Introduction

The neurotransmitters dopamine (DA) and serotonin (5-hydroxytryptamine, or 5-HT) play an important role in the central nervous, cardiovascular, hormonal and renal systems [1,2,3,4]. On the other hand, L-tryptophan (L-Trp) is one of the most important essential amino acids having clinical and biochemical significance in humans and herbivores [5, 6]. Due to the coexistence of DA, 5-HT and L-Trp in biological system, their simultaneous determination is an important task. However, the oxidation peaks of these compounds are at a close potential and overlap during simultaneous determination. Moreover, the oxidation potential of ascorbic acid (AA) is very close to that of DA and 5-HT which results in an overlapping voltammetric response.

A great deal of research effort has been directed towards separating the anodic peak of AA, DA, 5-HT, and L-Trp by using different electrochemical techniques [7,8,9]. Among all analytical methods, electrochemical techniques have been paid much more attention due to high sensitivity, high accuracy and simple operation [10,11,12,13]. Poly amino acids modified electrodes have been the focus of intense research interest in the field of electrochemical sensors due to their extraordinary electrocatalytic properties [14, 15]. Among them, poly(L–arginine) has attracted significant attention which can be easily electropolymerized on electrode surface through combination of -NH2 and -COOH [16, 17]. Such polymer films can notably improve the stability of the electrode response and improve the electroactivity properties of analytes. For preparing electroactive and functionalized polymers at electrodes surface, electropolymerization is the incredibly convenient method [15].

Few researchers combined polymers and graphene oxide (GO) composites to improve the electrocatalytic performance of the electrodes [15, 18]. The guanidyl group of P-Arg engaged in hydrogen bonds with unique feature which can electrostatically interact favorably with this partial negative charge groups of GO. Moreover, the free amine group of P-Arg can easily interact with the carboxyl groups of GO. It is well known that GO-based materials provide an effective sensing platform toward the selective detection of bio-entities due to their tunable basal plane oxygen functionalities. The two-dimensional honeycomb sp2 carbon lattice of reduced graphene oxide (rGO) holds a great potential for sensing applications as it offers a rapid location of redox potentials of the electroactive species when employed as an electrode substrate [19, 20]. Preparation of rGO by electrochemical reduction method is recognized as one of the most promising methods that provides highly conductive, low-oxygen-containing graphene sheets without relying on toxic reagents [21, 22].

Moreover, metallic nanoparticles functionalized with rGO have been implemented in different modified electrodes to improve the electrical, thermal, and optical properties of the devices. Among them, Au-nanoparticle (AuNP) decorated graphene oxides are one of the most studied and widely employed in electrochemical sensing, due to the ease of preparation, simple surface functionalization and high analytical sensitivity [23, 24]. Zhang et al. also reported that rGO/AuNP composite shows better catalytic activities than AuNP/GO composite or free AuNP [25].

This work describes a novel electrochemical sensor based on the advantages of poly-Arg, ErGO, AuNP film modified GCE electrode (GCE/P-Arg/ErGO/AuNP) for the simultaneous determination of DA, 5-HT and L-Trp by DPV technique. The formation of ‘layer-by-layer’ polymer, reduced graphene oxide and metal nanoparticles film was firstly monitored. The electrocatalytic activity of the different modified electrodes was also evaluated. Furthermore, the potential application of the modified electrode for determination of these analytes in the real human urine samples was also investigated.

Materials and methods

Reagents

L-Arginine, dopamine hydrochloride, serotonin hydrochloride, L-Tryptophan, L-Ascorbic acid, Tetrachloroaurate (HAuCl4) trihydrate were purchased from Aladdins Reagent, Shanghai, China (http://www.aladdin-e.com). Modified Hummers method was used to prepare graphene nanosheet from natural graphite powder as described in previous work [26] and it was kindly supplied by Prof. Yuta Nishina, Japan. All the reagents for GO nanosheet preparation were purchased from Sigma Aldrich (http://www.sigmaaldrich.com/). The solutions of AA, DA, UA and L-Trp were prepared daily by dissolving the required amount of reagent in 0.1 M PB (pH 7.0). A redox probe solution was prepared in 0.1 M KCl, which contained 5.0 mM Fe(CN)6 3− and 5.0 mM Fe(CN)64−.

Preparation of the modified electrode (GCE/P-Arg/ErGO/AuNP)

The sequential ‘layer-by-layer’ film was fabricated on a GCE with a diameter of 3 mm by cyclic voltammetry (CV). Prior to modification, the GCE electrode was mirror fine-polished with 0.05 μm alumna slurry on microcloth pads and then subsequently ultrasonically washed using nitric acid (1:1), ethanol, and deionized water. Next, the electrode was rinsed with doubly distilled water and dry under N2. Scheme 1 illustrates the details of the modification procedure. The cleaned GCE was dipped into 0.1 M PB (pH 7.0) containing 2.5 mM L-Arg and cyclic voltammetry (CV) was performed in the potential range from −2.2 to 2.0 V at 100 mV s−1 for a defined cycle. After electropolymerization, the GCE/P-Arg electrode was carefully washed with ultrapure water and dried in air. To prepare GCE/P-Arg/ErGO film, 0.5 mg ml−1 GO was dispersed in 0.1 M PB (pH 7.0) solution in nanosheet form and was prepared according to our previous report [27]. Later, CV method (condition: 15 cycles @ −1.4 and 0.7 V vs. Ag/AgCl at a scan rate of 50 mVs−1) was used for electrodeposition of GO on GCE/P-Arg electrode under deoxygenated conditions. After electrodeposition, the modified electrode GCE/P-Arg/ErGO was washed with ultra-pure water, and then air-dried. Subsequently, the modified electrode was immersed in 0.5 mM HAuCl4 + 0.1 M KNO3 aqueous solution and CV was run for 20 continuous cycles (−0.5 to 0 V at a scan rate of 50 mVs−1) to obtain GCE/P-Arg/ErGO/AuNP modified electrode. At last, the GCE/P-Arg/ErGO/AuNP was carefully washed with ultra-pure water and stored in 4 °C for use. For electrode characterization and comparison, four different electrodes were also prepared with or without P-Arg, ErGO, and AuNP respectively.

Scheme 1
scheme 1

Step by step modification technique of GCE/P-Arg/ErGO/AuNP modified electrode

Full size image

Detection of DA, 5-HT and L-Trp in real sample

The sensor was applied to the analysis of real urine samples in order to assess its analytical performance. Following the standard addition method, 0.1 mL of human urine sample was spiked to 10 ml of PB solution (0.1 M, pH 7.0) and purged N2 to removal oxygen. Different concentrations of standard analyte solution were added to the diluted urine samples. We then measured the DPV curves in a potential range of 0.0 V to 1.0 V with amplitude of 50 mV and a pulse width of 0.2 s.

Apparatus

CHI electrochemical workstation (Shanghai CH Instruments, Model CHI-660E) (http://www.chinstruments.com/) with conventional three electrode system was used for electrochemical measurements. Electrochemical impedance spectroscopy (EIS) studies were analyzed using ZAHNER impedance analyzer EIM6ex ZAHNER (Kroanch, Germany) (http://zahner.de/). Before the EIS measurements, the electrode was cycled for 3 cycles and then measured in a frequency range from 0.01 Hz to 1.0 MHz. A small ac signal of 10 mV in amplitude was used as the perturbation of the system throughout the tests. Hitachi S-3000H model (https://www.hitachi-hightech.com) scanning electron microscopy (SEM) was used for surface characterization. All voltammetric measurements were carried out at room temperature under nitrogen (N2) atmosphere.

Results and discussion

Choice of materials

Graphene is an outstanding material which exhibits larger surface area, higher electrical conductivity and better stability than carbon nanotubes or carbon dots. The combination of graphene oxide and polymer composite enhance the performance of biosensors in terms of sensitivity, selectivity and efficiency. Compared with other polymers, the multidentate characteristics of P-Arg provide electrostatic interactions with negatively charged groups of GO. Owing to low-cost, versatility and easiness of the electrodeposition, P-Arg/ErGo composite has been used for the modification of electrode. In contrast to graphene and GO, reduced graphene oxide ensures high electrochemical activity, more reactive sites, as well as the more efficient charge transport. Moreover, AuNP were employed for improving the analytical sensitivity of the modified electrode.

Characterization of the modified electrode (GCE/P-Arg/ErGO/AuNP)

The details observation of the layer-by-layer electrodeposition process was described in electronic supporting material (ESM) section S1–3. We adopted cyclic voltammetry technique for both electropolymerization and electrodeposition along with AuNP deposition without any reducing agents onto the electrode surface. SEM measurements were done to investigate the surface morphology of the bare and different modified GCE and presented in Fig. 1. The image of Fig. 1b shows the P-Arg layer clearly deposited on GCE with regular and patterned interconnected net-like structure. The observed surface morphology indicates that P-Arg film has been successfully adhered to the electrode surface. After electrochemical deposition/reduction of GO, the ErGO nanosheets were observed as well connected and closely arranged thin folded sheets as shown in Fig. 1c. Finally, AuNP with an average size of ~30 nm uniformly and densely decorated along the surface of ErGO film by electrodeposition at 50 mVs−1 (Fig. 1d). The size distribution of AuNP in the electrodeposited GCE/P-Arg/ErGO/AuNP film was presented in Fig. S4. However, the electro-deposited AuNP were slightly aggregated due to the absence of any stabilizing functional groups in ErGO film.

Fig. 1
figure 1

SEM images of bare GCE (a), GCE/P-Arg film (b), GCE/P-Arg/ErGO film (c), and GCE/P-Arg/ErGO/AuNP film (d)

Full size image

The electrochemical characterization of the bare and modified electrodes was performed via CV and EIS in 0.1 M KCl including 5.0 mM Fe(CN)63−/4–. A pair of well-defined redox peaks corresponding to Fe(CN)63−/4– appeared with the lowest peak currents at the bare GCE. The peak to peak separation (ΔEp) value was calculated as 254 mV for bare GCE electrode which indicates the occurrence of a slow electron-transfer kinetics at the surface of the bare GCE. However, after electrodeposition of L-Arginine, the anodic and cathodic peaks increased with a ΔEp value of 185 mV. As reported by earlier researchers, the electrochemical reaction of Fe(CN)63−/4– effectively facilitated by the positive charges of poly(L-arginine) [5, 15]. After electrodeposition of GO, the peak current significantly increased as shown in Fig. 2a, with a ΔEp value of 123 mV which indicates increased electrochemical active sites with ErGO modification. The highest redox current and lowest potential difference were observed with GCE/P-Arg/ErGO/AuNP modified electrode. The possible reason may be ascribed to that the conductive graphene sheets connected individual AuNP and effectively facilitated electron transfer between the modifying layer and GCE substrate, thus resulting in synergic effect. These observations demonstrated excellent electro-catalytic prosperities of the modified GCE/P-Arg/ErGO/AuNP electrode.

Fig. 2
figure 2

Cyclic voltammograms (a) and Nyquist plot of EIS (b) of the different electrodes measured in 0.1 M KCl including 5.0 mM Fe(CN)63−/4–: (a) bare GCE, (b) GCE/P-Arg, (c) GCE/P-Arg/ErGO, and (d) GCE/P-Arg/ErGO/AuNP. CV response for GCE/P-Arg/ErGO/AuNP at different scan rates (from inner to outer): 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 150 and 200 mV/s in 0.1 M KCl including 5.0 mM Fe(CN)63−/4– solution (c). All potentials are given vs. Ag/AgCl. The inset shows the dependence of peak currents on the scan rates

Full size image

We have also calculated the electrochemical active surface area for bare and modified electrodes based on the Randles–Sevcik equation (Eq. 1) [28].

$$ {I}_p=2.69\times {10}^5{AD}^{1/2}\ {n}^{3/2}\ {\upgamma}^{1/2}C $$
(1)

Where A is the electroactive surface area, C is the concentration of the probe molecule, D is the diffusion coefficient (6.70 ± 0.02 × 10−6 cm2/s), γ is the scan rate (V/s), and n is the number of electrons in the redox reaction. The calculated A value for the GCE, GCE/P-Arg, GCE/P-Arg/ErGO, and GCE/P-Arg/ErGO/AuNP were 0.085 cm2, 0.135 cm2, 0.204 cm2 and 0.288 cm2 respectively. This finding proved that GCE/P-Arg/ErGO/AuNP modified electrode has the maximum electrochemical active surface area and high electron transfer kinetics.

Electrochemical impedance spectroscopy (EIS) experiments were performed to confirm the synergy between various modified electrodes and the results were shown in Fig. 2b. Theoretically, the electron-transfer kinetics of the redox probe at the electrode interface controlled by the charge transfer resistance (Rct) which represents the diameter of the semicircular portion presented in the Nyquist plot at higher frequencies of the EIS. It was observed that the bare GCE shows the largest semicircle. The Rct value of 922 Ω underpins the low transfer rate. Whereas GCE/P-Arg modified electrode shows a Rct value of 379 Ω. The diameter of semicircle obviously decreases after ErGO modification owing to the high electrical conductivity. The smallest semicircle (Rct value of 115 Ω) was observed at the GCE/P-Arg/ErGO/AuNP indicating that the synergistic effect of the conductivity of ErGO and AuNP layer, thereby promoting a higher electron transfer rate in the redox probe.

Moreover, the relationship between different scan rate and peak current was also studied to investigate the electrochemical mechanism of the electrode. From Fig. 2c, it can be observed that the peak currents increased with scan rate (in the range of 10–200 mV/s) for GCE/P-Arg/ErGO/AuNP modified electrode. A linear relationship with good correlation coefficients was observed between the peak current values and the scan rate as shown in the inset of Fig. 2c.

Electrochemical behavior of target analytes at different electrodes

CV behaviors of the bare and modified GCE electrodes were individually investigated in 0.1 M PB (pH 7.0) by adding target analytes. Figure 3 displays the cyclic voltammograms of AA, DA, 5-HT, and L-Trp at the bare GCE (a), GCE/P-Arg (b), GCE/P-Arg/ErGO (c), and GCE/P-Arg/ErGO/AuNP (d), respectively. A broad peak was observed at the bare GCE due to the overlapping oxidation peaks of AA, DA. However, the GCE/P-Arg/ErGO/AuNP electrode successfully separated the voltammetry peaks into four well-separated peaks at the potentials of 0.048 V (AA), 0.202 V (DA), 0.381 V (5-HT) and 0.719 V (L-Trp), which were large enough to determine AA, DA, 5-HT and L-Trp individually and simultaneously. The separations of Ep of AA–DA, DA–5-HT and 5-HT–L-Trp were of 154, 179 and 338 mV, respectively. These results indicate that the simultaneous detection of DA, 5-HT and L-Trp is feasible with the sensor. The excellent electrochemical performance can be ascribed to the synergistic effects between the ErGO and AuNP that would be beneficial for improved electron transfer kinetics.

Fig. 3
figure 3

The CV curves of AA (200 μM), DA (5 μM), 5-HT (10 μM) and L-Trp (10 μM) recorded on bare GCE (a), GCE/P-Arg (b), GCE/P-Arg/ErGO (c), and GCE/P-Arg/ErGO/AuNP (d) in phosphate buffer of pH 7.0, scan rate set at 100 mV s−1

Full size image

Individual voltammetric responses of DA, 5-HT, and L-Trp

The electrochemical oxidation mechanism of target analytes at GCE/P-Arg/ErGO/AuNP modified electrode was illustrated in Scheme 2. The electrochemical behavior of modified electrode in the presence of DA, 5-HT, L-Trp was investigated by differential pulse voltammograms (DPV) in wide potential range. All experimental procedures were performed in 0.1 M PB (pH 7.0) at room temperature. For all the DPV analysis, we have employed the following parameters: constant time interval of 90 s; N2 gas purging into the electrolyte solution for 5 min before the start of each experiment. Figure 4a, b and c, illustrates the DPVs for the individual detection of DA, 5-HT, and L-Trp and their calibration plots. As shown in Fig. 4a, the change of DPVs indicates that the oxidative peak current (Ipa) has linear relationship with the DA concentration. The calibration plot for DA exhibits two linear segments in the range from 1.0 to 50 nM (R2 = 0.997), and 1.0 to 50 μM (R2 = 0.997), with a detection limit (S/N = 3) of 1 nM. On the other hand, a linear relation between the concentration of 5-HT and the peak current (Fig. 4b) was in the range of 10 to 500 nM (R2 = 0.998) and 1.0 to 10 μM (R2 = 0.992), with a detection limit of 30 nM (S/N = 3). As shown in Fig. 4c, the peak current of L-Trp increases linearly with its two concentration range from 10 to 70 nM (R2 = 0.999), and 10 to 100 μM (R2 = 0.998), with a detection limit of 0.1 μM (S/N = 3). The sensor has a better ability to determine target analytes over a broader range of concentrations. The observed two linear concentration ranges correspond to low and high concentrations can be attributed to the sensor nature; due to the variation in its analytes concentration tolerance. Same results were verified with repeated tests. The calculated sensitivities for DA, 5-HT, and L-Trp are 2.48 μA·μM−1·cm−2, 5.97 μA·μM−1·cm−2, and 0.35 μA·μM−1·cm−2 respectively. Moreover, the analytical findings of this study were compared with several modified electrodes reported in last few years and presented in Table 1.

Scheme 2
scheme 2

Electrochemical oxidation of DA, 5-HT, and L-Trp at GCE/P-Arg/ErGO/AuNP modified electrode

Full size image
Fig. 4
figure 4

a DPVs for determination of DA using GCE/P-Arg/ErGO/AuNP electrode in pH 7.0 phosphate buffer at scan rate of 50 mV s−1 with a wide range of concentrations from 0.001 to 50 μM of DA; (b) Determinations of 5-HT with a wide range of concentrations from 0.01 to 10 μM; and (c) Determination of L-Trp with a wide range of concentrations from 0.01 to 100 μM. All the inset show the corresponding calibration plots for high and low concentrations ranges

Full size image
Table 1 Comparison of the parameters from the electrochemical detection of DA, 5-HT, and L-Trp with different methods and modified electrodes
Full size table

Repeatability, reproducibility, stability and interference study

A series of repetitive voltammetric measurements were carried out at the same GCE/P-Arg/ErGO/AuNP electrode to evaluate the precision of this method. The relative standard deviations (R.S.D.) for each 10 μM DA, 5-HT, and Trp measurements (n = 10) were 1.72, 1.23 and 2.87% respectively, indicating an excellent detecting reproducibility. For stability test, the modified electrode was kept at 4 °C for 2 weeks and the peak current did not have an obvious change (lower than 5%) for the same concentration of 10 μM analyte solution, illustrating the good film stability of the sensor.

Since the presence of high concentration of AA in biological systems together with monoamine neurotransmitters, interference study was performed before target analytes determination. From Fig. 5, we observed that AA doesn’t interfere with the measurement of DA, 5-HT and L-Trp. The effect of some possible other main interfering substances was also examined. Indeed, it was observed that 100-fold excess L-cysteine, Na+, K+, Ca2+, and Zn2+, 300-fold excess glucose, 200-fold excess urea, and 100-fold excess citric acid caused only negligible change and did not interfere with the measurement of the mixer of 10 μM of DA, 5-HT and L-Trp in 0.1 M PB (pH 7.0). Hence, the sensor has good selectivity towards the simultaneous determination of DA, 5-HT, and L-Trp. The relative response of the sensor in the presence of the various examined interferants was shown in the Table S 1 .

Fig. 5
figure 5

DPVs of simultaneous determination of DA, 5-HT, and L-Trp in the presence of AA using GCE/P-Arg/ErGO/AuNP electrode in pH 7.0 phosphate buffer at scan rate of 50 mV s−1. a 0.1–5.0 μM of DA (a-i) in the presence of 400 μM AA, 0.5 μM 5-HT, and 5 μM L-Trp; b 0.01–1.0 μM of 5-HT (a-h) in the presence of 400 μM AA, 0.1 μM DA, and 5.0 μM L-Trp; c 5.0–60 μM of L-Trp (a-h) in the presence of 400 μM AA, 0.1 μM DA, and 0.5 μM 5-HT. The insets show the calibration plot for each analyte (n = 3)

Full size image

Real sample analysis

To investigate the practicality of the method, GCE/P-Arg/ErGO/AuNP electrode was applied to the determination of DA, 5-HT, and L-Trp in healthy human urine using the standard addition method. The human urine samples were diluted 100 times using 0.1 M PB (pH 7.0). The recovery results ranged from 98.0 to 102.6% reveals the possibility of the sensor for real biological samples. Good recovery values were observed and the results are summarized in Table 2.

Table 2 Simultaneous determination of DA, 5-HT, and L-Trp in human urine samples. (n = 3)
Full size table

However, the major limitations of this sensor are the complex modification procedure and operational capabilities. Meanwhile, this technique has limitations in simultaneous multiple analyte detection and sometimes results overlapped voltammetric response.

Conclusion

A novel electrochemical sensor was prepared for the individual and simultaneous determination of DA, 5-HT, and L-Trp in the presence of ascorbic acid. The GCE/P-Arg/ErGO/AuNP electrode successfully resolved the overlapped voltammetric signals of target analytes into well-separated peaks by the DPV method. In particular, the electrode showed wide linear concentration ranges, very low detection limit and good selectivity. The sensor shows high electrocatalytic activity, efficacy, excellent stability and reproducibility towards the electro-oxidation of DA, 5-HT, and L-Trp. Hence, the prepared GCE/PArg/ ErGO/AuNP sensor provides a potential platform for the analysis of DA, 5-HT, and L-Trp in biological samples for diagnostic research. Despite of some limitations, this sensor may contribute for point-to-care applications. Further extensive studies are suggested on the clinical application and long-term biocompatibility of the electrode.