Download PDF

Analytical and Bioanalytical Chemistry

, Volume 410, Issue 2, pp 543–552 | Cite as

Luminol, horseradish peroxidase, and glucose oxidase ternary functionalized graphene oxide for ultrasensitive glucose sensing

  • Fang Li
  • Wenjing Ma
  • Jiachang Liu
  • Xiang Wu
  • Yan Wang
  • Jianbo He
Research Paper
First Online:
Received:
Revised:
Accepted:

Abstract

Luminol, horseradish peroxidase (HRP), and glucose oxidase (GOx) ternary functionalized graphene oxide (HRP/GOx-luminol-GO) with excellent chemiluminescence (CL) activity and specific enzymatic property was prepared via a simple and general strategy for the first time. In this approach, luminol functionalized GO (luminol-GO) was prepared by gently stirring GO with luminol. Then HRP and GOx were further co-immobilized onto the surface of luminol-GO by storing HRP and GOx with luminol-GO at 4 °C overnight, to form HRP/GOx-luminol-GO bionanocomposites. The synthesized HRP/GOx-luminol-GO could react with H2O2 generated from GOx catalyzed glucose oxidization reaction, to produce strong CL emission in the presence of co-immobilized HRP. Thus, we developed an ultrasensitive, homogeneous, reagentless, selective, and simple CL sensing system for glucose detection. The resulting biosensors exhibited ultra-wide linear range from 5.0 nM to 5.0 mM, and an ultra-low detection limit of 1.2 nM, which was more than 3 orders of magnitude lower than previously reported methods. Furthermore, the sensing system was successfully applied for the detection of glucose in human blood samples.

Keywords

Graphene oxide Horseradish peroxidase Glucose oxidase Glucose Chemiluminescence 

Introduction

Graphene oxide (GO), as a derivative of graphene, was an ideal matrix for constructing novel hybrid materials with a new feature due to its unique characteristics including incredibly large specific surface area, large π–π conjugated structure, abundant oxygen-containing surface functionalities, high water solubility, and special catalytic property [1, 2, 3, 4, 5]. A series of functionalized GO hybrid materials with excellent catalytic activity, unique fluorescence activity, electrochemical activity, and chemiluminescence (CL) property have been successfully synthesized [6, 7, 8, 9, 10]. CL property is of great significance due to its high sensitivity, wide linear detection range, simple equipment, and inexpensive cost. CL functionalized GO were synthesized by the immobilizations of CL reagents including lucigenin, luminol, and N-(aminobutyl)-N-(ethylisoluminol) (ABEI) onto the surface of GO [9, 11]. Recently, in order to further enhance the CL intensity, CL catalysts such as transition metal ions, metal complexes, and enzymes have been further assembled onto the surface of CL reagents functionalized GO. So far, three kinds of CL reagents and catalysts bifunctionalized GO hybrids have been reported, including lucigenin and Co(II) complex bifunctionalized GO [12], ABEI and hemin bifunctionalized GO [13], and ABEI and horseradish peroxidase (HRP) bifunctionalized GO [14]. However, CL functionalized GO hybrids with excellent CL activity, physical, chemical, and biological features are still far from fully developed. Luminol is a kind of common and low-cost CL reagent. The luminol–H2O2 CL reaction is one of the most sensitive CL reaction and still plays an important role in modern chemical and biochemical analysis [15]. So far, luminol and catalyst bifunctionalized GO has not been synthesized. Therefore, the application of CL functionalized GO hybrids for the fabrication of highly sensitive and selective sensors need to be further explored.

Glucose is one of the essential substances for life activities, which plays a critical role in the metabolic process. Several diseases, including hypoglycemia and diabetes, have been associated with an abnormal blood glucose level [16]. Glucose oxidase (GOx) is an oxido-reductase that specifically catalyzes the oxidation of glucose to gluconic acid and H2O2, since it is widely used for the determination of glucose [17, 18, 19]. HRP is a prototypical peroxidase that catalyzes the luminol-H2O2 CL reaction, and has been widely used for the amplified CL detection of glucose [20, 21]. GOx and HRP act in cascade in the glucose oxidation reaction, in which GOx oxidizes glucose to produce H2O2, a substrate for HRP. So far, several bienzymatic electrochemical (EC) and CL glucose sensing methods based on the co-immobilization of GOx and HRP on modified electrodes, membrane, or modified solid substrates have been reported [22, 23, 24, 25]. Although some of the bienzymatic glucose biosensors exhibited low detection limit and wide linear range, most of them were time-consuming and laborious, suffering from complicated immobilization, washing, and operation procedures. Therefore, simple, rapid, homogeneous, ultrasensitive, and selective sensors for the detection of glucose are highly desired.

Recently, some attention has been paid for the immobilization of enzymes onto the surface of nanomaterials [26, 27, 28]. GO was considered to be an ideal platform for the immobilization of enzymes due to its unique irreversible protein absorption ability [29]. The immobilized enzymes on the surface of GO have been reported to possess several advantages over free ones, such as enhanced stability and activity, easy separation and recovery from the reaction mixture, and synergistic catalytic properties [30]. However, the co-immobilization of GOx and HRP onto the surface of homogeneous GO has never been reported. We suppose if luminol, GOx, and HRP are co-immobilized onto the surface of GO sheets, novel multifunctional bionanocomposites with unique CL features and specific enzymatic activity might be obtained.

In this work, luminol, GOx, and HRP ternary functionalized GO (HRP/GOx-luminol-GO) was successfully synthesized using a simple and general strategy. The assembly of HRP/GOx-luminol-GO was studied by high-resolution transmission electron microscopy (HRTEM), atomic force microscopy (AFM), UV–visible (UV-vis) spectrometry, and circular dichroism (CD). The CL property of HRP/GOx-luminol-GO was exploited by a static injection method. It was found that the synthesized HRP/GOx-luminol-GO exhibited stable and strong CL response toward directly added glucose. The analytical performances of the proposed sensing method for glucose detection were examined. The utility of the developed sensing system for human blood samples was also explored.

Experimental section

Chemicals and materials

Luminol was purchased from Sigma-Aldrich (USA) and a 0.01 M stock solution of luminol was prepared by dissolving luminol in 0.1 M NaOH solution. HRP and GOx were purchased from Solarbio Biotechnology Co. Ltd. (Beijing, China). GO was obtained from XFNANO Materials Tech Co., Ltd. (Nanjing, China). Glucose and H2O2 were purchased from Sinopharm Chemical Reagent CO. (Shanghai, China). Working solutions of H2O2 were prepared fresh daily from 30% (v/v) H2O2. All other reagents were of analytical grade. Ultrapure water was used throughout.

Apparatus

The static injection CL detections were conducted on a MPI-A luminescence analyzer (Xi’an, China) with a photomultiplier tube (PMT). UV–vis spectra (Agilent Cary 60, USA), CD spectra (Jasco-810 CD spectrometer, Japan), AFM (Multimode V, Veeco, USA), and HRTEM (JEM-2010, Hitachi, Japan) were used for characterization.

Synthesis of HRP/GOx-luminol-GO

Firstly, luminol functionalized GO (luminol-GO) was synthesized according to the previous report with some modification [14]. Then, 1 mL of 10 mM luminol solution was mixed with 100 mL of 0.1 mg/mL homogeneous GO dispersion and stirred gently at room temperature for 24 h until the color of the mixture turned from light brown to stable dark brown, indicating the immobilization of luminol molecules on the surface of GO. Then, 100 μL 1 mg/mL HRP dissolved in pH 5.6 Britton-Robinson (B-R) buffer and 50 μL 100 U/mL GOx dissolved in pH 4.5 B-R buffer were further added into every 2 mL of the above solution and stored at 4 °C for 9 h. The resulting solution was centrifuged at a speed of 12,500 rpm for 30 min twice to remove residual luminol and enzymes, and the resulting pellet was resuspended in B-R buffer (pH 7.54), to obtain HRP/GOx-luminol-GO nanocomposites. The as-synthesized HRP/GOx-luminol-GO nanocomposites were subsequently characterized by TEM, AFM, UV-vis, and CD spectra. Luminol and GOx co-immobilized GO (GOx-luminol-GO), luminol and HRP co-immobilized GO (HRP-luminol-GO), and luminol and bovine serum albumin co-immobilized GO (BSA-luminol-GO) were assembled with similar procedures by mixing GOx, HRP, and BSA, respectively, with luminol-GO and stored at 4 °C for 9 h.

Quantifying enzymes concentration and activity on HRP/GOx-luminol-GO

The concentration of HRP on HRP/GOx-luminol-GO was quantified by the heme absorption at 403 nm. The absorption was operated on the UV–vis spectra (Agilent Cary 60, USA) with a quartz cuvette of 1 cm path length. The concentration of GOx on HRP/GOx-luminol-GO was quantified by fluorescence assay. The fluorescence was carried out on a fluorescence spectrophotometer (Hitachi F-7200, Japan) with a quartz cuvette of 1 cm path length. Emission was recorded from 300 to 380 nm with the excitation wavelength set at 290 nm.

The activity of HRP and GOx on HRP/GOx-luminol-GO were measured by tetramethylbenzidine (TMB) assay operated on the UV–vis spectra. HRP/GOx-luminol-GO were centrifuged at a speed of 12,500 rpm for 30 min and the resulting pellet was resuspended in 2 mL 0.1 M phosphate buffer (pH 7.2) before the measurements. For measuring the activity of HRP immobilized on HRP/GOx-luminol-GO, 100 μL 0.25 mM TMB solution and different concentrations of H2O2 (0.1–10 mM) were further added, and the activity was measured spectrophotometrically at 652 nm (ε = 39,000 M−1 cm−1). For measuring the activity of GOx, 100 μL 0.25 mM TMB solution and different concentrations of glucose (0.5–50 mM) were further added. Kinetic data were collected for 20 min.

CL measurement

The CL property of HRP/GOx-luminol-GO was exploited by a static injection method on a luminescence analyzer. In a typical CL measurement, 500 μL of the as-prepared HRP/GOx-luminol-GO was pipetted into a cylindrical cell. After injection of 500 μL of 10 mM H2O2 in 0.1 M NaOH, CL emission was recorded immediately.

CL detection of glucose

In a typical glucose assay, 500 μL of the as-prepared HRP/GOx-luminol-GO nanocomposites was pipetted into a cylindrical cell. After injecting 500 μL of glucose with various concentrations in 0.1 M NaOH, CL emission was detected. To investigate the selectivity of the sensing system, various interferents that can be found in biological media, including K+, Na+, ascorbic acid, cysteine, human immunoglobulin G (IgG), goat-anti human IgG, bovine serum albumin, lactose, maltose, and sucrose, were measured instead of glucose.

Results and discussion

Synthesis and characterization of HRP/GOx-luminol-GO

In Scheme 1, luminol molecules with aromatic rings first were attached to the large π-conjugated structure of GO via π–π stacking by mixing GO dispersion with luminol solution. Subsequently, GOx and HRP were further added into the as-prepared luminol-GO aqueous solution so that the enzymes could be self-assembled onto the surface of luminol-GO to form HRP/GOx-luminol-GO bionanocomposites. According to the previous work, the interactions between GOx, HRP, and luminol-GO mainly include electrostatic interaction, hydrophobic interaction, Van der Waals, π–π stacking, and hydrogen bonding interactions [27, 28, 29, 31, 32]. The co-immobilized GOx and HRP could maintain maximum of their bioactivity with this non-covalent binding strategy by avoiding chemical conjugations. The HRP/GOx-luminol-GO bionanocomposites could be stored for at least 2 months without any visible precipitates at 4 °C.
Scheme 1

Schematic illustration for the assembly of HRP/GOx-luminol-GO and fabrication of CL biosensor

The morphology and surface composition of the as-prepared HRP/GOx-luminol-GO were investigated by HRTEM, AFM, UV-vis, and CD. The specimen of HRTEM was prepared by dropping the sample onto a copper net covered by a carbon film. The HRTEM images of luminol-GO and HRP/GOx-luminol-GO are shown in Fig. S1 in the Electronic Supplementary Material (ESM). Both luminol-GO and HRP/GOx-luminol-GO have been shown to be single-layered carbon structures, indicating the good stability of HRP/GOx-luminol-GO. However, wrinkles appeared for luminol-GO (ESM Fig. S1A), while HRP/GOx-luminol-GO (ESM Fig. S1B) dispersed more easily on the carbon film, demonstrating the surface characteristic change of luminol-GO after the co-immobilization of GOx and HRP. According to the AFM results as shown in Fig. 1, the thickness of GO nanosheet is about 0.9 nm (Fig. 1a), demonstrating a single atomic layer thickness structure feature of GO [33]. The thickness of luminol-GO was 1.6 nm (Fig. 1b), 0.7 nm thicker than that of single-layered GO, indicating that monolayer luminol molecules are attached to both sides of GO via π–π stacking, which was in good agreement with the previous work [9, 13]. The thickness of GOx-luminol-GO and HRP-luminol-GO were 2.2 and 2.8 nm (Fig. S2 in the ESM), respectively, which were 1.3 and 1.9 nm thicker than that of single-layered GO, indicating that GOx and HRP could be adsorbed on the surface of GO in a monolayer, respectively [11, 14]. The thickness of HRP/GOx-luminol-GO (Fig. 1c) shows two increments of 1.3 and 1.9 nm, which were in good agreement with the increments when GOx and HRP was immobilized individually on the surface of luminol-GO. The results confirmed that GOx and HRP could be co-immobilized on the surface of luminol-GO in monolayer directly.
Fig. 1

Tapping mode AFM images of a exfoliated GO, b luminol-GO, and c HRP/GOx-luminol-GO

Furthermore, the UV–vis spectra of GO, luminol, GOx, HRP, luminol-GO, and HRP/GOx-luminol-GO were measured, as shown in Fig. 2. The luminol peaks at 300 and 350 nm can be clearly seen in the UV–vis absorbent spectra of luminol-GO and HRP/GOx-luminol-GO, indicating the successful assembly of luminol on the surface of GO. The peak around 200–220 nm is corresponding to the peptide bond in the protein, the peak around 280 nm is corresponding to the amino acid residues in proteins, and the peak 400 nm is corresponding to the heme group of HRP, which were all observed in the absorption spectrum of HRP/GOx-luminol-GO, further confirming the successful assembly of HRP and GOx on the surface of HRP/GOx-luminol-GO.
Fig. 2

UV–vis spectra of GO, luminol, GOx, HRP, luminol-GO, and HRP/GOx-luminol-GO

Generally, most of the biomacromolecules are chiral and exhibit CD absorption band in their CD spectrum. Thus, CD spectrum is a useful probe for monitoring the existence of active enzymes and the possible conformational changes in the protein secondary structure [34]. GO is a kind of two-dimensional material without chirality and does not exhibit CD absorption band in CD spectrum. Thus, the existence of active GOx and HRP on the surface of HRP/GOx-luminol-GO was further demonstrated by CD spectra. The CD spectra of GOx, HRP, luminol-GO, GOx-luminol-GO, and HRP-luminol-GO were measured and compared. As shown in Fig. S3 (see ESM), luminol-GO does not exhibit CD absorption band. In comparison, GOx-luminol-GO exhibited similar optically active band with GOx, while HRP-luminol-GO exhibited similar optically active band with HRP, indicating that GOx and HRP could be assembled onto the surface of luminol-GO, respectively, meanwhile maintaining their enzymatic activities. Then the CD spectra of HRP/GOx-luminol-GO was further compared with the spectra of GOx and HRP. As shown in Fig. 3, the CD spectrum of HRP/GOx-luminol-GO exhibited integrated optically active bands of GOx and HRP, indicating that GOx and HRP were co-immobilized on the surface of luminol-GO successfully. The secondary structures of the co-immobilized GOx and HRP on HRP/GOx-luminol-GO were well maintained, which were important to keep the native activity of the enzymes.
Fig. 3

CD spectra of GOx, HRP, luminol-GO, and HRP/GOx-luminol-GO

Quantifying enzymes concentration and activity on HRP/GOx-luminol-GO

As shown in Fig. 2, HRP heme shows a Soret peak at 403 nm. In comparison, due to the lack of heme, GOx has minimal absorption in this region. Thus, the Soret peak can be used to quantify the concentration of HRP. GOx is tryptophan rich, resulting in strong tryptophan fluorescence peak at 330 nm, while HRP exhibit very weak tryptophan fluorescence as shown in Fig. S4C in the ESM. Thus, tryptophan fluorescence can be used to quantify the GOx concentration [35]. Contribution of HRP to the tryptophan fluorescence emission can be subtracted out based on the HRP concentration determined by heme absorption. Then, the concentrations of HRP and GOx that have been immobilized on HRP/GOx-luminol-GO were determined by subtraction of the initial quantity of enzymes by the concentrations of non-immobilized enzymes in the supernatants after incubation. The concentrations of HRP and GOx in the supernatant were calculated by using Soret absorption and tryptophan fluorescence, respectively. The calibration curves for HRP and GOx were shown in Fig. S4 in the Electronic Supplementary Material. The initial concentration of HRP was 50 μg/mL, whereas the concentration of HRP in the supernatants was calculated to be 16.3 μg/mL. Thus, the concentration of HRP on HRP/GOx-luminol-GO was 33.7 μg/mL. It indicated that the percentage of HRP that have been immobilized on the surface of HRP/GOx-luminol-GO was 67.4%. The initial concentration of GOx was 2.5 U/mL, whereas the concentration of GOx in the supernatants was calculated to be 1.54 U/mL. Thus, the concentration of GOx on HRP/GOx-luminol-GO was 0.96 U/mL. It indicated that 38.4% of HRP have been immobilized on the surface of HRP/GOx-luminol-GO.

The activity of HRP and GOx on HRP/GOx-luminol-GO were measured colorimetrically based on TMB assay. The Michaelis–Menten constant (K M ), reflection of the enzyme-substrate affinity, was calculated from Lineweaver-Burk plots according to the Michaelis–Menten equation: ν = ν max[S] / (K M + [S]), where [S] is the concentration of substrate, ν is rate of reaction, ν max is the maximum rate of reaction under saturated substrate condition (Fig. S5 in the ESM). The enzymatic affinity of HRP was determined by incubating HRP/GOx-luminol-GO with TMB and different concentration of H2O2. K M of HRP on HRP/GOx-luminol-GO was calculated to be 0.695 mM, slightly higher than that for free HRP (K M  = 0.476 mM) [35]. It indicated that the immobilized HRP was able to bind its substrate H2O2, and the binding pocket of immobilized HRP was not perturbed by GO. The GOx activity was determined by incubating HRP/GOx-luminol-GO with TMB and different concentration of glucose. K M of GOx on HRP/GOx-luminol-GO was calculated to be 13.7 mM, comparable with that for free GOx (K M  = 13 mM) [35], indicating the good substrate binding affinity of the immobilized GOx to its substrate glucose, just as well as free GOx.

CL property of HRP/GOx-luminol-GO

Considering the fact that CL reagent luminol, enzymatic catalyst HRP, and GOx were co-immobilized on the surface of GO, HRP/GOx-luminol-GO was supposed to have special CL property. Accordingly, the CL behaviors of various as-prepared composites including GO, luminol-GO, BSA-luminol-GO, GOx-luminol-GO, HRP-luminol-GO, HRP/GOx-luminol-GO, and GOx-luminol-GO mixed with free HRP (the same amount as HRP to prepare HRP/GOx-luminol-GO) reacted with H2O2 were studied and compared. As shown in Fig. 4, it can be seen that no CL emission was observed with GO, and very weak CL signals were observed with luminol-GO (150 a.u.), BSA-luminol-GO (300 a.u.), and GOx-luminol-GO (1260 a.u.), while high CL signals were observed with HRP-luminol-GO (10,500 a.u.), HRP/GOx-luminol-GO (15,000 a.u.), and GOx-luminol-GO mixed with free HRP (4300 a.u.). BSA is a kind of commonly used blocking protein in the fabrication of biosensors. The CL intensity of BSA-luminol-GO was only two times higher than that of luminol-GO, indicating that the immobilized BSA on the surface of luminol-GO had little effect on the CL system. The CL intensities of GOx-luminol-GO was eight times higher than that of luminol-GO, indicating that the immobilized GOx on the surface of luminol-GO had small catalytic effect on the luminol-H2O2 CL system. This may be due to that GOx as an oxido-reductase could facilitate the generation of oxygen-related radicals, accelerating the CL reaction [36, 37]. In comparison, the CL intensities of HRP-luminol-GO was 70 times higher than that of luminol-GO, indicating that the immobilized HRP on the surface of luminol-GO had great catalytic effect on the luminol-H2O2 CL reaction. Furthermore, the CL intensity of HRP/GOx-luminol-GO was much higher than that of GOx-luminol-GO mixed with free HRP, indicating that the catalytic effect of immobilized HRP was much better than that of free HRP. The CL intensity of HRP/GOx-luminol-GO was highest, which was about 2 orders of magnitude higher than that of luminol-GO, indicating that the co-immobilized GOx and HRP on the surface of luminol-GO exhibited superior catalytic effect on the luminol-H2O2 CL reaction.
Fig. 4

CL kinetic curves of GO, luminol-GO, luminol-BSA-GO, GOx-luminol-GO, HRP-luminol-GO, HRP/GOx-luminol-GO, and GOx-luminol-GO mixed with free HRP reacted with H2O2. Insert: histogram of the CL intensities. Reaction conditions: 10 mM H2O2 in 0.1 M NaOH. CL measurement: PMT voltage was set at −250 V

The excellent CL activity of HRP/GOx-luminol-GO might be due to the synergistic catalytic effect of immobilized HRP, GOx, and GO on the CL reaction of luminol-H2O2. HRP has been widely utilized to enhance CL of the luminol-H2O2 system. According to the previous reports, HRP as an oxidoreductase could catalyze the decomposition of H2O2 to form active oxidant compounds, which further reacted with luminol to produce luminol radicals (luminol•–), accelerating the CL reaction [20]. Recent studies demonstrated that GO possesses an intrinsic peroxidase activity [38, 39] and the catalytic efficiency of HRP immobilized on the surface of GO could be improved [26]. Furthermore, GO as nanosized reaction platform could facilitate radical generations of superoxide anion radical (O2 •−) and hydroxyl radical (HO•), and stimulated electron transfer in radical-involved CL reaction, further accelerating the CL reaction [14, 38, 40]. In addition, GOx may facilitate the conversion of H2O2 to HO• in presence of Fe2+ from the heme group of HRP [36, 37]. Therefore, the possible CL mechanism of HRP/GOx-luminol-GO reacted with H2O2 was proposed as follows. Firstly, a complex formed between HRP and H2O2 triggered the oxidation of luminol anions (luminol−) to produce luminol•–. The formed luminol•– could react with the dissolved O2 to generate O2 •–. Meanwhile, GO facilitated the generation of O2 •− and OH• on the surface of GO. GOx facilitated the conversion of H2O2 to HO•. The generated OH• could reacted with HO2 − in alkaline solution to generate more O2 •−. Finally, luminol•− reacted with O2 •− to produce excited-state 3-aminophthalate anion (AP2−*), which emitted a strong CL emission when returning to the ground state. Moreover, since multiple luminol and HRP molecules were co-immobilized on the surface of GO in proximity, the reagent and generated radicals could diffuse to a nearby active site and exchange electrons with each other easily, facilitating the CL reaction rate and improving CL efficiency [23].

CL response of HRP/GOx-luminol-GO to glucose

Luminol and two enzymes, one for H2O2 generation (GOx) and another for H2O2 consumption (HRP), were co-immobilized on the surface of the nanometric GO in proximity. As a result, when glucose in 0.1 M NaOH was directly injected into the HRP/GOx-luminol-GO dispersion solution, strong and stable CL signal was observed. This was due to the fact that glucose could be oxidized by the dissolved O2 with the immobilized GOx as catalyst to yield H2O2. Then the in situ generated H2O2 could oxidize luminol immediately with the surrounded HRP and GO as synergistic catalyst, resulting in strong CL emission. This bienzymatic reaction system confined the GOx catalyzed glucose oxidization reaction and HRP catalyzed luminol-H2O2 CL reaction together on the surface of the nanosized GO, which act in cascade, thereby greatly enhancing the catalytic activity, reaction rate, and amplifying the CL efficiency. The result indicated that HRP/GOx-luminol-GO exhibited both excellent CL activity and specific enzymatic property, which can be used as an ideal platform for the ultrasensitive detection of glucose.

Optimization of CL detection system

The amounts of GOx and HRP, and the ratio of the two enzymes, were key factors in the fabrication of this bienzymatic glucose sensing system. Thus, the CL response at different volume ratios of HRP to GOx was investigated. As shown in Fig. S6 (see ESM), the CL intensity of HRP/GOx-luminol-GO reacted with H2O2 increased with the increasing of the volume ratio of HRP to GOx. This was due to the fact that H2O2 can be consumed more efficiently in the HRP catalyzed CL reaction step with the increasing amount of immobilized HRP. However, the CL response trended to level off when the volume ratio of HRP to GOx was greater than 2:1, probably due to the fact that GO was saturated. Moreover, HRP/GOx-luminol-GO reacted with glucose more rapidly with the increasing of the volume ratio of GOx to HRP. It was due to the fact that more H2O2 could be generated in the GOx catalyzed glucose oxidation reaction step with the increasing amount of immobilized GOx. However, the CL response decreased when the volume ratio of GOx to HRP was greater than 1:2. It was probably due to the fact that the amount of co-immobilized CL catalyst HRP decreased. Thus, the optimized volume ratio of HRP to GOx was 2:1 (100 to 50 μL). The reaction time for the immobilization of GOx and HRP onto the surface of luminol-GO, and pH of the dispersion buffer for HRP/GOx-luminol-GO were also optimized. The experimental results (Fig. S6) showed that the assembly time for GOx and HRP was optimized at 9 h and the CL intensity of HRP/GOx-luminol-GO resuspended in pH 7.54 B-R buffer was highest.

CL detection of glucose

As mentioned above, the synthesized HRP/GOx-luminol-GO nanocomposites exhibited excellent CL activity when directly reacted with glucose. Accordingly, a simple, homogeneous, and reagentless CL biosensor was designed for glucose detection. As shown in Fig. 5a, the CL intensity increased gradually with a larger concentration of glucose, and the logarithm of the CL intensity exhibits a good linear relationship with the logarithm of the glucose concentration over 6 orders of magnitude range from 5.0 nM to 5.0 mM. The linear calibration plot for glucose detection was shown in inset in Fig. 5a. The regression equation was log I = 4.293 + 0.231 × log C with a correlation coefficient of 0.997, where I refers to CL intensity and C the concentration of glucose. The limit of detection (LOD) at a signal-to-noise ratio of 3 (S/N = 3) for glucose was calculated to be 1.2 nM. The integration value of CL curve within 20 s was also used to determine the concentration of glucose. As shown in Fig. S7 (see ESM), the integration values of CL curve show little change with increasing glucose concentrations from 5.0 nM to 5.0 μM, and begin to increase with the increasing of glucose concentration above 5.0 μΜ. Thus, much wider detection range and lower detection limit could be obtained by using the maximal intensity of the CL curve for glucose determination. As shown in Table 1, compared with the previously reported bienzymatic glucose biosensors, the developed sensing system shows much wider linear range and about 3 orders of magnitude lower detection limit except one report. Such ultrahigh sensitivity was mainly ascribed to the co-immobilization of luminol, GOx, and HRP on nanometric GO. HRP catalyzed luminol-H2O2 CL reaction coupled with GOx catalyzed glucose oxidization reaction on nanometric GO surface could synergistically facilitate substrate mobility, amplify enzymatic reaction rate, and improve CL efficiency, resulting in highly enhanced CL signal for ultrasensitive glucose detection. The relative standard deviation (RSD) of five replicate detections of 1 μM glucose within a day and in different days was 3.42 and 4.1%, respectively, indicating good repeatability of the proposed glucose sensing system. Moreover, the present biosensor is convenient, simple, homogeneous, rapid, and low-cost, which renders it more potential to be applied in practical applications.
Fig. 5

a CL kinetic curves of luminol-GOx-HRP-GO in the presence of glucose at varying concentrations. Inset: Calibration curves for glucose detection. b CL response of glucose (1 mg/ml) and other interfering substances (10 mg/mL) including K+, Na+, ascorbic acid, cysteine, IgG, goat-anti human IgG, BSA, lactose, maltose, and sucrose. Reaction conditions: 10 mM H2O2 in 0.1 M NaOH. CL measurement: PMT voltage was set at −350 V

Table 1

A comparison of the proposed glucose sensor with previously reported bienzymatic glucose biosensor

Method

Glucose linear range (mM)

Glucose LOD (μM)

Reference

ECa

0.005–0.83

1

[41]

EC

0.08–3

50

[42]

EC

0.0165–10

5.4

[43]

EC

0.03–2.43

30

[24]

EC

0.1–5

9.4

[44]

EC

0.4–15

400

[45]

CL

0.01–1

5

[46]

CRETb

1 × 10−5–1

3.4 × 10−3

[23]

CL

5 × 10−6–5

1.2 × 10−3

This work

a EC electrochemical

b CRET chemiluminescence resonance energy transfer

To evaluate the specificity of the proposed biosensor, the CL responses of the assay toward potentially interfering substances including relevant ions, proteins, amino acids, and glucose analogs (sucrose, lactose, and maltose) were tested. As shown in Fig. 5b, the CL responses of these interfering substances were negligible even when their concentrations were ten times higher than that of glucose. The results demonstrated that the proposed sensing system has a high selectivity for glucose detection. Such high selectivity of the assay may be ascribed to the specificity of GOx to catalyze glucose oxidation to produce H2O2, and the catalytic specificity of HRP for luminol-H2O2 CL reaction.

Application

In order to explore the applicability of the present biosensor for the detection of glucose in complex biological samples, the method was applied to detect glucose in human blood samples. The experimental protocol was approved by the Research Ethics Committee of Hefei University of Technology, China. The blood samples were collected from healthy volunteers. All participants provided written informed consent. Owing to the ultrahigh sensitivity and ultra-wide linear range of the proposed detection method, only 10 μL blood sample was required. To reduce the matrix interference effect, the blood samples were diluted 100 times with phosphate buffer (pH 7.0) and 5 times with 0.1 M NaOH solution prior to the measurements. The analytical results are presented in Table 2. It clearly revealed that the results obtained by the proposed sensing system were in good agreement with those obtained by the commercial glucometer method, and the RSD values of the measurements were from 1.6 to 4.8%. In addition, standard addition experiments display that the recoveries were in the range from 91.4 to 107.6% (ESM Table S1). These results verified that the sensing strategy has high selectivity and is applicable for monitoring glucose levels in real blood samples.
Table 2

Determination of glucose in human blood samples with the present CL sensor

Samples

CL sensor (mM)

Glucometer (mM)

RSD (%, n = 3)

1

6.2 ± 0.21

6.4 ± 0.23

3.2

2

5.2 ± 0.18

5.15 ± 0.32

2.7

3

5.6 ± 0.13

5.87 ± 0.41

2.6

Conclusion

In conclusion, CL reagent and bienzyme multifunctionalized GO bionanocomposites HRP/GOx-luminol-GO with excellent CL activity and specific enzymatic property was prepared through a simple and effective strategy for the first time. The co-immobilization of CL reagent luminol and bienzyme, GOx for H2O2 generation and HRP for H2O2 consumption, on the nanometric GO sheet simultaneously could facilitate in situ H2O2 generation, substrate transportation and consumption, thus accelerating the cascaded enzymatic reactions and greatly amplifying the CL efficiency. Then, a simple, homogenous, and reagentless sensing system was successfully developed for glucose detection based on the excellent CL property, catalytic activity, and enzyme specificity of HRP/GOx-luminol-GO. The detection range of the presented sensing system was ultra-wide ranged from 5.0 nM to 5.0 mM (6 order of magnitude), and the detection limit of 1.2 nM was ultra-low. Moreover, the HRP/GOx-luminol-GO-based sensing system has been successfully applied for the detection of glucose in human blood samples. The proposed sensing strategies were ultrasensitive, selective, convenient, low-cost, and quick response, which hold great application potential in diabetes mellitus research, clinical diagnosis, food industry, and biomedical research. Furthermore, this sensing strategy may be extended for the detection of various species involving H2O2-generation reactions, such as cholesterol, uric acid, lactic acid, and so on, by simply replacing GOx with their specific oxidoreductases. This work reveals that multifunctionalization of GO by CL reagent and enzymes leads to novel functionalized bionanocomposites with excellent CL feature and specific enzymatic property, which hold great potential in medical diagnostics and biomedical applications.

Notes

Acknowledgements

This work was supported by the National Natural Science Foundation of PR China (No. 21605032, 21672049) and the Fundamental Research Funds for the Central Universities (No. JZ2015HGBZ0455).

Compliance with ethical standards

The authors have declared that no conflict of interest exists. The experimental protocol was approved by the Research Ethics Committee of Hefei University of Technology, China. All participants provided written informed consent.

Supplementary material

216_2017_752_MOESM1_ESM.pdf (1 mb)
ESM 1 (PDF 1069 kb)

References

  1. 1.
    Li D, Mueller MB, Gilje S, Kaner RB, Wallace GG. Processable aqueous dispersions of graphene nanosheets. Nature Nanotech. 2008;3(2):101–5.CrossRefGoogle Scholar
  2. 2.
    Park S, Ruoff RS. Chemical methods for the production of graphenes. Nature Nanotech. 2009;4(4):217–24.CrossRefGoogle Scholar
  3. 3.
    Allen MJ, Tung VC, Kaner RB. Honeycomb carbon: a review of graphene. Chem Rev. 2010;110(1):132–45.CrossRefGoogle Scholar
  4. 4.
    Marcano DC, Kosynkin DV, Berlin JM, Sinitskii A, Sun Z, Slesarev A, et al. Improved synthesis of graphene oxide. ACS Nano. 2010;4(8):4806–14.CrossRefGoogle Scholar
  5. 5.
    Zhu Y, Murali S, Cai W, Li X, Suk JW, Potts JR, et al. Graphene and graphene oxide: synthesis, properties, and applications. Adv Mater. 2010;22(35):3906–24.CrossRefGoogle Scholar
  6. 6.
    Elias DC, Nair RR, Mohiuddin TMG, Morozov SV, Blake P, Halsall MP, et al. Control of graphene's properties by reversible hydrogenation: evidence for graphane. Science. 2009;323(5914):610–3.CrossRefGoogle Scholar
  7. 7.
    He S, Song B, Li D, Zhu C, Qi W, Wen Y, et al. A graphene nanoprobe for rapid, sensitive, and multicolor fluorescent DNA analysis. Adv Funct Mater. 2010;20(3):453–9.CrossRefGoogle Scholar
  8. 8.
    Li Y, Gao W, Ci L, Wang C, Ajayan PM. Catalytic performance of Pt nanoparticles on reduced graphene oxide for methanol electro-oxidation. Carbon. 2010;48(4):1124–30.CrossRefGoogle Scholar
  9. 9.
    Shen W, Yu Y, Shu J, Cui H. A graphene-based composite material noncovalently functionalized with a chemiluminescence reagent: synthesis and intrinsic chemiluminescence activity. Chem Commun. 2012;48(23):2894–6.CrossRefGoogle Scholar
  10. 10.
    Wang Y, Li Z, Wang J, Li J, Lin Y. Graphene and graphene oxide: biofunctionalization and applications in biotechnology. Trends Biotechnol. 2011;29(5):205–12.CrossRefGoogle Scholar
  11. 11.
    Gao L, Zhang H, Cui H. A general strategy to prepare homogeneous and reagentless GO/lucigenin&enzyme biosensors for detection of small biomolecules. Biosens Bioelectron. 2014;57:65–70.CrossRefGoogle Scholar
  12. 12.
    Li Q, Li F, Shen W, Liu X, Cui H. Lucigenin/co(tryptophan)2complex bifunctionalized graphene oxide: facile synthesis and unique chemiluminescence. J Mater Chem C. 2016;4(16):3477–84.CrossRefGoogle Scholar
  13. 13.
    Liu D, Huang G, Yu Y, He Y, Zhang H, Cui H. N-(Aminobutyl)-N-(ethylisoluminol) and hemin dual-functionalized graphene hybrids with high chemiluminescence. Chem Commun. 2013;49(84):9794–6.CrossRefGoogle Scholar
  14. 14.
    Liu X, Han Z, Li F, Gao L, Liang G, Cui H. Highly chemiluminescent graphene oxide hybrids bifunctionalized by N-(Aminobutyl)-N-(Ethylisoluminol)/horseradish peroxidase and sensitive sensing of hydrogen peroxide. ACS Appl Mater Inter. 2015;7(33):18283–91.CrossRefGoogle Scholar
  15. 15.
    Dodeigne C, Thunus L, Lejeune R. Chemiluminescence as diagnostic tool. Talanta. 2000;51(3):415–39.CrossRefGoogle Scholar
  16. 16.
    Liu J, Lu L, Li A, Tang J, Wang S, Xu S, et al. Simultaneous detection of hydrogen peroxide and glucose in human serum with upconversion luminescence. Biosens Bioelectron. 2015;68:204–9.CrossRefGoogle Scholar
  17. 17.
    Gill R, Bahshi L, Freeman R, Willner I. Optical detection of glucose and acetylcholine esterase inhibitors by H2O2-sensitive CdSe/ZnS quantum dots. Angew Chem Int Ed. 2008;47(9):1676–9.CrossRefGoogle Scholar
  18. 18.
    Kang X, Wang J, Wu H, Aksay IA, Liu J, Lin Y. Glucose oxidase-graphene-chitosan modified electrode for direct electrochemistry and glucose sensing. Biosens Bioelectron. 2009;25(4):901–5.CrossRefGoogle Scholar
  19. 19.
    Lin YH, Lu F, Tu Y, Ren ZF. Glucose biosensors based on carbon nanotube nanoelectrode ensembles. Nano Lett. 2004;4(2):191–5.CrossRefGoogle Scholar
  20. 20.
    Thorpe GH, Kricka LJ. Enhanced chemiluminescent reactions catalyzed by horseradish peroxidase. Methods Enzymol. 1986;133:331–53.CrossRefGoogle Scholar
  21. 21.
    Pavlov V, Xiao Y, Gill R, Dishon A, Kotler M, Willner I. Amplified chemiluminescence surface detection of DNA and telomerase activity using catalytic nucleic acid labels. Anal Chem. 2004;76(7):2152–6.CrossRefGoogle Scholar
  22. 22.
    Xu S, Li X, Li C, Li J, Zhang X, Wu P, et al. In situ generation and consumption of H2O2 by bienzyme-quantum dots bioconjugates for improved chemiluminescence resonance energy transfer. Anal Chem. 2016;88(12):6418–24.CrossRefGoogle Scholar
  23. 23.
    Zhu L, Yang R, Zhai J, Tian C. Bienzymatic glucose biosensor based on co-immobilization of peroxidase and glucose oxidase on a carbon nanotubes electrode. Biosens Bioelectron. 2007;23(4):528–35.CrossRefGoogle Scholar
  24. 24.
    Wang Z, Liu F, Lu C. Chemiluminescence flow biosensor for glucose using Mg-Al carbonate layered double hydroxides as catalysts and buffer solutions. Biosens Bioelectron. 2012;38(1):284–8.CrossRefGoogle Scholar
  25. 25.
    Bornscheuer UT. Immobilizing enzymes: how to create more suitable biocatalysts. Angew Chem Int Ed. 2003;42(29):3336–7.CrossRefGoogle Scholar
  26. 26.
    Zhang Y, Zhang J, Huang X, Zhou X, Wu H, Guo S. Assembly of graphene oxide-enzyme conjugates through hydrophobic interaction. Small. 2012;8(1):154–9.CrossRefGoogle Scholar
  27. 27.
    Wang Z, Zhou X, Zhang J, Boey F, Zhang H. Direct electrochemical reduction of single-layer graphene oxide and subsequent functionalization with glucose oxidase. J Phys Chem C. 2009;113(32):14071–5.CrossRefGoogle Scholar
  28. 28.
    Zhang J, Zhang F, Yang H, Huang X, Liu H, Zhang J, et al. Graphene oxide as a matrix for enzyme immobilization. Langmuir. 2010;26(9):6083–5.CrossRefGoogle Scholar
  29. 29.
    Zhang Q, Yang S, Zhang J, Zhang L, Kang P, Li J, et al. Fabrication of an electrochemical platform based on the self-assembly of graphene oxide-multiwall carbon nanotube nanocomposite and horseradish peroxidase: direct electrochemistry and electrocatalysis. Nanotechnology. 2011;22(49).Google Scholar
  30. 30.
    Zhang F, Zheng B, Zhang J, Huang X, Liu H, Guo S, et al. Horseradish peroxidase immobilized on graphene oxide: physical properties and applications in phenolic compound removal. J Phys Chem C. 2010;114(18):8469–73.CrossRefGoogle Scholar
  31. 31.
    Zhang Y, Wu C, Guo S, Zhang J. Interactions of graphene and graphene oxide with proteins and peptides. Nanotechnol Rev. 2013;2(1):27–45.CrossRefGoogle Scholar
  32. 32.
    Stankovich S, Dikin DA, Piner RD, Kohlhaas KA, Kleinhammes A, Jia Y, et al. Synthesis of graphene-based nanosheets via chemical reduction of exfoliated graphite oxide. Carbon. 2007;45(7):1558–65.CrossRefGoogle Scholar
  33. 33.
    Li TH, Park HG, Lee HS, Choi SH. Circular dichroism study of chiral biomolecules conjugated nanoparticles. Nanotechnology. 2004;15(10):S660–S3.CrossRefGoogle Scholar
  34. 34.
    Ciaurriz P, Bravo E, Hamad-Schifferli K. Effect of architecture on the activity of glucose oxidase/horseradish peroxidase/carbon nanoparticle conjugates. J Colloid Interf Sci. 2014;414:73–81.CrossRefGoogle Scholar
  35. 35.
    Shenoy R, Bowman CN. Kinetics of interfacial radical polymerization initiated by a glucose-oxidase mediated redox system. Biomaterials. 2012;33(29):6909–14.CrossRefGoogle Scholar
  36. 36.
    Johnson LM, Fairbanks BD, Anseth KS, Bowman CN. Enzyme-mediated redox initiation for hydrogel generation and cellular encapsulation. Biomacromolecules. 2009;10(11):3114–21.CrossRefGoogle Scholar
  37. 37.
    Yang L, Zhang R, Liu B, Wang J, Wang S, Han M-Y, et al. Pi-conjugated carbon radicals at graphene oxide to initiate ultrastrong chemiluminescence. Angew Chem Int Ed. 2014;53(38):10109–13.CrossRefGoogle Scholar
  38. 38.
    Song Y, Qu K, Zhao C, Ren J, Qu X. Graphene oxide: intrinsic peroxidase catalytic activity and its application to glucose detection. Adv Mater. 2010;22(19):2206–10.CrossRefGoogle Scholar
  39. 39.
    Wang DM, Zhang Y, Zheng LL, Yang XX, Wang Y, Huang CZ. Singlet oxygen involved luminol chemiluminescence catalyzed by graphene oxide. J Phys Chem C. 2012;116(40):21622–8.CrossRefGoogle Scholar
  40. 40.
    Huang Y, Wang W, Li Z, Qin X, Bu L, Tang Z, et al. Horseradish peroxidase-catalyzed synthesis of poly(thiophene-3-boronic acid) biocomposites for mono-/bi-enzyme immobilization and amperometric biosensing. Biosens Bioelectron. 2013;44:41–7.CrossRefGoogle Scholar
  41. 41.
    Wang F, Gong W, Wang L, Chen Z. Enhanced amperometric response of a glucose oxidase and horseradish peroxidase based bienzyme glucose biosensor modified with a film of polymerized toluidine blue containing reduced graphene oxide. Microchim Acta. 2015;182(11–12):1949–56.CrossRefGoogle Scholar
  42. 42.
    Gong C, Chen J, Song Y, Sun M, Song Y, Guo Q, et al. A glucose biosensor based on the polymerization of aniline induced by a bio-interphase of glucose oxidase and horseradish peroxidase. Anal Methods. 2016;8(7):1513–9.CrossRefGoogle Scholar
  43. 43.
    Wang Y, Hasebe Y. Glucose oxidase-modified carbon-felt-reactor coupled with peroxidase-modified carbon-felt-detector for amperometric flow determination of glucose. Mater Sci Eng: C. 2012;32(3):432–9.CrossRefGoogle Scholar
  44. 44.
    Xu S, Qi H, Zhou S, Zhang X, Zhang C. Mediatorless amperometric bienzyme glucose biosensor based on horseradish peroxidase and glucose oxidase cross-linked to multiwall carbon nanotubes. Microchim Acta. 2014;181(5–6):535–41.CrossRefGoogle Scholar
  45. 45.
    Lan D, Li B, Zhang Z. Chemiluminescence flow biosensor for glucose based on gold nanoparticle-enhanced activities of glucose oxidase and horseradish peroxidase. Biosens Bioelectron. 2008;24(4):940–4.CrossRefGoogle Scholar
  46. 46.
    Lan D, Li B, Zhang Z. Chemiluminescence flow biosensor for glucose based on gold nanoparticle-enhanced activities of glucose oxidase and horseradish peroxidase. Biosens Bioelectron. 2008;24(4):934–8.Google Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2017

Authors and Affiliations

  1. 1.Anhui Province Key Laboratory of Advanced Catalytic Materials and Reaction Engineering, School of Chemistry and Chemical EngineeringHefei University of TechnologyHefeiChina

Personalised recommendations