Direct electrochemistry of cholesterol oxidase and biosensing of cholesterol based on PSS/polymeric ionic liquid–graphene nanocomposite

Abstract

A novel graphene nanocomposite, functionalized by polymeric ionic liquids (PILs) and poly(sodium-p-styrenesulfonate) (PSS), was successfully prepared and exhibited excellent conductivity, favourable biocompatibility and good film-forming properties as an electrode material. TEM showed that the nanocomposite possessed an individual nanosheet-like structure. Owing to the surface modification of graphene, PSS/PILs–GP can not only be well dispersed in aqueous solution, but also possesses a strong negative charge. Due to electrostatic interactions, positively charged cholesterol oxidase (ChOx) can be immobilized onto the surface of PSS/PILs–GP to form the ChOx/PSS/PILs–GP/GC electrode material. UV-vis and FT-IR spectroscopy were used to monitor the assembly process of the nanocomposite. Due to the conductivity and biocompatibility of PSS/PILs–GP, the immobilized ChOx exhibited enhanced direct electron transfer (DET) at a glassy carbon (GC) electrode. Furthermore, the ChOx/PSS/PILs–GP/GC electrode displayed excellent catalytic performance with a wide linear range of 10.5 × 10⁻⁶ to 10.4 × 10⁻³ mol L⁻¹, and a low detection limit of 3.5 μmol L⁻¹ for the detection of cholesterol.

---

1. Introduction

Cholesterol, one of the most abundant steroids in the body, is essential for maintaining a normal metabolism in humans.¹ High cholesterol levels in the blood are related to heart disease, diabetes, atherosclerosis, nephrotic syndrome, liver disease and other complications due to hepatobiliary diseases.^2–4 The determination of cholesterol concentration in serum has become a very important reference index for clinical diagnosis. The traditional determination method of cholesterol in serum is based on spectrophotometry, involving complicated and detailed experimental procedures, slow testing speeds, large sample consumption and so on.⁵ Therefore, developing a highly sensitive and selective cholesterol detection method is of particular significance. Electrochemical methods based on cholesterol oxidase (ChOx) have attracted more and more attention in recent years, having advantages of sensitive, rapid and simple determination of cholesterol.^6–8 In order to achieve high performance of an electrochemical biosensor, it is particularly important to select an appropriate support matrix, which can provide the right environment to preserve the biology and catalytic activity of an enzyme, and allow electron transfer between the enzyme and electrode.

Graphene (GP), a two-dimensional carbon based nanomaterial, has sparked great interest in various application fields.^9–12 In recent years, it has also been found that graphene is an outstanding material for the fabrication of electrochemical biosensors because of its unique properties, such as a large accessible surface area and excellent electrical conductivity.^13,14 For instance, it has been found that the direct electron transfer (DET) reaction of redox enzymes can be greatly enhanced via the utilization of graphene as the electrode material,^15–17 and due to the achieved DET, the performance of the as-fabricated electrochemical biosensors can be greatly improved.¹⁸ However, pure graphene exhibits poor solubility in aqueous solution, and the hydrophobicity of graphene prevents it from interacting with and immobilizing water soluble redox enzymes. Moreover, without functionalization, graphene tends to aggregate or indeed re-stack to form graphite in aqueous solutions owing to the strong van der Waals energy and π–π interactions. Such aggregation causes graphene to lose its individual sheet-like structure, and reduces the surface area of the electrode material, thus further affecting the performance of the biosensor. Therefore, the utilization of water soluble, individual graphene nanosheets for the immobilization of redox proteins on an electrode surface is important for the fabrication of electrochemical biosensors.

A variety of materials have been functionalized onto graphene to generate hydrophilic graphene nanocomposites, such as surfactants,¹⁹ metal nanoparticles,²⁰ ionic liquids²¹ and polymers,²² etc. In particular, polymeric ionic liquids (PILs) and polyelectrolytes, which possess excellent binding capabilities, high ionic conductivity and electrochemical stability, have been widely used to modify graphene to improve the individual dispersibility and increase the electron transport performance of graphene. For instance, Yang et al. developed a polyelectrolyte-functionalized ionic liquid (PFIL) incorporated into a sol–gel organic–inorganic hybrid material (PFIL/sol–gel), which was used to immobilize glucose oxidase (GOD) on a glassy carbon (GC) electrode, which exhibited an enhanced current response towards glucose.²³ In our previous work, the electrode obtained by immobilizing GOD on a PIL–GP (polymerized ionic liquid–graphene) nanocomposite film was found to have a very wide linear range and good stability towards glucose determination.²⁴ Liu et al. prepared graphene nanosheets functionalized with poly(diallyldimethylammonium chloride) (PDDA-G), which were combined with a room temperature ionic liquid (RTIL). The resulting RTIL/PDDA-G composite displayed an enhanced capability for the immobilization of hemoglobin to realize its direct electrochemistry and excellent electrocatalytic activity for the detection of nitrate with a wide linear range.²² Hong et al. prepared composite films of polystyrenesulfonate doped poly(3,4-ethylenedioxythiophene) (PEDOT-PSS) and graphene. The composite films possessed higher conductivity and much better film-forming ability than graphene, and were used as the counter electrodes of dye-sensitized solar cells with improved performance.²⁵ Moreover, it has been reported that the combination of ionic liquids with polyelectrolytes or conducting polymers could produce excellent conductivity and synergistic effects. Cheng et al. synthesised new polymer electrolytes containing 1-butyl-4-methylpyridinium bis(trifluoromethanesulfonyl)imide (BMPyTFSI) ionic liquid, and the addition of BMPyTFSI to the polyelectrolyte resulted in an increase of the ionic conductivity.²⁶ Therefore, the polyelectrolytes, when composited with PILs instead of their counter anions, are expected to endow graphene with improved conductivity for an increased electrochemical response, increased solubility in aqueous solution and good film-forming properties for the modification of electrodes.

In this work, a novel functionalized graphene nanocomposite, PSS/PILs–GP nanocomposite, was prepared by a self-assembly process with PILs and PSS via π–π interaction and electrostatic forces. Based on the electrostatic interaction between the negatively charged PSS/PILs–GP and positively charged cholesterol oxidase (ChOx), the redox enzyme could be immobilized onto PSS/PILs–GP under mild conditions to fabricate a ChOx/PSS/PILs–GP/GC electrode. The immobilized ChOx not only showed direct electrochemistry on the GC electrode, but also exhibited favourable catalytic performance for cholesterol, owing to the excellent conductivity and biocompatibility of the hydrophilic graphene nanocomposite.

2. Chemicals and methods

2.1. Chemicals

Cholesterol oxidase (ChOx, EC 1.1.3.6, 500U) and cholesterol esterase (ChEt, EC 3.1.1.13, 500U) were purchased from Sigma Chemical Co. Graphite powder (99.95%, 325 mesh), and hydrazine hydrate solution (85 wt%) were purchased from Sinopharm Chemical Reagent Co., Ltd. PSS (M_w = 70 000), cholesterol, Triton X-100, isopropanol, ascorbic acid (AA), glucose, uric acid (UA) and lactic acid (LA) were obtained from Alfa Aesar and used as received. 1-Vinyl-3-ethylimidazolebromide (ViEtIm⁺Br⁻) was synthesized according to the published literature.²⁷

A stock solution of 0.1 mmol L⁻¹ cholesterol was prepared in 0.1 mol L⁻¹ PBS (pH 6.5) containing 0.5% (v/v) isopropanol and 0.2% (v/v) Triton X-100 in a bath at 60 °C. The cholesterol solution was stored at 4 °C when not in use.

2.2. Preparation of graphene oxide

Graphene oxide (GO) was synthesized from graphite powder by the Hummers method.²⁸ In brief, a mixture of 2 g powered flake graphite and 1.6 g NaNO₃ were added to 67.5 mL of 98% H₂SO₄ in an ice bath. Afterwards, 9 g KMnO₄ was gently added with stirring. The mixture was then held at 35 °C for 30 min. Then, the compound was kept at room temperature with mechanical stirring for 5 days, following which 560 mL of warm water was gradually stirred into the paste, which was further treated with 30% H₂O₂; the color of the suspension turned into brilliant yellow. The mixture was then centrifuged, washed with distilled water repeatedly and dried under ambient conditions overnight to obtain GO.

2.3. Preparation of PSS/PILs–GP nanocomposite

The PILs–GP nanocomposite was synthesized according to the reported method with a slight modification.²⁹ Firstly, GO (50 mg) was dispersed in a 100 mL aqueous solution containing PILs that were obtained by polymerizing ViEtIm⁺Br⁻ with initiator 2,2-azobisisobutyronitrile. The suspension was homogenized for 1 h using a magnetic stirrer and then sonicated for 20 min. After that, the pH of the solution was regulated to 10 with 1 mol L⁻¹ aqueous NaOH. The resulting GO dispersion was then reduced with 0.53 mL 85% hydrazine hydrate at 100 °C for 12 h. Then, the black poly(ViEtIm⁺Br⁻)–GP (PILs–GP) suspension was obtained by centrifuging (14 000 rpm) for 30 min, and washed with deionized water three times to remove excess hydrazine hydrate. Subsequently, the compound was dispersed in deionized water to prepare a 1 mg mL⁻¹ dispersion solution. Finally, 10 mL of 1 mg mL⁻¹ PILs–GP was mixed with 20 mL PSS and sonicated for 4 h to form a homogenous dispersion. The mixture was centrifuged (14 000 rpm) for 30 min and washed with deionized water three times to obtain PSS/PILs–GP.

2.4. Preparation of ChOx/PSS/PILs–GP/GC electrode

Prior to modification, the bare glassy carbon (GC) electrode was carefully polished with 1.0, 0.3 and 0.05 μm alumina powder consecutively. Then, it was thoroughly rinsed with deionized water, followed by sonication in acetone, ethanol and deionized water successively. To prepare the ChOx/PSS/PILs–GP/GC electrode, 5 μL PSS/PILs–GP dispersion (1 mg mL⁻¹) was dropped onto the surface of the pretreated GC electrode. A beaker was placed over the electrode so that the water could evaporate slowly in air, forming a uniform film on the electrode. Then the modified electrode was soaked in 0.1 mol L⁻¹ phosphate buffer solution (PBS) containing 5 mg mL⁻¹ ChOx (pH 4.0) to construct the ChOx/PSS/PILs–GP/GC electrode. The ChOx/PILs–GP/GC electrode was fabricated by immersion in a 5 mg mL⁻¹ ChOx (pH 8.0) solution with negative charge. The as-prepared enzyme electrodes were stored at 4 °C in refrigerator for 24 h.

For comparison, PSS/PILs–GP/GC was prepared by the same procedures as described above.

2.5. Apparatus and measurements

All electrochemical experiments were performed at room temperature with a CHI660E workstation (Shanghai Chenhua Instruments Co., China). A three-electrode system was used in the measurements, with a modified electrode (3 mm in diameter) as the working electrode, a platinum wire as the counter electrode, and a saturated Ag/AgCl electrode as the auxiliary electrode. Cyclic voltammetric measurements were performed in buffer solution degassed with highly purified nitrogen for at least 30 min, and a nitrogen atmosphere was maintained throughout the electrochemical measurements without special illustration. Differential pulse voltammetric (DPV) measurements were carried out from −0.1 to −0.7 V in 10 mL of 0.1 mol L⁻¹ air-saturated PBS (pH 6.5). Impedance measurements were performed in the frequency range of 0.1 Hz to 100 kHz using an ac voltage of 5 mV in 5.0 mmol L⁻¹ K₃[Fe(CN)₆]/K₄[Fe(CN)₆] solution containing 0.1 mol L⁻¹ KCl.

UV-vis experiments were carried out using a Lambda 35 ultraviolet/visible spectrophotometer (Perkin Elmer Corp., USA). Fourier transform infrared (FT-IR) spectra of the samples in KBr pellets were recorded on a Perkin Elmer Spectrum One instrument. X-ray diffraction (XRD) patterns were recorded with a Bruker D8-ADVANCE diffractometer. The morphologies of PSS/PILs–GP were observed using a Hitachi model H-800 transmission electron microscope (TEM) operated at an accelerating voltage of 100 kV, and a Hitachi SU-8010 scanning electron microscope (SEM) equipped with an energy-dispersive X-ray spectroscopy (EDX) analyzer. Zeta potential data were obtained using a Zetasizer Nano-ZS particle analyser (Malvern Corp., England).

3. Results and discussion

3.1. Characterization of PSS/PILs–GP nanocomposite

It is known that most of the specific properties of graphene are dependent on its individual sheet structure, so preserving the configuration of graphene is of great importance for the application of graphene in biosensors.³⁰ The morphology of PSS/PILs–GP was characterized by transmission electron microscopy (TEM). As shown in Fig. 1A, PSS/PILs–GP nanocomposite exhibits a similar appearance to graphene nanoplatelets. It can be observed that PSS/PILs–GP was present in the form of flake-like individual sheets, ranging from 500 nm to 1 μm in size. The XRD patterns of graphite (a), GO (b), and PSS/PILs–GP (c) are shown in Fig. 1B. As opposed to the strong peak of graphite at 26.5°, a new peak appeared at 11.3° in (b), corresponding to the (001) reflection in the spectrum of GO,³¹ confirming that the graphite had been greatly oxidized. After the reduction of GO to graphene by hydrazine, a broad peak centered at 24.1° appeared, which represented the complete reduction of GO. The photographs of GO (left), unmodified GP (middle) and PSS/PILs–GP (right) dispersed in water are shown in the inset of Fig. 1B. The reduction of GO was also proved by the change in color from brown to black. It is notable that the unfunctionalized graphene aggregated readily in water. However, PSS/PILs–GP could be easily dispersed in water to form a homogeneous black suspension, illustrating that the nanocomposite was extremely hydrophilic. Fig. 1C and D present the zeta potential data of PILs–GP and PSS/PILs–GP in PBS (pH 6.5). It can be observed that PILs–GP displayed an obvious positive charge of 43.2 mV, and PSS/PILs–GP possessed a negative charge of −32.5 mV at pH 6.5. This negative charge can not only significantly stabilize the dispersion of PSS/PILs–GP in aqueous solution due to the electrostatic repulsion between the nanosheets, but also provides the driving force for further enzymatic immobilization. The chemical compositions of the prepared PILs–GP and PSS/PILs–GP nanocomposites were determined by EDX. The elements C, O, N and Br were detected in the EDX spectrum of the PILs–GP hybrid film on a silicon chip (Fig. 1E), where the N and Br peaks originated from the imidazolium of the PIL and its counter ion, and the C and O peaks arise from the reduced graphene oxide (GP). In the spectrum of PSS/PILs–GP (Fig. 1F), besides the aforementioned peaks, an additional peak from the element S, derived from PSS, was observed. The sharp decrease of the Br peak in the spectrum of PSS/PILs–GP, compared to that in the EDX spectrum of PILs–GP, further proves that Br⁻ counter ions in PILs–GP have been replaced by negatively charged PSS.

Fig. 1 (A) TEM image of the PSS/PILs–GP nanocomposite. (B) XRD patterns of graphite (a), GO (b), and PSS/PILs–GP (c). Inset: photos of GO (left), unmodified graphene (middle), and PSS/PILs–GP (right) dispersed in water. (C and D) Zeta potential data of PILs–GP and PSS/PILs–GP in pH 6.5 PBS. EDX spectra of (E) PILs–GP and (F) PSS/PILs–GP nanocomposite.

3.2. UV-vis and FT-IR spectroscopic analysis of PSS/PILs–GP nanocomposite

UV-vis absorption spectroscopy was used to evaluate the formation of the PSS/PILs–GP nanocomposite. Fig. 2A shows the UV-vis spectra of GO (a), PILs–GP (b), and PSS/PILs–GP (c) in aqueous solution. The GO suspension displayed an absorption peak centered at 230 nm and a shoulder peak at 302 nm (curve a), attributed to the π–π transitions of aromatic C–C bonds and the n–π transitions of C=O bonds.³² However, after the reduction by hydrazine hydrate, the typical absorption peak of GO shifted from 231 nm to 271 nm, indicating the formation of graphene nanosheets (curve b). Furthermore, due to the modification with PSS through electrostatic interactions, a new absorption peak at 225 nm (curve c) can also be observed in the spectrum of PSS/PILs–GP (curve c), ascribed to the π–π* electron transition of PSS.

Fig. 2 (A) UV-vis absorption spectra of GO (a), PILs–GP (b) and PSS/PILs–GP (c) dispersions. (B) FT-IR spectra of GO (a), PILs–GP (b) and PSS/PILs–GP (c).

Fig. 2B shows the FT-IR spectra of GO (a), PILs–GP (b), and PSS/PILs–GP (c). As shown in curve (a), the spectrum of GO exhibited C=O stretching absorption bands at 1726 cm⁻¹, a strong C=C stretching peak at 1632 cm⁻¹, a C–O carboxyl absorption peak at 1392 cm⁻¹, and the C–O stretching band at 1121 cm⁻¹. It can be observed that the FT-IR of PILs–GP (curve b) was quite different from that of GO, with the disappearance of the oxide groups, indicating the efficient reduction of GO. After the modification of PSS onto PILs–GP, new absorption peaks located at 1040, 1012, 830 and 678 cm⁻¹ appeared in the FT-IR spectrum of the PSS/PILs–GP nanocomposite (curve c), which can be ascribed to the characteristic bands of PSS.³³

3.3. Fabrication of ChOx/PSS/PILs–GP/GC electrode

The multi-step self-assembly method for the preparation of the ChOx/PSS/PILs–GP/GC electrode is shown in Scheme 1. Firstly, GO was synthesized by the Hummers method, and was then reduced by hydrazine in the presence of PILs. Owing to the strong π–π stacking interactions between the imidazolium ion moieties and graphene,²⁹ the imidazolium-based PIL was able to attach to the surface of graphene, resulting in the formation of PILs–GP during the chemical reduction procedure. PSS could be further functionalized onto the PILs–GP through electrostatic interactions, forming a PSS/PILs–GP nanocomposite with a layer of negative charge generated on the surface of PILs–GP, which prevented the agglomeration of the graphene in water to maintain its good conductivity. The effect of PSS on PILs–GP not only enhanced the electron transfer capacity of the nanocomposite, but also endowed the nanosheets with negative charge. Meanwhile, the ChOx exhibited positive charge in PBS at pH 4.0. Therefore, when the PSS/PILs–GP/GC electrode was immersed in ChOx at pH 4.0 PBS, the enzyme could be readily immobilized onto PSS/PILs–GP through electrostatic forces, thus forming the ChOx/PSS/PILs–GP/GC electrode. According to the above results, it is obvious that the self-assembly technique played a crucial role during the fabrication of the ChOx/PSS/PILs–GP/GC electrode, not only acting as the driving force for the synthesis of water soluble graphene, but also resulting in the immobilization of the redox enzyme onto the graphene based electrode under mild conditions.

Scheme 1 Schematic explanation of the process of ChOx immobilization on the surface of PSS/PILs–GP, and the direct electrochemistry of ChOx.

3.4. Electrochemical performance of ChOx/PSS/PILs–GP/GC electrode

The modification process of the electrode was characterized by electrochemical impedance spectroscopy (EIS), meanwhile the intensive electrical conductivity of the PSS/PILs–GP nanocomposite was also demonstrated. The EIS technique is a means of detecting interface properties, consisting of a semicircular part and a linear part.³⁴ The diameter of the high-frequency semicircle part is equal to the interfacial electron transfer resistance (R_et) in the impedance spectrum.³⁵ On this basis, the change in impedance value was used to observe changes on the electrode surface. Fig. 3A shows the impedance diagrams of the different modified electrodes. The R_et of the bare GC electrode is 429 Ω (curve a, Fig. 3A). In comparison, the R_et value of the PILs–GP/GC electrode was obviously decreased to 321 Ω, due to its good conductivity (curve b, Fig. 3A), which showed that PILs–GP was successfully self-assembled onto the bare GC electrode surface. The R_et value of PSS/PILs–GP/GC was further decreased to 232 Ω (curve c, Fig. 3A), attributed to the acceleration of the electron transfer between electrode and electrolyte. When the ChOx molecule was adsorbed onto the modified electrode through electrostatic interactions, the R_et value significantly increased to 1429 Ω (curve d, Fig. 3A), which showed that the ChOx was successfully immobilized onto the electrode surface, and the non-conductive nature of the ChOx hindered the electron transfer.

Fig. 3 (A) Nyquist plots of bare GC (a), PILs–GP/GC (b), PSS/PILs–GP/GC (c) and ChOx/PSS/PILs–GP/GC (d) electrodes in 5.0 mmol L⁻¹ Fe(CN)₆^3−/4− solution containing 0.1 mol L⁻¹ KCl in the frequency range from 0.1 to 1 × 10⁵ Hz. (B) Cyclic voltammograms of bare GC (a), PILs–GP/GC (b), PSS/PILs–GP/GC (c) and ChOx/PSS/PILs–GP/GC (d) electrodes in 5.0 mmol L⁻¹ K₃Fe(CN)₆ solution containing 0.1 mol L⁻¹ KCl. Scan rate: 200 mV s⁻¹.

Cyclic voltammetry (CV) was used to monitor the modification process of the electrode in 5 mmol L⁻¹ of ferricyanide solution in the potential range −0.2 to 0.8 V. As shown in Fig. 3B, when the bare GC electrode was used as the working electrode, an obvious redox peak appeared (curve a, Fig. 3B), but it was weaker than that of the PILs–GP/GC electrode (curve b, Fig. 3B). After PSS was modified on the PILs–GP, the oxidation and reduction peak current of the modified electrode increased significantly (curve c, Fig. 3B), indicating that the electron transmission between the surface of the electrode and the electrolyte was accelerated. As shown in curve (d) (Fig. 3B), when ChOx was immobilized on the surface of modified electrode, the current response diminished due to the non-conductive protein layer which hindered electron transfer. The results are consistent with the impedance spectra and further demonstrate the good synergistic effect of PSS/PILs–GP.

The direct electrochemistry of the ChOx/PSS/PILs–GP/GC electrode was investigated by cyclic voltammetry. Fig. 4 shows the cyclic voltammograms (CVs) of PSS/PILs–GP/GC (a), ChOx/PILs–GP/GC (b) and ChOx/PSS/PILs–GP/GC (c) electrodes in the potential range from −0.1 to −0.7 V. No redox peak was observed at the PSS/PILs–GP/GC electrode (curve a), indicating that PSS/PILs–GP was not electroactive in the observed potential range, which is consistent with reports of similar materials.^22,24,36 However, after the immobilization of ChOx, a pair of well-defined and quasi-reversible redox peaks appeared at the ChOx/PSS/PILs–GP/GC electrode (curve c), and furthermore, the response of the peak current is higher than that of the ChOx/PILs–GP/GC (curve b) electrode, which showed that the PSS/PILs–GP nanocomposite was superior to PILs–GP in the direct electron transfer of the FAD/FADH₂ redox couple of the ChOx. The cathodic and anodic peak potentials were found at −0.409 and −0.361 V respectively for the ChOx/PSS/PILs–GP/GC modified electrode. The formal potential (E_p), calculated from the average value of the cathodic and anodic peak potentials, was −0.385 V. The experimental results obtained from the ChOx/PSS/PILs–GP/GC electrode demonstrated that PSS/PILs–GP provided a biocompatible and conductive microenvironment and played an important role in the achievement of enhanced direct electrochemistry of the immobilized ChOx, owing to its hydrophilic surface, excellent electron transfer properties and large surface area. The excellent electron transfer properties of the PSS/PILs–GP nanocomposite, which resulted in the intensive electrochemical response, should be attributed to its higher conductivity than PILs–GP (Fig. 3A). The superior conductivity of the PSS/PILs–GP nanocomposite might be due to the replacement of the small molecular counter ion Br⁻ of PILs–GP with the polymeric counter ion PSS, which is worthy of further investigation.

Fig. 4 Cyclic voltammograms of PSS/PILs–GP/GC (a), ChOx/PILs–GP/GC (b) and ChOx/PSS/PILs–GP/GC (c) electrodes in 0.1 mol L⁻¹ PBS (pH 6.5). Scan rate: 200 mV s⁻¹.

The effect of scan rate on the voltammetric response of the ChOx/PSS/PILs–GP/GC electrode was also investigated. Fig. 5A shows the CVs of the ChOx/PSS/PILs–GP/GC electrode at different scan rates in the range 50 to 800 mV s⁻¹. As shown, upon increasing the scan rate, both the cathodic (I_pc) and anodic (I_pa) peak current increased linearly with the scan rate (Fig. 5B), indicating that a surface-controlled process occurred at the modified electrode. The surface average concentration of electroactive ChOx (Γ*) at the ChOx/PSS/PILs–GP/GC electrode can be calculated from the charge integration of the cathodic peak. According to Faraday’s law, Q = nFAΓ* (where F = 96 485, Q can be obtained by integrating the cathodic peak of ChOx, Γ* is the surface concentration of electroactive ChOx, and n and A represent the number of electrons transferred and the area of the GC electrode, respectively), the electroactive ChOx (Γ*) on the ChOx/PSS/PILs–GP/GC electrode was estimated to be 2.94 × 10⁻¹⁰ mol cm⁻², which was much larger than that in other reports,^37,38 demonstrating that multiple layers of immobilized ChOx entrapped in the PSS/PILs–GP film participated in the electron transfer process. The reason for the multi-layer electron transfer process was possibly due to the good conductivity, large specific surface area and the favorable orientation of the PSS/PILs–GP nanocomposite for the immobilized ChOx.

Fig. 5 (A) Cyclic voltammograms of the ChOx/PSS/PILs–GP/GC electrode at scan rates of 50, 100, 200, 300, 400, 500, 600, 700 and 800 mV s⁻¹ (from a to i) in 0.1 mol L⁻¹ PBS (pH 6.5), respectively. (B) Plots of anodic peak current (■) and cathodic peak current (●) vs. scan rate for the ChOx/PSS/PILs–GP/GC electrode.

As is well known, the direct electron transfer process of ChOx is a two-electron and two-proton transfer process in an electrochemical reaction that can be expressed as follows:³⁷

ChOx(FAD) + 2H⁺ + 2e⁻ ↔ ChOx(FADH₂) (1)

Accordingly, the pH values of the solution influence the electrochemical behavior of ChOx immobilized on the PSS/PILs–GP film. As seen in Fig. 6A, with increasing pH, both the cathodic peak and the anodic peak potentials were negatively shifted. The maximum peak current appeared at pH 6.5. Therefore, pH 6.5 PBS was used in the subsequent electrochemical catalytic test. The relationship between formal potential and pH is shown in Fig. 6B. A clear linear relationship was observed with a slope of 54 mV per pH unit, which is close to the theoretical value (59 mV per pH),³⁹ proving a two-electron and two-proton reaction process.

Fig. 6 (A) Cyclic voltammograms of the ChOx/PSS/PILs–GP/GC electrode in N₂-saturated PBS at pH 4.0, 5.0, 5.7, 6.5, 7.0, 7.4 and 8.0 (from a to g), respectively. Scan rate: 200 mV s⁻¹. (B) Plot of E⁰ vs. pH. (C) Plot of the peak current (I_pc) (C_ChOx: 5.0 mg mL⁻¹) vs. the concentration of 5 μL PSS/PIL–GP dropped on the GC electrode: 0.1, 0.5, 1.0 and 2.0 mg mL⁻¹. (D) Plots of the peak current (●) and Γ* (■) vs. the concentration of ChOx. The PSS/PIL–GP/GC electrode was immersed in different concentrations of ChOx: 1.0, 3.0, 5.0 and 8.0 mg mL⁻¹.

In addition, in order to optimize the performance of the enzyme–nanocomposite modified electrode, the effects of the volume of PSS/PIL–GP dispersion and the concentration of the enzyme used for the fabrication of the electrode on the biosensor performance were also investigated. 5 μL of PSS/PILs–GP dispersion, an optimal volume of the nanocomposite dispersion, was chosen to drop onto the surface of electrode, so that the droplet could completely cover the electrode surface, but couldn’t overflow the bounds of the electrode. When the volume of nanocomposite dispersion dropped onto the electrodes is fixed, the thickness of the film is dependent on the nanocomposite concentration. The influence of PSS/PIL–GP concentration on the current response was investigated. When 5 μL of 1 mg mL⁻¹ PSS/PIL–GP was dropped onto the GC electrode, the cathodic peak current of the ChOx/PSS/PIL–GP/GC electrode was the maximum, as shown in Fig. 6C. The weaker peak currents at concentrations lower than 5 μL of 1 mg mL⁻¹ PSS/PIL–GP might be attributed to an insufficient electrode surface area, while the declining peak currents at concentrations higher than 5 μL of 1 mg mL⁻¹ PSS/PIL–GP might be due to the fact that the PSS/PIL–GP film was too thick to promote the electronic transmission between the enzyme and the GC electrode surface. The thickness of the film at the maximum cathodic peak current was determined to be about 350 nm by atomic force microscopy and scanning electron microscopy (images not shown). The influence of the concentration of ChOx used for the fabrication of the electrode on the current response has also been investigated. As shown in Fig. 6D, at first the cathodic peak current of the ChOx/PSS/PIL–GP/GC electrode and the corresponding Γ* of ChOx increased with the increasing concentration of ChOx, and then reached a maximum value at the concentration of 5 mg mL⁻¹. When the ChOx exceeded this concentration, the cathodic peak current began to slightly decrease, indicating that the adsorption of enzyme reached saturation near this concentration. Based on the results, the concentrations of PSS/PIL–GP and ChOx were selected to be 1 mg mL⁻¹ and 5 mg mL⁻¹ respectively as the optimized conditions for the preparation of the ChOx/PSS/PIL–GP/GC electrode in all electrochemical experiments.

3.5. Electrocatalysis of ChOx/PSS/PILs–GP/GC electrode for cholesterol

In order to study the catalytic activity of ChOx immobilized onto PSS/PILs–GP, the CV and DPV measurements of the ChOx/PSS/PILs–GP/GC electrode were performed with cholesterol. Fig. 7 shows the CVs of ChOx/PSS/PILs–GP/GC electrode in the deoxygenated and air saturated PBS with and without the presence of cholesterol. A pair of obvious redox peaks can be observed in both nitrogen (curve a) and air-saturated (curve c) PBS. Whereas, the reduction peak was larger and the oxidation peak was smaller in air-saturated than the homologous peak in nitrogen-saturated PBS, which demonstrated the dissolved oxygen was involved in the redox reaction of ChOx. When cholesterol was added into air-saturated PBS, reductive peak current decreased (curve b), certifying enzymatic reaction happened between oxidation state of ChOx and cholesterol.

Fig. 7 Cyclic voltammograms of the ChOx/PSS/PILs–GP/GC electrode in N₂-saturated (a), O₂-saturated with 1.0 mmol L⁻¹ cholesterol (final concentration) (b) and O₂-saturated (c) 0.1 mol L⁻¹ PBS at pH 6.5. Scan rate: 200 mV s⁻¹.

The DPVs were carried out to evaluate the catalytic performance of ChOx/PSS/PILs–GP/GC electrode, due to the high sensitivity and resolution of this method. As shown in Fig. 8A, the reduction peak current at about −0.38 V was greatly decreased when cholesterol was added into the PBS. Furthermore, the reduction peak currents further decreased when the concentration of cholesterol in the PBS was increased gradually, suggesting that the immobilized ChOx exhibited excellent catalytic activity toward the reduction of cholesterol. However, no reduction peak was observed at the PSS/PILs–GP/GC electrode in the presence of cholesterol under the same experimental conditions (figure not shown). The catalytic process of the immobilized ChOx to cholesterol can be described as follows:⁴⁰

Fig. 8 (A) DPVs of ChOx/PSS/PILs–GP/GC electrode in 0.1 mol L⁻¹ O₂-saturated PBS (pH 6.5) in the presence of different concentrations of cholesterol: 0, 10.5 × 10⁻⁶, 30.5 × 10⁻⁶, 50.5 × 10⁻⁶, 0.3 × 10⁻³, 0.9 × 10⁻³, 1.6 × 10⁻³, 4.7 × 10⁻³, 6.4 × 10⁻³ mol L⁻¹ (from a to l). (B) Plot of the catalytic current (ΔI) vs. cholesterol concentration for the ChOx/PSS/PILs–GP/GC electrode. Inset: plot of the cholesterol concentration/catalytic current (c/ΔI) vs. cholesterol concentration for the ChOx/PSS/PILs–GP/GC electrode. Amplitude: 50 mV; pulse width: 25 ms.

The relationship between the decrease of catalytic reduction current and the concentration of cholesterol is shown in Fig. 8B. It can be observed that the decrease of cathodic peak currents increased with cholesterol concentration. Meanwhile, cholesterol concentration/catalytic current increased linearly with cholesterol concentration from 10.5 × 10⁻⁶ mol L⁻¹ to 10.4 × 10⁻³ mol L⁻¹ (inset, Fig. 8B). The detection limit was estimated to 3.5 μmol L⁻¹ at a signal-to-noise ratio of 3. This biosensor exhibited a wider linear range and a lower detection limit than other reported cholesterol biosensors based on ChOx (Table 1). According to the Lineweaver–Burk equation (1/I_ss = 1/I_max + K_m/(C)I_max) (I_ss: steady-state response current; I_max: maximum catalytic current under saturated concentration; C: substrate concentration; K_m: apparent Michaelis–Menten constant). The K_m of the ChOx/PSS/PILs–GP/GC electrode was calculated to be 4.38 mmol L⁻¹, illustrating that the ChOx immobilized on the PSS/PILs–GP exhibited a higher affinity for the cholesterol substrate to be detected, in contrast to some other reports (7.17 mmol L⁻¹,⁴¹ 16 mmol L⁻¹ (ref. 42)), and therefore, the modified electrodes showed good electrochemical catalytic performance. The wide linear range and low detection limit of the ChOx/PSS/PILs–GP/GC electrode may be due to the preserved bioactivity of the ChOx immobilized on the conductive and biocompatible PSS/PILs–GP nanocomposite.

Table 1 Comparison of different cholesterol biosensors based on cholesterol oxidase

Electrode	Detection method	Linear range (mmol L^−1)	Detection limit (μmol L^−1)	Reference
a ChOx/multi-walled carbon nanotubes/GC electrode.b (ChEt/ChOx)-functionalized graphene/graphite electrode.c Gold electrode/dithiol/gold nanoparticles/11-mercaptoundecanoicacid/ChOx.d Ti/nanoporous gold/ChOx–horseradish peroxidase–ChEt.e GC electrode/poly(thionine)/ChOx/horseradish peroxidase.f Nafion/ChOx/gold nanoparticles–multi-walled carbon nanotubes/GC electrode.g ChOx/chitosan–graphene/GC electrode.
ChOx/MWCNTs/GCE^a	Amperometry	0.0468–0.279	46.8	43
(ChEt/ChOx)–FG/Gr electrode^b	Amperometry	0.05–0.30	15.0	44
AuE/dithiol/AuNPs/MUA/ChOx^c	CV	0.04–0.22	34.6	45
Ti/NPAu/ChOx–HRP–ChEt^d	CV	0.97–7.8	—	6
GCE/PTH/ChOx/HRP^e	DPV	0.025–0.125	6.3	46
Nafion/ChOx/GNPs–MWCNTs/GCE^f	DPV	0.01–5.0	4.3	40
ChOx/CS–GR/GCE^g	DPV	0.005–1.0	0.715	38
ChOx/PSS/PILs–GP/GCE	DPV	0.0105–10.4	3.5	Present work

---

3.6. Reproducibility and stability of ChOx/PSS/PILs–GP/GC electrode

The reproducibility and stability of the ChOx/PSS/PILs–GP/GC electrode was investigated. When 0.2 mmol L⁻¹ cholesterol was measured at five different modified electrodes independently, the relative standard deviations (RSD) for the response current were 3.4%. The RSD of six successive determinations was 3.9%, indicating that the ChOx/PSS/PILs–GP/GC electrode possessed good reproducibility. To investigate the stability of the ChOx/PSS/PILs–GP/GC electrode, the modified electrode was successively scanned for 50 cycles. No obvious change in the peak current could be observed. The long term stability of the ChOx/PSS/PILs–GP/GC electrode was also investigated. After the modified electrode was stored in pH 6.5 PBS at 4 °C for about 1 week, the responses to 0.2 mmol L⁻¹ cholesterol were almost unchanged. Even after storage for 1 month, the reduction current of 0.2 mmol L⁻¹ cholesterol was still 91% of its initial value. The good stability may be attributed to the hydrophilic and biocompatible microenvironment of the immobilized ChOx provided by PSS/PILs–GP.

3.7. Interference study and real sample determination

In order to evaluate the selectivity of the enzyme electrode, the influence of several interferents, such as glucose, AA, LA and UA were investigated using the DPV method. As shown in Fig. 9, the experimental results indicated no obvious change in response after the addition of 0.1 mmol L⁻¹ glucose, 0.1 mmol L⁻¹ AA, 0.1 mmol L⁻¹ LA and 0.1 mmol L⁻¹ UA, compared with the apparent response to 1.0 mmol L⁻¹ cholesterol, due to the low working potential decreasing the electroactivity of the interfering substances. The results demonstrated that the modified electrode showed high selectivity for cholesterol.

Fig. 9 DPVs of the ChOx/PSS/PILs–GP/GC electrode in 0.1 mol L⁻¹ O₂-saturated PBS (pH 6.5) in the presence of 0.015 mmol L⁻¹ cholesterol (a), 0.1 mmol L⁻¹ cholesterol (b), 0.1 mmol L⁻¹ AA (c), 0.1 mmol L⁻¹ UA (d), 0.1 mmol L⁻¹ LA (e), 0.1 mmol L⁻¹ glucose (f) and 1 mmol L⁻¹ cholesterol (g).

To assess the potential practicality of the biosensor, the fabricated electrode was applied in the determination of cholesterol in serum. Cholesterol mainly exists as free cholesterol and cholesterol ester in serum.⁴⁷ The total cholesterol content in serum is the sum of the free and esterified cholesterol. Accordingly, before the detection, 0.5 mg mL⁻¹ cholesterol esterase (ChEt) was added to the serum samples at 37 °C and incubated for 30 min to hydrolyze the esterified cholesterol into free cholesterol. The experimental results showed the determined cholesterol concentrations of the serum samples were consistent with the values evaluated by the spectrophotometric method. A control experiment was also carried on by spiking real serum samples with various concentrations of cholesterol. The results are shown in Table 2. The recoveries were obtained in the range 93–104%. Good recovery results indicated that the biosensor has good performance for testing actual samples. The results indicate that the biosensor has good prospects for application in the determination of cholesterol in serum.

Table 2 Application of the as-prepared biosensor in human serum

Sample	Added cholesterol (mmol L^−1)	Found cholesterol (mmol L^−1)	Recovery (%)	Hospital data (mmol L^−1)
1	—	4.68 ± 0.13	—	4.81
	0.3	4.97 ± 0.12	103.3
	0.5	5.16 ± 0.08	96.0
2	—	3.05 ± 0.09		3.14
	0.3	3.33 ± 0.22	93.3
	0.5	3.53 ± 0.17	96.0

---

4. Conclusions

In summary, hydrophilic and negatively charged PSS/PILs–GP was synthesized via non-covalent modification. This method is advantageous because it is convenient and allows the retention of the intrinsic electronic structure of graphene nanosheets. Through further electrostatic self-assembly, ChOx was immobilized onto PSS/PILs–GP to form a novel enzyme electrode. By taking advantage of the biocompatible and conductive graphene nanocomposite, the ChOx/PSS/PILs–GP/GC electrode can not only exhibit enhanced direct electron transfer of the immobilized ChOx, but also displays excellent catalytic performance for cholesterol with a wide linear range from 10.5 × 10⁻⁶ to 10.4 × 10⁻³ mol L⁻¹, a low detection limit of 3.5 μmol L⁻¹, high stability and selectivity. Therefore, the novel PSS/PILs–GP nanocomposite can be applied as a promising platform for enzymatic immobilization and electrochemical biosensing.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 51273087 and 51203072).

Notes and references

1. S. K. Arya, M. Datta and B. D. Malhotra, Biosens. Bioelectron., 2008, 23, 1083.
2. V. D. G. Shumyantseva, T. Bulko, S. Carrara, C. Nicolini and S. A. A. Usanov, Biosens. Bioelectron., 2004, 19, 971.
3. S. Singh, R. Singhal and B. D. Malhotra, Anal. Chim. Acta, 2007, 582, 335.
4. M. Sungano and A. C. Beynen, Dietary proteins, cholesterol metabolism and atherosclerosis, CAB Direct, Oxfordshire, 1990, vol. 16.
5. A. Gopalan, K. Lee and D. Ragupathy, Biosens. Bioelectron., 2009, 24, 2211.
6. A. Asieh and C. Aicheng, Biosens. Bioelectron., 2011, 26, 4508.
7. R. Manjunatha, G. S. Suresh, J. S. Melo and S. F. D’Souza, J. Electroanal. Chem., 2011, 651, 24.
8. N. Tripathy, R. Ahmad, E. Y. Kim, G. Khang and Y.-B. Hahn, RSC Adv., 2014, 4, 46049.
9. Y. Wang, Y. Shao, D. W. Matson, J. H. Li and Y. H. Lin, ACS Nano, 2010, 4, 1790.
10. S. J. Xu, Y. Liu, T. H. Wang and J. H. Li, Anal. Chem., 2011, 83, 3817.
11. Q. Zeng, J. S. Cheng, L. H. Tang, X. F. Liu, Y. Z. Liu, J. H. Li and J. H. Jiang, Adv. Funct. Mater., 2010, 20, 3366.
12. H. Zhang, X. J. Lv, Y. M. Li, Y. Wang and J. H. Li, ACS Nano, 2010, 4, 380.
13. Y. Wang, Z. Li, D. Hu, C. T. Lin, J. H. Li and Y. H. Lin, J. Am. Chem. Soc., 2010, 132, 9274.
14. Z. Qian, Q. Yun, H. Fei, Z. Ling, W. Shuyao, L. Ying, L. Jinghong and S. Xi Ming, Chem. - Eur. J., 2010, 16, 8133.
15. D. Chen, L. H. Tang and J. H. Li, Chem. Soc. Rev., 2010, 39, 3157.
16. C. Shan, H. Yang, D. Han, Q. Zhang, A. Ivaska and N. Li, Biosens. Bioelectron., 2010, 25, 1070.
17. Q. Zhang, Y. Qiao, L. Zhang, S. Wu, H. Zhou, J. Xu and X. M. Song, Electroanalysis, 2011, 23, 900.
18. B. Thirumalraj, S. Palanisamy, S.-M. Chen, C.-Y. Yang, P. Periakaruppan and B.-S. Lou, RSC Adv., 2015, 5, 77651.
19. J. R. Lomeda, C. D. Doyle, D. V. Kosynkin, H. Wen-Fang and J. M. Tour, J. Am. Chem. Soc., 2008, 130, 16201.
20. K. Zhou, Y. Zhu, X. Yang, L. Jie, C. Li and S. Luan, Electrochim. Acta, 2010, 55, 3055.
21. H. Yang, C. Shan, F. Li, D. Han, Q. Zhang and L. Niu, Chem. Commun., 2009, 45, 3880.
22. K. Liu, J. Zhang, G. Yang, C. Wang and J. J. Zhu, Electrochem. Commun., 2010, 12, 402.
23. F. Yang, L. Jiao, Y. Shen, X. Xu, Y. Zhang and L. Niu, J. Electroanal. Chem., 2007, 608, 78.
24. Z. Qian, W. Shuyao, Z. Ling, L. Jin, V. Francis, L. Yang, X. Zhiqiang, L. Jinghong and S. Xi-Ming, Biosens. Bioelectron., 2011, 26, 2632.
25. W. Hong, Y. Xu, G. Lu, C. Li and G. Shi, Electrochem. Commun., 2008, 10, 1555.
26. C. Hu, C. Zhu, B. Huang, M. Lu and Y. Yang, Electrochim. Acta, 2007, 52, 5789.
27. Z. J. Wang, Q. X. Zhang, D. Kuehner, X. Y. Xu, A. Ivaska and L. Niu, Carbon, 2008, 46, 1687.
28. W. S. Hummers and R. E. Offeman, J. Am. Chem. Soc., 1958, 80, 1339.
29. K. Taeyoung, L. Hyunwook, K. Jongeun and K. S. Suh, ACS Nano, 2010, 4, 1612.
30. C. S. Shan, H. F. Yang, J. F. Song, D. X. Han, A. Ivaska and L. Niu, Anal. Chem., 2009, 81, 2378.
31. J. Balamurugan, T. D. Thanh, S. B. Heo, N. H. Kim and J. H. Lee, Carbon, 2015, 94, 962.
32. Y. Zhou, Q. Bao, L. A. L. Tang, Y. Zhong and K. P. Loh, Chem. Mater., 2009, 21, 2950.
33. W. Xiuming, Z. Bo, W. Ping, Z. Hui and C. Chenxin, J. Phys. Chem. B, 2009, 113, 13365.
34. D. Li, J. Yang, J. Zhou, Q. Wei and F. Huang, RSC Adv., 2014, 4, 61831.
35. X. Kang, J. Wang, H. Wu, I. A. Aksay, J. Liu and Y. Lin, Biosens. Bioelectron., 2009, 25, 901.
36. S. Wu, H. X. Ju and Y. Liu, Adv. Funct. Mater., 2007, 17, 585.
37. X.-R. Li, J.-J. Xu and H.-Y. Chen, Electrochim. Acta, 2011, 56, 9378.
38. Z. Li, C. Xie, J. Wang, A. Meng and F. Zhang, Sens. Actuators, B, 2015, 208, 505.
39. S. Liu and H. Ju, Biosens. Bioelectron., 2003, 19, 177.
40. L. Zhu, L. Xu, L. Tan, H. Tan, S. Yang and S. Yao, Talanta, 2013, 106, 192.
41. M. Guo, J. Chen, J. Li, L. Nie and S. Yao, Electroanalysis, 2004, 16, 1992.
42. B. K. Jena and C. R. Raj, Biosens. Bioelectron., 2011, 26, 2960.
43. J. Y. Yang, Y. Li, S. M. Chen and K. C. Lin, Int. J. Electrochem. Sci., 2011, 6, 2223.
44. R. Manjunatha, G. S. Suresh, J. S. Melo and S. F. D’Souza, Talanta, 2012, 99, 302.
45. U. Saxena, M. Chakraborty and P. Goswami, Biosens. Bioelectron., 2011, 26, 3037.
46. M. M. Rahman, X. B. Li, J. Kim, B. O. Lim, A. J. S. Ahammad and J. J. Lee, Sens. Actuators, B, 2014, 202, 536.
47. E. García-Ruiz, J. C. Vidal, M. T. Aramendía and J. R. Castillo, Electroanalysis, 2004, 16, 497.

---