epoxide functional groups had disappeared and also overall peak intensities decreased significantly, which confirm reduction of pristine GO leads to the formation of ZrO2/rGO nanocomposite.

The chemical composition was further confirmed by X-ray photoelectron spectrometry (XPS). The wide-survey scan spectrum of the ZrO2-doped rGO nanocomposite (ZrO2/rGO) is shown in Figure S6. The major peaks at 182.5, 284.9, and 530.2 eV are attributed to Zr 4p, C 1s, and O 1s, respectively. In addition, the peaks at 27, 333, and 433 eV attributed to Zr 4p, Zr 3p, and Zr 3s, respectively. The deconvolution spectrum of the Zr 3d peak, Figure S6 (inset), shows binding energies at 182.4 and 184.9 eV attributed to Zr 3d5/2 and Zr 3d3/2, respectively, which can be assigned to the Zr(IV) oxidation state. On the basis of these results, we confirmed that ZrO2 is well embedded into the wrinkled rGO.

Electrochemical Behavior of REG

Figure shows the cyclic voltammograms for the electrocatalytic oxidation of REG on the bare and modified GCE electrodes, recorded in the supporting electrolyte solution (phosphate buffered saline (PBS) 0.1 M, pH 7.0) in the presence of 0.01 mM REG at 50 mV s–1. The voltammograms recorded on the bare GCE in the absence of REG did not show any redox peaks (Figure a), indicating that no faradic reactions occurred on the surface of the unmodified GCE electrode. Figure b shows that the addition of 0.01 mM REG to the supporting electrolyte solution results in the GCE exhibiting a lowest sensitivity reversible couple peak of high separation; ΔEp = 208 mV, which suggests a slow electron transfer. Figure c,d shows the recognizable electrochemical response of the ZrO2/GCE and ZrO2/rGO/GCE, respectively, during the oxidation of 0.01 mM REG, which is interpreted as a result of the enhanced sensitivity, electrode surface area, and improvement of the electrochemical activity of the GO support with ZrO2 NPs. Figure d shows the ZrO2/rGO/GCE and reveals well-defined reversible couple peaks at about Epa = 275 mV and Epc = 306 mV, attributed to the high catalytic effect during the oxidation of 0.1 mM REG. The ZrO2/rGO/GCE remarkably improved the reversible couple peaks and it should be emphasized that the peak-to-peak separation, that is ΔEp, decreased to 31 mV. Finally, the results confirmed that the prepared ZrO2/rGO/GCE significantly improves the electrocatalytic ability to oxidize REG.

Figure 2

Cyclic voltammograms recorded in 0.1 M PBS, pH 7.0, at the scan rate of 100 mV s–1 (a) blank and bare GCE electrode without REG, (b) bare GCE in the presence of 0.01 mM REG, (c) ZrO2-modified GC electrode in the presence of 0.01 mM REG, and (d) ZrO2/rGO/GCE in the presence of 0.01 mM REG.

Effect of pH

The significant effects of the electrolyte pH on the determination of REG by electrocatalysis of ZrO2/rGO/GCE were studied for both current and potential. Figure shows the effect of different pH values, in the range 5.5–8.0, investigated by differential pulse voltammetry (DPV) in a 0.01 mM REG solution and the relationship between Ipa and Epa (anionic peak current and potential, respectively) with the buffer pH. Figure b shows that the anodic peak current of the REG electro-oxidation increases until the pH value becomes 7.0 and then decreases until the end of the experiment (pH 8.0). For this reason, the electrolytic solution with pH 7.0 was chosen for the complete electrocatalytic study. Figure b also shows that the formal REG potential shifts toward lower values with the increase in the supporting electrolytic solution. A better correlation coefficient was obtained for the pH vs Epa, which was confirmed by a slope of 0.053 09 V/pH (R2 = 0.9621), in the range 5.5–8.0. According to the linear regression analysis, the slope of the dEp/dpH being close to the theoretical value of 0.059 V/pH indicates that the irreversible couple peaks involved the transfer of the same number of electrons and protons,[42] in agreement with literature data.[43] The ZrO2/rGO/GCE responds to the oxidation of REG according to the mechanism presented in Scheme . Thus, REG-keto oxidizes to REG-enol (the amide derivative) after an exchange of one electron and one proton via ZrO2/rGO nanocomposite.[36]

Figure 3

(a) DPV voltammograms obtained with the ZrO2/rGO/GCE in an electrolyte solution at different pH values, with 0.01 mM REG. (b) Calibration plot of the anodic peak current (curve-I) and the anodic peak potential (curve-II) vs the pH of the 0.1 M PBS solution, during the electro-oxidation of 0.01 mM REG, at a scan rate of 100 mV s–1.

Scheme 2

Electrochemical Redox Process of REG by ZrO2/rGO/GCE

Influence of the Scan Rate

Figure a shows the influence of the scan rate, from 10 to 100 mV s–1, on the cyclic voltammetry (CV) peak potential and current of 0.01 mM REG in the presence of 0.1 M PBS, pH 7.0, at the ZrO2/rGO/GCE. The REG reversible couple peak current increases gradually with an increase in the scan rate. In addition, the REG oxidation and reduction peak currents (Ipa and Ipc) showed good linear correlation coefficients (R2) as a function of the square root of the scan rates of the anodic and cathodic peaks with 10–100 mV s–1 changing scan rates (Figure b), obeying the following linear regression equationsThese results indicate that the electrochemical reactions of REG on the ZrO2/rGO/GCE are diffusion-controlled processes. The estimation of the REG electrochemical parameters, at different scan rates, was made with Laviron’s eqs –3.[44]where α is the electron transfer coefficient (0.76), n is the number of electrons, F is the Faraday constant (96 485 C mol–1), R and T are the universal gas constant and temperature (K), respectively, and ks is the standard heterogeneous rate constant (1.18) determining the slowest step of the REG electrochemical oxidation.

Figure 4

(a) Cyclic voltammograms recorded at the ZrO2/rGO/GCE in the electrolyte solution at different scan rates from 10 to 100 mV s–1. (b) Calibration plot of the anodic and cathode peak currents versus the square root of the scan rate, during the electro-oxidation of 0.01 mM REG in the presence of 0.1 M PBS, pH 7.0.

Analytical Performance of ZrO2/rGO/GCE for REG Detection

The ZrO2/rGO/GCE was tested by differential pulse voltammetry (DPV), to investigate the sensitivity of its response to various REG concentrations in the linear dynamic range of 11–343 nM REG in 0.1 M PBS, pH 7.0, at pulse height 60, pulse width 10, and scan rate 100 mV s–1, as shown in Figure a. The intensity of the anodic peak current increased with the REG concentration. Figure b shows the linear dynamic-range plotting of the anodic peak current (Ipa) versus the REG concentrations and its linear regression equation Ipa = 9.4911(REG) + 3.434, (R2 = 0.9963). The limits of detection and quantification can be calculated according to eqs and 5(45,46)where SD is the standard deviation and B is the slope of the calibration plot. From the calibration plot, a detection limit of 11 nM and quantification limit of 59 nM were calculated, on the basis of S/N = 3 (signal to noise). These results confirmed that the ZrO2/rGO/GCE is a promising platform for the electrochemical determination of ultratrace of REG concentration.

Figure 5

(a) Differential pulse voltammograms recorded at the ZrO2/rGO/GCE over an REG concentration of 11–343 nM in 0.1 M PBS at pH 7.0. (b) Linear calibration plot of the anodic peak current versus REG concentration.

Simultaneous Detection of REG, AA, and UA by ZrO2/rGO/GCE

The main objective of this study was sensing of REG, AA, and UA in a mixture. REG, AA, and UA mixture solution with different concentrations in 0.1 M PBS with pH 7 was prepared. DPV result for the simultaneous detection of REG, AA, and UA is presented in Figure . The oxidation peak current increased synchronously on increasing the concentration of REG, AA, and UA. The DPV signal shows the linear relationship between the oxidation peak current and REG concentrations (Figure inset) with LDR 0.32–0.66 μM and the linear regression equation as follows: Ipa (μA) = 15.46CREG + 0.5232 (μM) (R2 = 0.9943). The oxidation current increased parallel AA and UA concentrations with LDRs 0.58–1.26 and 0.08–0.52 μM, respectively, and the linear regression equations as follows: Ipa (μA) = 3.92CAA + 2.417 (μM) (R2 = 0.9749) and Ipa (μA) = 15.85CUA + 4.97 (μM) (R2 = 0.9795). This result indicated that the proposed electrochemical sensor enables the synergetic and sensitive detection of REG in the presence of AA and UA without significant interference from each other.

Figure 6

DPVs recorded on the ZrO2/rGO/GCE during simultaneous determination of 0.32–0.66 μM REG, 0.58–1.26 μM AA, and 0.08–0.52 μM UA in 0.1 M PBS, pH 7.0. Insets: plots of the anodic peak currents against concentrations of REG, AA, and UA.

Effect of Interference Compounds

The effect of interferences for the determination of REG, AA, and UA mixture solution was investigated in 0.1 M PBS (pH 7.0) electrolyte solution in the presence of a possible interfering factor such as metal ions (Ca2+, Mg2+, and Zn2+), glutathione, folic acid, or l-cysteine. The observed results are summarized in Table S1. No significant signal intensity change (less than 5% difference from their original signal intensity) was observed in the presence of interfering ions and molecules. The results indicate that the designed ZrO2/rGO/GCE displays good ability for simultaneous sensing of AA, UA, and REG in a matrix mixture, without any interference of the above-mentioned species.

Real-Sample Analysis

The real-sample monitoring of the performance of the ZrO2/rGO/GCE was validated by the DPV determination of REG in human blood serum samples. To determine the accuracy of the results, 1.0 mL human blood serum samples were diluted 50 times with PBS to prevent the matrix effects of analytical determinations. The human blood serum samples were centrifuged before the measurements; the results are summarized in Table . The recovery rates for the different volumes of the samples ranged between 96.5 and 101.9% and the relative standard deviations were in the range 0.3–2.4%, showing the accuracy and efficiency of the constructed electrochemical sensor. Therefore, the ZrO2/rGO/GCE can be applied to real bioclinical samples.

Table 1

Real-Sample Analysis of REG Using the Proposed Method by Triplicate (n = 3) Readings

samples	spiked sample (mM)	found (mM)	recovery (%)	RSD (%) ±SE
pharmaceutical formulation	0.010	0.0098	98	2.54 ± 0.05
	0.050	0.0509	101.8	1.98 ± 0.08
	0.100	0.0990	99	1.32 ± 0.06
blood serum	0.010	0.0098	98	2.42 ± 0.11
	0.050	0.0486	97.2	1.65 ± 0.07
	0.100	0.1026	102.6	1.09 ± 0.05

Comparison with Other Established Methods

The sensitivity of the developed electroanalytical method was compared to that of some of the chromatography and spectrophotometry methods. Recently, Fujita et al., reported one method using high-performance liquid chromatography (HPLC) for simultaneous quantitative determination of REG in the human plasma and achieved LOQ of 10 ng mL–1. In addition, Erp et al., reported liquid chromatography tandem mass spectrometry (LC-MS/MS) can achieve LOQ of 100 ng mL–1 for REG. The other spectroscopic method can analyze REG with an LOQ of 290 ng mL–1. Compared with the reported methods, the presented method shows a better limit of detection and limit of quantification limit (Table ). Due to its low cost, simplicity, high sensitivity, and rapid analysis time, the presented method has advantages over the other analytical methods.

Table 2

Comparison of the Electroanalytical Method with Reported Analytical Techniques

method	LDR (ng mL–1)	LOD (ng mL–1)	LOQ (ng mL–1)	refs
HPLC	10–10 000		10.0	(13)
LC-MS/MS	25–25 000		25.0	(14)
LC-MS/MS	100–4000		100	(15)
spectrophotometric method	500–25 000	110	290	(20)
electroanalytical method	1.5–107 (11–343 nM)	5.00 (17 nM)	18.5 (59 nM)	present work

Conclusions

Inorganic metal-oxide nanoparticles have been enormously employed as electrode material for a developed efficient electrochemical sensor. A simple one-pot hydrothermal synthesis of ZrO2/rGO nanocomposite was successfully synthesized. The pristine ZrO2 nanoparticles are about 7 nm and uniformly dispersed on the reduced graphene oxide sheet. Due to reduced graphene oxide support, the ZrO2 metal-oxide nanoparticles illustrate excellent electrocatalytic performance toward REG anticancer drug. The developed ZrO2/rGO nanocomposite was characterized by TEM, EDX, PXRD, FT-IR, and XPS measurements. TEM images clearly show the zirconia nanoparticles have uniform size and uniform distribution on the surface of rGO. The inorganic nanocomposite-based electrochemical sensor was successfully applied first time for the detection of REG anticancer drug. The fabricated ZrO2/rGO nanocomposite exhibited an excellent electrocatalytic activity toward REG, with a linear dynamic range of 11–343 nM and the detection limit as low as 17 nM toward the detection of REG drug. The ZrO2/rGO/GCE was applicable for the joint determination of REG and a commonly reported interference AA and UA at pH 7.0. The calculated LOD/LOQ values by the present electrochemical method are better than those from chromatography and spectrophotometry methods. This electrochemical sensor shows promise for future exploration of rapid detection of other anticancer drugs in human blood serum and pharmaceutical formulation.

Experimental Section

Modification of the Glassy Carbon Electrode

Preparation of the ZrO2/rGO/GCE: prior to use, the bare GCE was polished with alumina powder (1, 0.3, and 0.05 μm) and washed with an ethanol solution, followed by Millipore water under ultrasonication. For the preparation of the electrochemical sensor, 3 mg of the prepared ZrO2/rGO was dispersed in 3 mL of Millipore water with 0.3 mL of Nafion solution and then ultrasonicated for 30 min until a uniformly dispersed ink was obtained. The ZrO2/rGO ink (7 μL) was drop cast onto the clean GCE surface and allowed to dry for 15 min at room temperature. The ZrO2/GCE was fabricated in a similar manner.

Reproducibility and Stability of the Modified Electrode

Reproducibility and stability are key elements for electrode performance. To evaluate the reproducibility of the ZrO2/rGO/GCE, we made five sensing electrodes and used them to investigate their CV current response on 1 μM REG in phosphate buffered saline (PBS), pH 7.0, as shown in Figure S1a. The calibrated histograms in Figure S1b show a relatively standard deviation (RSD) of 0.54%. The repeatability of the modified electrochemical sensor values was obtained for the detection of 1 μM REG in presence of the supporting electrolyte, PBS (pH 7.0). The ZrO2/rGO/GCE average voltammetric response for seven successive determinations was 3.34% (Figure S2a). Moreover, the stability of the ZrO2/rGO/GCE was verified by the daily detection for 7 weeks of 1 μM REG solution in presence of the supporting PBS electrolyte (pH 7.0). After each test of stability, the electrode was washed with deionized water, dried under an argon stream, and kept in empty glass tubes at room temperature. The electrochemical oxidation of the 1 μM REG solution, in presence of the supporting PBS electrolyte (pH 7.0), using the ZrO2/rGO/GCE diminished by about 9.6% of their initial response during the 7 weeks, as shown in Figure S2b. Hence, the proposed method and the modified electrochemical sensor determined REG with higher reproducibility and stability than the ZrO2/GCE sensor.

Preparation of Real Samples for Analysis

A powdered Stivarga tablet (Nexus Lifecare Pvt. Ltd., Mumbai, India) containing 40 mg of REG was dissolved by ultrasonication in 25 mL of 0.1 M PBS buffer solution at pH 7.0. This solution was filtered and quantitatively diluted with buffer solution to get 0.1 mM REG solution that was used for the analyses. Fresh human blood serum samples were collected from healthy volunteers (S. V. University Health Center, S. V. University, Tirupati, India). Approximately, 2.0 mL of human blood serum was diluted with 100 mL of 0.1 M PBS, at pH 7.0, and the solution thus prepared was used for analysis, without any further treatment.