Introduction

Cytochrome c (Cyt C) is a vital heme-containing metalloprotein which is located in the intermembrane space of mitochondria. It acts as an electron transport between complex III and complex IV of the breathing chain. Cyt C is the key biomolecule in the electron transport chain and is also an intermediate species in apoptosis. When mitochondria are damaged under pathological conditions, Cyt C enters the cytosol of the cell. This release of Cyt C from mitochondria into the cytosol is a critical event in the activation of intracellular signaling and causes caspase activation and leads to programmed cell death (apoptosis). Cytochrome c which is released from the mitochondria into the cytosol can be a trigger of the apoptosis. In a review article in 2008, Brown and Borutaite explained that the oxidized form of cytochrome c (Fe3+) can induce caspase activation via the apoptosome, while the reduced form (Fe2+) cannot; however, it is unclear whether this difference in behavior is entirely due to the oxidized and reduced forms of cytochrome c or not. Cytosolic cytochrome c of healthy cells is rapidly reduced by various enzymes or reductant species and this blocks the apoptosis. Whereas, in apoptotic cells, the oxidized cytosolic cytochrome c is rapidly oxidized by mitochondrial cytochrome oxidase and accessible due to permeabilization of the outer membrane. Thus, regulation of the redox state of cytochrome c potentially enables regulation of the intrinsic pathway of apoptosis [1, 2].

Therefore, monitoring of Cyt C is of great significance in clinical diagnostics and therapeutic research. The determination of Cyt C release was used for the screening of anticancer drugs, too [3]. Study of Cyt C redox activity was successfully used in the detection of respiratory poisons [4]. Recently, the analysis of Cyt C has attracted considerable attention and some techniques including cyclic voltammetry [5, 6], electrochemical impedance spectroscopy [7, 8], fluorescence spectrophotometry [9–12], and electrochemiluminescence [10, 13] have been reported for Cyt C measurement.

Electrochemiluminescence (ECL) is a well-known process in which the species generated at electrodes undergo high-energy electron transfer reactions to form excited states that emit light [14]. Because the ECL method shows some remarkable features such as low background, high sensitivity, good temporal/spatial control, low detection limit, and wide dynamic range, it has attracted much attention in recent years and is used extensively for biosensing purposes [15].

A solid-state ECL sensor based on the immobilization of tris(2,2-bipyridyl)ruthenium(II) (Ru(bpy)3 2+) on the solid electrode surface can significantly reduce the use of expensive reagents, enhances the ECL signal, simplifies experimental design, and improves the detection sensitivity [16]. Ru(bpy)3 2+ is widely used as a reagent in designing ECL sensors for a variety of analytes owing to its high emission efficiency, good stability in different pH, and superior electrochemical reversibility [17].

Use of nanomaterials in designing sensors is the most direct and simple strategy for signal amplification, because of their remarkable biocompatibility, conductivity, and the high loading of signal molecules for synergistic amplification of the target response [18–20]. Graphene (G), an allotrope of carbon in the form of a densely packed honeycomb two-dimensional lattice, has generated tremendous interest since its discovery in 2004 [21]. Graphene compounds are used in a wide variety of applications in the fields of photoelectric devices, fuel cells, sensors, and biosensors owing to their unique properties, such as high mechanical strength, good biocompatibility, high surface area, fast electron transfer rate, and potentially low manufacturing cost [22]. Recently, use of reduced graphene oxide (RGO) in construction of biosensors has been extensively studied [23]. Peng et al. indicated that nanocomposite RGO enhanced electrochemiluminescence of peroxydisulfate and provided a novel sensing strategy for biomolecules [24]. Zhou et al. fabricated graphene-Fe3O4 composite-modified indium tin oxide (ITO) electrodes to construct H2O2 biosensors [25]. These results indicate that RGO shows great potential as an amplification material in the fabrication of ECL biosensors.

Ceria (CeO2), an oxide of cerium, has received great attention for its unique catalytic properties in redox reactions [26]. It has two common oxidation states, Ce3+ and Ce4+, that make it a suitable compound for catalytic or electrocatalytic reactions. Also, owing to the high surface-to-volume ratio, CeO2 has a high capacity to absorb molecules such as proteins, amino acids, and sugars [27].

Decoration of RGO with various nanoparticles produces a new class of hybrid materials with properties that are different from those of each individual component. For example, several metal oxides, such as SnO2, Fe3O4, ZnO, TiO2, Co3O4, and MnO2, have been decorated on RGO for a variety of applications in energy storage, catalysis, and chemical and biosensing [28–32].

In this work, we constructed a new sensitive solid-state ECL nanocomposite biosensor for detection of Cyt C. The CeO2 NPs-RGO/ Ru(bpy)3 2+/chitosan (CHIT) modified glassy carbon electrode was designed to provide a large specific surface area and fast electron transfer rate enhanced ECL intensity. This ECL biosensor has a lower detection limit for cytochrome c in comparison with previously reported sensors.

Experimental

Materials and reagents

All the chemicals were of analytical reagent grade and used as received without further purification. Graphite powder (99.995 %), sulfuric acid (95 %), potassium permanganate, potassium nitrate, Ce(NO3)3·6H2O, aqueous NH4OH, and N2H4 were purchased from Shanghai, China. Tris(2,2-bipyridyl)ruthenium(II) (Ru(bpy)3 2+) chloride hexahydrate and chitosan were obtained from Sigma and used without further purification. Cytochrome c (bovine heart, molecular weight 12,588) and bovine serum albumin (98 %) were purchased from Sigma (St. Louis, MO, USA). Phosphate buffer solutions (PBS) with pH 7.4 were prepared by mixing K2HPO4, NaH2PO4, and KCl and adjusting the pH.

Apparatus

Cyclic voltammetry (CV) was performed using a PalmSens PC potentiostat–galvanostat (PalmSens, Houten, the Netherlands) with a conventional three-electrode setup in which a modified glassy carbon electrode, an Ag|AgCl|KCl sat. electrode, and a platinum wire served as the working, reference, and auxiliary electrodes, respectively. Electrochemical and ECL measurements were carried out in a 4-mL quartz cell. The working electrode was placed in an equatorial position in a quartz cell with its surface exactly in front of the window of the Fl-win lab photomultiplier LS 50 (Perkin-Elmer) and fixed on a holder. The detector and ECL cell were enclosed in a light-tight black box. Electrochemical impedance measurements were carried out on AUTOLAB PGSTAT 30. The size and morphology of the nanoparticles were measured by scanning electron microscopy (SEM) using a KYKY-EM 3200 digital scanning electron microscope (China).

Preparation of CeO2NPs -RGO nanocomposite

The CeO2NPs-RGO nanocomposite was prepared according to a simple sonochemical method previously reported [29]. The procedure was briefly as follows. Initially, 1 mmol of Ce(NO3)3·6H2O was dissolved in deionized water (30 mL). Then, during the ultrasound irradiation (in an ultrasonic bath), a 32 % NH4OH aqueous solution (2 mL) was added dropwise to the solution. The process was completed within 66 min. Next, 60 mL aqueous solution of GO suspension (0.5 mg mL−1), was added to the suspension and heated to 90 °C. Then 3 mL N2H4 was added and the suspension solution was heated at reflux for about 1 h. The resulting precipitate was washed several times with ethanol and distilled water before drying at 60 °C for 24 h. The final obtained product was CeO2NPs-RGO.

Fabrication of ECL biosensor

A glassy carbon electrode (GCE) was polished well with 1.0, 0.3, and 0.05 μm alumina slurry on a polishing cloth, rinsed thoroughly with water, and sonicated in ethanol and then distilled water. A 100-μL aliquot of 2.5 × 10−3 mol L−1 Ru(bpy)3 2+ aqueous solution was dispersed in 600 μL solution of 0.5 wt% chitosan which was prepared by dispersing 50.0 mg of chitosan in 10.0 mL of 3 % acetic acid. The CeO2NPs-RGO/ Ru(bpy)3 2+/ CHIT modified electrode was prepared by casting 4 μl of CeO2NPs-RGO nanocomposite suspension (0.5 mg mL−1 and 1:2 mass ratio of CeO2/G) on the electrode surface and dried at room temperature. Then, 5 μl of Ru(bpy)3 2+/ CHIT suspension was dropped on the CeO2NPs-RGO/GCE surface and dried at room temperature (Scheme 1).

Scheme 1
scheme 1

Fabrication of ECL sensor

Full size image

Results and discussion

Characterization of CeO2NPs-RGO

The crystal phase of the used CeO2NPs-RGO nanocomposite was studied by x-ray diffraction (XRD). As shown in Fig. 1, the diffraction peaks at 28.58, 33.17, 47.50, 56.41, 59.14, 69.49, 76.80, 79.21, and 88.38, correspond to (111), (200), (220), (311), (322), (400), (331), (420), and (422) planes, respectively. The obtained pattern matches with a typical cubic fluorite-type structure of CeO2 (JCPDS no. 34-0394), indicating that CeO2 nanoparticles were successfully anchored on the RGO [33].

Fig. 1
figure 1

XRD pattern of CeO2 NPs-RGO

Full size image

The FE-SEM image of the CeO2 nanoparticles before and after combining with RGO is shown in Fig. 2. As can be seen, a large number of CeO2 NPs were well dispersed on the RGO surface, indicating a good combination between nanoparticles and RGO. Through this combination, the agglomeration of CeO2 nanoparticles can be prevented effectively. And on the other hand, the CeO2 nanoparticles on the RGO’s surfaces can act as spacers and prevent the RGOs from restacking [29]. Consequently, the new nanostructure can increase the stability of the RGO and provide a much more active surface area [34, 35].

Fig. 2
figure 2

SEM image of CeO2 NPs before (a) and after combining with RGO (b)

Full size image

Electrochemical and ECL signal

The catalytic effect of rare earth oxide NPs such as Sm2O3 [36], CeO2 [37–39], and CeO2–graphene [27] on the ECL reaction has been shown in previous studies. To study the effect of the CeO2NPs-RGO nanocomposite on the oxidation of Ru(bpy)3 2+ and corresponding ECL signal, cyclic voltammetry using Fe(CN)6 3−/Fe(CN)6 4−as redox probe and ECL potential study by the modified electrodes was performed in the potential range between 0.0 and 1.5 V vs. Ag|AgCl|KCl sat. at a scan rate of 100 mV s−1. Figure 3A shows the CV curves for probing the electrode modification process. A redox peak of Fe(CN)6 3−/Fe(CN)6 4−was observed at the bare GCE (curve a). When CeO2NPs-RGO-CHIT nanocomposites were coated on the electrode, the peak current markedly increased as a result of the increase in the electrode surface area and the excellent conductivity of CeO2NPs-RGO nanocomposites (curve b). The peak current decreased in the absence of nanocomposites and the presence of Ru-CHIT (curve c). However, when CeO2NPs-RGO added to Ru-CHIT, the peak current increased significantly (curve d). To investigate the effect of CeO2NPs and RGO, two nanocomposites RGO -Ru-CHIT (curve e) and CeO2NPs -Ru-CHIT (curve f) were also examinated. As shown in Fig. 3, the RGO-Ru-CHIT nanocomposites increased the peak current more than CeO2 -Ru-CHIT.

Fig. 3
figure 3

CV (A) and ECL signal (B) from a bare GCE, b CeO2-RGO-CHIT, c Ru-CHIT, d CeO2-RGO/ Ru(bpy)3 2+-CHIT, e RGO/ Ru(bpy)3 2+-CHIT, f CeO2/ Ru(bpy)3 2+-CHIT. CV in 10 mM Fe(CN)6 3−/Fe(CN)6 4−(1:1) containing 0.1 M KCl. The CVs proceeded between 0 and 0.6 V with a scan rate of 100 mV s−1. ECL curves in 0.1 M pH 7.4 phosphate buffer solution containing 0.5 mM TPA

Full size image

Figure 3B presents the ECL–potential curves of the bare GCE and CeO2-RGO-CHIT, Ru-CHIT, CeO2-RGO-Ru-CHIT modified GCE (curves a–d). Clearly, bare GCE and CeO2-RGO-CHIT modified electrode have negligible ECL emission (Fig. 3a,b). However, Ru- CHIT and CeO2-RGO-Ru-CHIT modified GCE show high ECL intensity (Fig. 3c,d). It can be concluded that electrodes without Ru(bpy)3 2+ are not able to produce ECL emission. Figure 3b shows that ECL emission of CeO2-RGO-Ru-CHI modified electrode (curve d) compared with Ru-CHIT-GCE (curve c) increased twofold by modification of the electrode with CeO2-RGO nanocomposite. Also ECL emission of CeO2NPs -Ru-CHIT (curve f) was higher compared with RGO -Ru-CHIT (curve e).This increase of ECL emission is attributed to the increase in the electrode surface area due to the presence of CeO2-RGO nanocomposite and excellent electrical conductivity of the nanocomposite film [33]. In addition, CeO2 NP decorated RGO can enhance the signal owing to the electrocatalytic activity. The redox behavior of Ce4+/Ce3+ in the CeO2 and Ce2O3 can significantly catalyze the oxidation of Ru(bpy)3 2+ to Ru(bpy)3 3+[37].

Electrochemical impedance spectroscopy (EIS) was employed to investigate the conductivity of the electrode surface due to the modification with CeO2-RGO nanocomposite. Figure 4 shows the Nyquist plots of bare GCE, Ru(bpy)3 2+-CHIT, and CeO2 NP-RGO-Ru(bpy)3 2+-CHIT in 0.1 M KCl solution with 5 mM K3Fe(CN)6/K4Fe(CN)6 electroactive probes. In the high frequency zone, the semicircle’s diameter represents the electron-transfer resistance (R et) of the electrochemical reaction, and the linear part of Z Im versus Z Re at lower frequencies represents diffusion-controlled electrode process. The EIS spectrum for the bare GCE exhibited a small semicircular domain, at high frequencies (Fig. 4, curve a), and its electron-transfer resistance (R et) is small, indicating a fast electron-transfer process of Fe(CN)6]3−/4−. The spectrum for the Ru(bpy)3 2+-CHIT electrode indicated much increased electron-transfer resistance (Fig. 4, curve b). After modification of the electrode with CeO2 NP-RGO-Ru(bpy)3 2+-CHIT, the semicircle diameter was much smaller, indicating a lower R et value for Fe(CN)6]3−/4− (Fig. 4, curve c). This showed that the CeO2 NP-RGO nanocomposite could increase the conductivity of the electrode surface.

Fig. 4
figure 4

Impedance of a bare GCE, b Ru(bpy)3 2+-CHIT, c CeO2-RGO/ Ru(bpy)3 2+-CHIT in 10 mM Fe(CN)6 -3/-4 in 0.1 M KCl solution in the frequency range of 0.01 Hz to 100 kHz and signal amplitude 5 mV

Full size image

Optimization of ESL signal

To optimize the performance of the proposed ECL biosensor towards cytochrome c detection, the effects of pH, concentration of CeO2-RGO on the electrode surface, concentration of Ru(bpy)3 2+, concentration of TPA, and scan rate on the intensity of ECL signal were investigated.

The relationship between the ECL intensity and pH was investigated over a pH range of 5.5 to 10. Figure S1 (A) in the Electronic Supplementary Material (ESM) shows that the ECL intensity increased considerably with the increase in pH value from 5.5 to 7.4 and then decreased. The maximum ECL intensity was observed at pH 7.4. Thus, phosphate buffer solution of pH 7.4 was chosen for further ECL biosensor determination.

The effect of CeO2-RGO concentration on the electrode response of the ECL signal was investigated. As shown in ESM Fig. S1 (B), the ECL signal increased correspondingly when the concentration of CeO2-RGO increased from 0.1 to 0.5 mg mL−1, since the CeO2-RGO promoted the electron transfer on the electrode surface. However, the ECL response decreased when the concentration of the CeO2-RGO was more than 0.5 mg mL−1 is because of the blackbody effect. High ECL response was obtained at 0.5 mg mL−1 because there is a balance between the promoted charge transfer and the blackbody effect [40].

The effect of the Ru(bpy)3 2+ concentration in the modified electrode on the intensity of the ECL signal was also investigated. As shown in ESM Fig. S1 (C) the ECL intensity increased with increasing Ru(bpy)3 2+ concentrations. The intensity of the ECL signal in the presence of TPA increased linearly with an increase in the concentration of Ru(bpy)3 2+ from 0.5 to 2.5 mM. However, further increase in the concentration of Ru(bpy)3 2+ (up to 2.5 mM) did not cause any further enhancement in the intensity of the ECL signal (Scheme 2). Therefore, 2.5 mM was selected as the optimum concentration of Ru(bpy)3 2+ in modified the ECL biosensor for detection of cytochrome c.

Figure S1 (D) in the ESM shows the relationship between ECL intensity and TPA concentration at modified electrodes. The ECL intensity was directly related to the TPA concentration at the modified electrode as a result of the more excited state of [Ru(bpy)3 2+]* produced with the increasing TPA concentration. The ECL intensity increased at 0.5 mM of TPA and decreased with further concentration of TPA.

Scheme 2
scheme 2

Proposed mechanism of ECL sensor

Full size image

The effect of scan rate (v) on ECL and CV signals of Ru(bpy)3 2+/TPA systems was investigated (ESM Fig. S1(E)). The results showed that with increasing scan rate in the range between 25 to 100 mV/s the intensity of the ECL signal of Ru(bpy)3 2+ increased and achieved the maximum value at 100 mV/s. At a lower scan rate, the formation of the intermediate is very slow, which causes a lower ECL intensity. At higher scan rates, shorter reaction time led immediately to lower concentration and smaller transmission of the reaction, resulting in a lower ECL intensity [41]. A scan rate of 100 mV/s was selected for further experiments because maximum ECL intensity was achieved for that scan rate.

ECL detection of Cyt C

Under the optimal experimental condition, the proposed sensor showed strong and stable ECL emission in the presence of TPA (Fig. 5a). However, the ECL intensities decreased dramatically by addition of Cyt C concentrations. The ECL sensor responses to different concentrations of Cyt C are illustrated in Fig. 5b. As shown, the ECL intensity decreased with an increase in the Cyt C concentrations in the range of 2.5 nM to 2 μM (R = 0.995, n = 9). The detection limit for Cyt C determination in this work was 0.7 nM (S/N = 3) and relative standard deviation (RSD) for the measurement of 10 nM Cyt C by six sensors was 4.4 %. Also, a comparative study was performed between the different reported analytical methods for the detection of Cyt C and this work. As shown in Table 1, the proposed Cyt C biosensor displays higher sensitivity, wider linear range, and lower detection limit than those of other analytical methods.

Fig. 5
figure 5

a Stability of ECL signal from CeO2-RGO/ Ru(bpy)3 2+-CHIT modified electrode in 0.1 M pH 7.4 phosphate buffer solution, containing 0.5 mM TPA under ten cycles of CV scan. b Typical ECL signals of cytochrome c concentrations of samples were 0, 2.5, 5.0, 25, 50, 100, 250, 500, 1000, and 2000.0 nM (a–g, respectively)

Full size image
Table 1 Comparison of the Cyt C biosensor with other reported methods
Full size table

Interference studies were carried out by adding glucose, bovine serum albumin (BSA), arginine, and cysteine to the measuring solution of Cyt C in various concentrations. As shown in Fig. 6, no remarkable changes were observed compared to the quenching effect of Cyt C on ECL responses. Thus, the interference study results indicated that the biosensor has a good selectivity in the determination of Cyt C in real samples.

Fig. 6
figure 6

Selectivity of the proposed ECL sensor in 0.1 M pH 7.4 phosphate buffer solution

Full size image

The applicability of the proposed sensor for the determination of Cyt C in human serum was examined. For this purpose, the plasma samples were diluted with pH 7.4 PBS to ensure that the concentrations were within the linear ranges of the method. Cyt C was spiked into the human serum samples. The results showed that the recoveries were in the range of 94–103 % and RSDs of 1.8–3.6 % (Table 2). The recoveries indicate that both the accuracy and repeatability of the proposed sensor are very satisfactory. It is very clear that the developed methodology has great potential for determination of Cyt C in complex biological matrixes.

Table 2 Recovery studies performed in spiked serum samples for applicability of the developed biosensor (n = 3)
Full size table

Study of ECL quenching mechanism

A series of experiments were performed to study the ECL quenching of the Ru(bpy)3 2+ /TPA system by Cyt C.

To verify whether Cyt C can interact with Ru(bpy)3 2+ or TPA, we investigated the ECL of the Ru(bpy)3 2+/Cyt C system without the presence of TPA (Fig. 7a). Curve a is related to the annihilation ECL mechanism of Ru(bpy)3 2+[16]. It can be seen that Cyt C can also quench the ECL of Ru(bpy)3 2+ in the absence of TPA (curve b). This result suggested that Cyt C most likely interacted directly with Ru(bpy)3 2+. TPA here only acted as a co-reactant to enhance the ECL intensity of the system [44].

Fig. 7
figure 7

a ECL of Ru(bpy)3 2+ system without the presence of TPA. a 2.5 mM Ru(bpy)3 2+, b 2.5 mM Ru(bpy)3 2+ + 300 nM Cyt C. b Fluorescence intensity of 1 × 10−5 M Ru(bpy)3 2+; Cyt C: 0, 25, 50, 75, 100 nm (a–e, respectively). c UV–visible absorption spectra for Ru(bpy)3 2+ with Cyt C in 0.1 M pH 7.4 phosphate buffer solution a 2 × 10−7 M Ru(bpy)3 2+, b 1 × 10−6 Cyt C, c a + b

Full size image

The fluorescence quenching was studied in the presence and absence of Cyt C. As shown in Fig. 7b, the fluorescence of Ru(bpy)3 2+ with excitation at 350 nm and fluorescence emission peak at 620 nm could be quenched by Cyt C. The results showed that increasing the concentration of Cyt C resulted in a decrease in fluorescence emission of Ru(bpy)3 2+. The fluorescence decrease of Ru(bpy)3 2+ by Cyt C indicated that Ru(bpy)3 2+ had a strong interaction with Cyt C [44].

Also, in order to know whether or not chemical reaction occurs between Ru(bpy)3 2+and Cyt C, UV–visible absorption spectra of the Ru(bpy)3 2+ Cyt C system were investigated. As shown in Fig. 7c, The Ru(bpy)3 2+ shows absorption peaks at 286 and 455 nm (curve a). Cyt C had an absorption peak at around 410 nm (curve b). The absorption spectra of the mixed solution of the Ru(bpy)3 2+-Cyt C were exactly the sum of that of the individuals (curve c). The results verify that no new compounds were produced by simply mixing Cyt C with Ru(bpy)3 2+, which indicated that there is no chemical reaction between Ru(bpy)3 2+ and Cyt C by simply mixing Cyt C with Ru(bpy)3 2+ in the ground state [45]. Thus, the quenching of ECL signal by Cyt C can be due to the inhibitory effect of Cyt C on TPA (Scheme 2). Cyt C prevents the TPA from reaching the electrode surface which causes the decrease of the ECL intensity. Such behavior has been already reported by Wang et al. [13] and Hu et al. [10].

The observed quenching behavior can be described by the Stern-Volmer equation: F 0/F = 1 + K SV[Q] where F 0 and F are the fluorescence intensities observed in the absence and presence of quencher, respectively. [Q] is the quencher concentration and K SV is the Stern-Volmer quenching constant. The value of K SV for ECL quenching of Ru(bpy)3 2+ by Cyt C was calculated to be 1.1 nM .

Conclusions

In this work, a CeO2NPs-RGO nanocomposite was prepared by a facile sonochemical method and used for making a novel ESL biosensor. This nanocomposite has large specific surface, good conductibility, and high stability that could provide a large active surface and promote the formation of radical species. The proposed CeO2NPs-RGO/ Ru(bpy)3 2+/ CHIT nanocomposite can enhance ECL intensity to provide a favorable initial ECL signal for construction of a quenching biosensor for detection of Cyt C. The proposed biosensor demonstrated its advantages of high sensitivity, selectivity, long-term stability, simple fabrication, and excellent specificity towards Cyt C. Therefore, this method has the ability to be extended for the detection of the protein in clinical analysis.