Signal amplification strategy using gold/N ‐trimethyl chitosan/iron oxide magnetic composite nanoparticles as a tracer tag for high‐sensitive electrochemical detection

Abstract

This study presents a novel signal amplification method for high‐sensitive electrochemical immunosensing. Gold (Au)/N ‐trimethyl chitosan (TMC)/iron oxide (Fe₃ O₄) (shell/shell/core) nanocomposite was used as a tracing tag to label antibody. The tag was shown to be capable of amplifying the recognition signal by high‐density assembly of Au nanoparticles (NPs) on TMC/Fe₃ O₄ particles. The remarkable conductivity of AuNPs provides a feasible pathway for electron transfer. The method was found to be simple, reliable and capable of high‐sensitive detection of human serum albumin as a model, down to 0.2 pg/ml in the range of 0.25–1000 pg/ml. Findings of the present study would create new opportunities for sensitive and rapid detection of various analytes.

1 Introduction

Several assays have been developed for reliable measurement of multiple biomarkers, of which immunoassays are mainly used for quantitative detection of specific substances in biological specimen [1]. Traditional immunoassays employing enzyme, fluorescent, chemoluminescent and radio isotype labels are usedfor determination of analytes in clinical, food safety and environmental laboratories [2, 3, 4, 5]. In comparison, electrochemical immunoassays with intrinsic advantages of simplicity, low cost, high sensitivity and compatibility have attracted much more attention [4].

Nanoparticles (NPs) with different characteristics are extensively used in various kinds of analytical techniques [1, 4, 6, 7]. Gold NPs (AuNPs) [8, 9], nanoAu‐attached carbon nanotube [10], core/shell iron oxide (Fe₃ O₄)/AuNPs [4], Au nanorods (AuNRs) [11], dumbbell‐like platinum (Pt)‐Fe₃ O₄ NPs [12] and nanoAu‐enwrapped graphene nanocomposites [13] have been used as label in electrochemical immunosensing devices. These metals and semiconductor NPs can be used directly as an electroactive label for significant amplification of electrochemical detection, thanks to their huge surface area to weight ratio, remarkable conductivity and electronic properties. Hybrid nanomaterials are applicable for their novel combined properties. Fe₃ O₄ magnetic beads have distinctive features of easy handling by magnets, narrow size distributions, large surface areas and controllable sizes [14, 15]. Protecting naked magnetic beads with shells has been proved to overcome some of their limitations such as high chemical reactivity, poor magnetism and dispersibility [4, 16, 17]. Chitosan is an attractive natural polymer for immobilisation of biomolecules due to excellent membrane‐forming ability, good biocompatibility, non‐toxicity, high mechanical strength and low cost [14, 17, 18]. However, its solubility in acidic solutions restrains its biomedical application. This weakness has been addressed by quaternisation of nitrogen atoms of free amino groups originally presented in the chitosan chain creating functionalised N ‐trimethyl chitosan (TMC) with permanently positive charged sites [19, 20].

Increased positive charge sites in TMC display high affinity for adsorption of NPs, leading to better stability and preventing them from aggregation [21].

AuNPs are distinguished as conductive and biocompatible labels for electrochemical signal amplification [1, 22]. Their rapid and simple chemical synthesis, narrow size distribution, good stability and convenient labelling to biomolecules offer perfect conditions for immobilising macromolecules such as protein or DNA [9, 23]. Therefore, AuNPs decorated TMC/Fe₃ O₄ nanocomposite compared with AuNPs would be promising for the future of electrochemical biosensing as it offers large surface area to volume ratio and excellent electrical conductivity. Although a few reports have been presented on the synthesis of Au/chitosan/Fe₃ O₄ NPs [7, 14], no report has documented the synthesis of Au/TMC/Fe₃ O₄ NPs or using them as label in electrochemical immunosensor yet. This study evaluates membrane‐forming ability of TMC and its capability to produce a new label for signal amplification in electrochemical immunosensor, compared with chitosan.

The detection process was based on the immunoreaction of analyte with specific conjugated antibody (Ab) (anti‐human serum albumin (HSA) as a model) to Au/TMC/Fe₃ O₄ NPs. Electrochemical signals produced by preoxidisation of this nanocomplex in 1 M hydrochloric acid (HCl) at 1.27 V for 30 s, followed by the reduction of chloroauric acid (AuCl₄ ⁻) to Au⁰ in differential pulse voltammetry (DPV) mode. Oxidoreduction properties of the AuNPs in acidic medium, makes it possible to quantify the NPs, and the analyte in the sample could be detected thereafter.

The excellent analytical performance of the immunosensor raised the prospect for its application in early detection of diseases.

2 Experimental section

2.1 Reagents and solutions

Iron (III) chloride hexahydrate (FeCl₃. 6H₂ O) (99.0%), iron (II) chloride tetrahydrate (FeCl₂. 4H₂ O) (99.0%), sodium hydroxide (NaOH) and acetic acid were purchased from Acros Organics (USA). Chloroauric acid (HAuCl₄), sodium dodecyl sulphate (SDS), sodium azide, vinyl alcohol (VA), bovine serum albumin (BSA) HSA and dialysis tube with molecular cut off 12,000 Da were purchased from Sigma (UK). The analytical HCl and sodium chloride (NaCl) were prepared from Merck (Merck, Germany). Chitosan with low molecular weight was obtained from Primex (Iceland). N ‐methyl pyrrolidone (NMP), iodomethane, NaOH, sodium iodide, acetone D‐glucose and glutaraldehyde were obtained from Merck (Darmstadt, Germany). De‐ionised water was used in all the experiments.

2.2 Procedure

2.2.1 Synthesis of AuNPs seeds

The 12 nm glucose‐reduced AuNPs seeds were synthesised by the previously reported method with slight modifications [24, 25]. About 50 ml of aqueous solution containing 6 × 10⁻⁴ M HAuCl₄ was prepared and 5.5 ml of D‐glucose 2.4 M was added to the solution. The mixture was heated to 60 °C, and then 111.11 µl of 2.4 M NaOH was added while stirring (10 s). The solution was cooled to room temperature and Au colloids were adjusted to pH 8.5 [8].

2.2.2 Synthesis and preparation of superparamagnetic NPs

Fe₃ O₄ NPs were prepared by co‐precipitation of Fe (II) and Fe (III) chlorides (1:2 mol ratio) using NaOH as reductant agent, according to the reported methods with slight modifications [26, 27]. In brief, 2.365 g FeCl₃. 6H₂ O and 0.995 g FeCl₂. 4H₂ O were dissolved in 100 ml of de‐ionised and de‐oxygenated water. The reaction solution was purified with nitrogen and stirred in a water bath at 80 °C for 1 h under continuous stirring and nitrogen bubbling, and then 45 ml of NaOH solution (2 M) was quickly injected to the solution while the dark Fe₃ O₄ NPs were produced, the resultant mixture was stirred for 5 min to completely oxidise the particles to γ ‐Fe₂ O₃. The final precipitate was stored in tetramethylammonium hydroxide (TMOH, 0.1 M) at room temperature.

2.2.3 Preparation of TMC

Methylation of chitosan was achieved based on a single treatment with iodomethane in the presence of NMP and NaOH. Briefly, low molecular weight chitosan (1 g) was suspended in a basic solution of NMP (50 ml) and stirred at room temperature for 12 h. Afterwards, 15% (w/v) aqueous solution of NaOH (8 ml) and sodium iodide (3 g) were added and stirred in a water bath under constant stirring for 15 min at 60 °C.

Consequently, iodomethane (8 ml) was added at about 3 h intervals and the resulting solution stirred for a further 24 h at 60 °C. The synthesised derivative was precipitated using acetone, and thereafter separated by centrifugation. The obtained product was suspended in 15% NaCl aqueous solution to replace the iodide. The final solution was transferred to dialysing tube and dialysed against distilled water for 3 days and then freeze‐dried to give cotton‐liked powder TMC chloride [28, 29].

2.2.4 Synthesis of Au assembled magnetic NPs microspheres

The synthesis of the Au assembled magnetic NPs microspheres was accomplished using literature procedure with minor modifications [14]. TMC‐coated superparamagnetic Fe₃ O₄ NPs were initially prepared for attachment to AuNPs.

For this purpose, the synthesis nano Fe₃ O₄ (0.5 g) was dispersed into 50 ml solution containing 0.5 M NaCl and 0.025 M SDS. Thereafter, TMC was dissolved in 10 ml de‐ionised water and added tothe mixture (TMC/Fe₃ O₄ molar ratio 0.75: 1). The reaction mixture was subjected to ultrasonic irradiation for 1 h at room temperature. The TMC‐coated magnetic NPs were mixed with 5 ml 25% glutaraldehyde solution for 5 h at 50 °C under stirring. After cross‐linking, the products were washed several times with water by a permanent magnet. The final precipitate was dried in an Avon at 120 °C for 30 min. Finally, the prepared 12 nm Au colloids (∼6.51 × 10¹² particles/ml [30]) were added to the TMC magnetic NPs under stirring at room temperature. Au/TMC/Fe₃ O₄ nanocomposites formation could be performed through electro statical assembling of Au on the surface of TMC/Fe₃ O₄ nanocomposite [14]. The pinkish red solution turned colourless by increasing the reaction time. The resultant product was rinsed twice with water, through magnetic separation method, and stored in a dark glass bottle containing 1 ml of phosphate buffer saline (PBS, 0.1 M pH 7.2) at 4 °C before use.

2.2.5 Conjugation of monoclonal antibody (mAb) to Au/TMC/Fe₃ O₄ nanocomposites

Hybridoma producing HSA‐specific mouse mAb were generated by fusion of Sp2/0 myeloma cells with mouse spleen cells from mice immunised with HSA antigen as previously described [31].

A suspension containing 20 mg of Au/TMC/Fe₃ O₄ nanocomposites was initially sonicated for 10 min in 1 ml PBS 0.1 M. The dispersion was resuspended in 500 µl of PBS and the pH of the suspension was brought to 9. For bioconjugation, 38 µl of mAb (40 µg/ml) was added and incubated for 3 h at 37 °C, with light rotating. Subsequently, BSA 1% was added 30 min at 37 °C for blocking. Ab–Au/TMC/Fe₃ O₄ nanocomposites were then collected by external magnetic separation, and stored in 1 ml of PBS containing 0.01% sodium azide at 4 °C [8].

2.2.6 Fabrication of the electrochemical immunosensor

Fig. 1 schematically shows the processes for electrochemical immunosensor detection of HSA. To design the immunosensor, polyvinyl alcohol (PVA) 2% was used for antigen stabilisation and also electrode surface modification, according to our previous reports [8, 9]. For this purpose, 5 µl of VA containing 500 ng/ml HSA was dropped on the surface of the electrode and polymerisation of VA was carried out under ultraviolet (UV) conditions (220–250 nm, 20 min). After the PVA–HSA coated electrode was dried, it was blocked with 15 µl of phosphate buffer (PB, 0.1 M pH 7.2) containing 1% skim milk and tween 20 for 15 min at 37 °C to eliminate non‐specific binding. The electrode was then washed once by 20 μl PBS. Finally, 20 µl of the prepared Ab–NPs solution and 5 µl of standard HSA or real sample incubated for 20 min, then dropped onto the modified electrodes surface and incubated for another 15 min. Subsequent to accumulation step, the electrode was washed three times by PBS to remove the unbounded Ab–Au/TMC/Fe₃ O₄ from the electrode surface.

Fig. 1.

Fabrication of the electrochemical immunosensor

a Schematic representation of the synthesised Au/TMC/Fe₃ O₄ NPs

b Steps for processing standard/samples with Ab–Au/TMC/Fe₃ O₄ NPs

c Preparation procedure of the ultrasensitive electrochemical immunoassay: (a) immobilisation of the PVA–HSA onto the working electrode of screen printed carbon electrode (SPCE), (b) skim milk blocking, (c) incubation of the standard/samples with Ab–Au/TMC/Fe₃ O₄ NPs on the SPCE, (d) Au/TMC/Fe₃ O₄ NPs preoxidation, and (e) signal recording by the voltammetric modes

2.2.7 Measurement protocol

The analytical measurements of electrochemical signals were recorded by cyclic voltammetry (CV) and DPV. The CV was made between −1 and +1 V at a scan rate of 50 mV/s. DPV was performed at the potential range from −0.2 to +0.8 V with a constant potential of 1.27 V for 30 s. The DPV measurements carried out in the presence of standard or biological samples.

3 Results and discussion

The type of tracer is critical for successful development of user‐friendly assays with adequate sensitivity and accuracy. Here, the sensitivity of the electrochemical immunosensor was enhanced using Au/TMC/Fe₃ O₄ NPs as a tracing tag to label Ab. In fact, remarkable conductivity of AuNPs provides a potential pathway for electron transfer that raises sensitivity and specificity for detection.

3.1 Characterisation of Au/TMC/Fe₃ O₄ nanocomposites

The size of individual Fe₃ O₄ as well as TMC/Fe₃ O₄ and Au/TMC/Fe₃ O₄ composite was characterised by TEM (Zeiss Model, 100 kV, Germany). Figs. 2 a and b show the TEM images of Fe₃ O₄ and TMC‐coated Fe₃ O₄, with approximate size of 10 ± 1 nm and 12 ± 1 nm, respectively. The images of Fe₃ O₄ NPs proved that the particles were monodisperse (Fig. 2 a) and they were successfully capped (Fig. 2 b). As shown in Fig. 2 c, increasing the diameter of the synthesised composite to 20 nm confirms the presence of AuNPs on the surface of TMC/Fe₃ O₄ NPs.

Fig. 2.

Characterisation of Au/TMC/Fe3O4 nanocomposites

a Transmission Electron Microscopy (TEM) images of naked Fe₃ O₄

b TMC/Fe₃ O₄

c Au/TMC/Fe₃ O₄ NPs

The effect of electrostatic interaction on the formation of Au/TMC/Fe₃ O₄ was further investigated by zeta potential measurement (Malvern Zetasiser 3000 photon correlation spectroscopy (PCS) system, Southborough, UK).

Table 1 shows the zeta potential of NPs, the zeta potential of freshly prepared TMC/Fe₃ O₄ NPs was in a positive range. However, after incubation with AuNPs, the zeta potential dropped to negative values in the range of −30.2 to −33.4 mV, indicating the attachment of anionic AuNPs to the cationic surface. The higher positive zeta potential of TMC‐coated Fe₃ O₄ NPs, compared with chitosan coated Fe₃ O₄ NPs, can be explained by the permanently positive charged sites in the TMC polymer. The comparable z values of NPs confirmed successful coating of Fe₃ O₄ NPs with TMC and AuNPs. The high zeta potential values elucidate the stability of the NP suspensions in water or buffer solution [21].

Table 1.

PCS analysis of NPs

NPs	Zeta, mV	Polydispersity Index (PDI)
Fe3 O4	−42.7 ± 0.82	0.08 ± 0.02
chitosan/Fe3 O4	+18.1 ± 1.51	0.45 ± 0.1
TMC/Fe3 O4	+41.2 ± 3.51	0.11 ± 0.03
Au/chitosan/Fe3 O4	−5.2 ± 0.81	0.51 ± 1.03
Au/TMC/Fe3 O4	−31.8 ± 1.61	0.19 ± 0.05

Fig. 3 a shows the infrared spectrums (Thermo Nicolet Nexus 870 Fourier transform infrared spectroscopy (FTIR), USA) of the magnetite, magnetic TMC and AuNPs coated magnetic TMC. The spectra clearly indicated the presence of TMC in the hybrid samples. The peak about 3430 cm⁻¹ observed in TMC, magnetic TMC and AuNPs coated magnetic. TMC curves indicated the combined O–H stretching and intermolecular hydrogen bonding vibrations, while a band near 1470 cm ^{− 1} observed for TMC's spectra indicated asymmetrical stretching of C–H in the methyl groups. The C–H stretching vibrations of the polymer were clearly seen around 2920 and 2850 cm⁻¹ for all samples. The absorption peak at 1629 cm⁻¹ came from the C=O amide stretching. All spectra showed a vibration near 1470 cm⁻¹ and also a broad absorption peak at 570 cm⁻¹, indicating the quaternised ammonium structure at the C‐2 position in the TMC [21] and formation of the Fe₃ O₄ framework, respectively. Since the surface of Fe₃ O₄ with negative charges has an affinity to TMC, protonated TMC could coat the magnetite NPs by the electrostatic interaction and chemical reaction through glutaraldehyde cross‐linking.

Fig. 3.

Infrared spectrums and magnetisation

a FTIR spectra of naked Fe₃ O₄, TMC, TMC/Fe₃ O₄ and Au/TMC/Fe₃ O₄ NPs

b VSM of naked Fe₃ O₄ and TMC/Fe₃ O₄ and Au/TMC/Fe₃ O₄ NPs

Fig. 3 b compares the magnetisation curves of three sample sets measured at room temperature (vibrating sample magnetometer (VSM), MAG‐3110, Freescale). The magnetisation results interestingly showed that the saturation magnetisation (M _s) was 66.52 emu/g, whereas the remanence and coercivity were zero. By self‐assembly of TMC and AuNPs on the surface of the Fe₃ O₄, the saturation magnetisation declined to 58.22 and 55.80 emu/g, while the superparamagnetism behaviour maintained. The magnetic measurements were aligned with the figures reported in previous studies [16, 27, 32]. Moreover, while the external field was removed, the remanence magnetisation of the Au/TMC/Fe₃ O₄ was zero, suggesting superparamagnetic behaviour of the Au/TMC/Fe₃ O₄ NPs. Supermagnetism in magnetic NPs prevents them from aggregation and allows the particles to be redispersed quickly as soon as the magnetic field is removed. This behaviour is important for their biomedical and bioengineering applications [33].

The reaction time between Au and magnetic TMC and chitosan plays a key role in the development of Au/TMC/Fe₃ O₄ and Au/chitosan/Fe₃ O₄ NPs. The number of AuNPs coated on the TMC/Fe₃ O₄ and chitosan/Fe₃ O₄ was evaluated at different reaction times by UV–visible absorption spectra (UV–vis spectrophotometer, Perkin Elmer, USA). The freshly prepared AuNPs solution showed maximum absorption at about 520 nm. However, after an initial incubation of 30 min with TMC/Fe₃ O₄, the absorbance value decreased significantly and the pinkish red solution of AuNPs turned colourless, indicating that ∼99.9% are coated on the TMC/Fe₃ O₄ (Fig. 4 a)_. On the other side, the absorbance value for chitosan/Fe₃ O₄ decreased gradually after 2 h incubation, tending to a steady‐state value for chitosan/Fe₃ O₄ (Fig. 4 b). The findings pointed the strong adsorption ability of TMC/Fe₃ O₄ for AuNPs, compared with chitosan/Fe₃ O₄ nanocomposite.

Fig. 4.

Evaluated the number of AuNPS coated on the TMC/Fe₃ O₄ and chitosan/Fe₃ O₄ at different reaction times by UV spectroscopy

a UV–vis adsorption spectra of AuNPs onto the TMC/Fe₃ O₄

b Chitosan/Fe₃ O₄ with different reaction time

3.2 Mechanism of signal amplification strategy

After conjugation of Ab between two different NPs, supernatants of Ab–Au/TMC/Fe₃ O₄ and Ab–Au/Chi/Fe₃ O₄ nanocomposites were separately examined by spectrophotometry. It was determined that bioconjugation amounts of Ab on Au/TMC/Fe₃ O₄ and Au/Chi/Fe₃ O₄ nanocomposites are about 80–90 and 50–60%, respectively.

Fig. 5 compares the signal amplification of two kinds of immunosensors, adopting two different NPs as label. The change in current responses (ΔI) was used to evaluate the effect of the labels in signal amplification strategies, according to the ΔI = I − I ₀ equation. The background noise was recorded as I ₀ toward zero analyte, whereas the cathodic current responses were recorded as I toward 500 ng HSA, according to different types of labels. As shown in Figs. 5 a and b, an obvious ΔI of 3 µA and 28 was achieved, while the working electrode of immunosensors were immobilised with 500 ng HSA and then incubated with the Ab–Au/chitosan/Fe₃ O₄ and Ab–Au/TMC/Fe₃ O₄ magnetic NPs labels, respectively.

Fig. 5.

Signal amplification strategy for sensitive electrochemical immunoassay by using

a Au/chitosan/Fe₃ O₄ NPs

b Au/TMC/Fe₃ O₄ NPs as tracing tag

Fig. 5 illustrates an approximately eight‐fold higher peak current of electrochemical immunosensor label with Au/TMC/Fe₃ O₄ nanocomposites, compared with the label without TMC. This can be explained by no aggregation and precipitation of nanocomposite particles on the electrode surface. Additionally, its smaller size and more importantly, the higher positive charge of TMC/Fe₃ O₄ particles with the capability of absorbing a greater number of negative charge electroactive probes, such as AuNPs, generates a more current signal on the immunosensor. Therefore, the rest of the procedure was carried out with the Au/TMC/Fe₃ O₄ NPs as label and Au/chitosan/Fe₃ O₄ was omitted in the subsequent experiments.

3.3 Electrochemical activity of immunosensor

The process of electrode modification can be effectively probed by the CV method. As expected, no detectable redox reaction signal was observed in the cyclic voltammograms of bare, PVA and HSA–PVA modified electrodes (Fig. 6 a) [8]. To detect non‐specific reactions, 20 µl of NPs solution including Au/TMC/Fe₃ O₄ NPs was added to the modified HSA–PVA electrode, and the current response was then calculated. Owing to the non‐specific interaction between NPs and PVA, a low background current was observed. After the immunoreactions, a significant increase in redox peak currents was recorded (Fig. 6 a curve e) as high‐content AuNPs on the label were electro‐oxidised to produce AuCl₄ ⁻, which in turn enhanced the electron transfer [8, 30]. The oxidation potential has high effect on the reduction peak current of AuCl₄ ⁻ ions since very high potential will damage the SPCE and a low potential will not be able to oxidise Au to AuCl₄ ⁻. This parameter was tested with DPV mode, after the preoxidation of immunosensor in 1 M HCl at a constant time of 40 s. The peak current increased with the increasing preoxidation potential from +1.25 to +1.40 V and reached the maximum value at +1.27 V (Fig. 6 b). Therefore, in this experiment, +1.27 V (against silver (Ag)/AgCl) was chosen as the best preoxidation potential. The oxidation time can also greatly influence the reduction peak current. Inadequate oxidation time is assumed to cause incomplete oxidation of the colloidal Au, whereas the redundant time would result in diffusion of the AuCl₄ ⁻ ions formed in the solution phase. Therefore, the DPV measurements were carried out at different times ranging from 20 to 90 s. The reduction of AuNPs signals was augmented by increasing time, reaching their highest level at 30 s and was stable thereafter. About 30 s was selected as the best preoxidation time, which demonstrated the feasibility of rapid detection, as well (Fig. 6 c).

Fig. 6.

Electrochemical activity of immunosensor

a Cyclic voltammograms of the various modified electrodes evaluated in 1 M HCl at the potential range of −1 to +1 V and the scan rate of 50 mV/s: (a) bare electrode, (b) PVA, (c) PVA–HAS, (d) PVA–HSA–Au/TMC/Fe₃ O₄ and (e) PVA–HSA–Ab–Au/TMC/Fe₃ O₄

b Remarks the measurement of preoxidation potential (1.25, 1.27, 1.3 and 1.4 V)

c Times (20, 30, 40, 60, 80 and 90 s) on DPV response to 500 ng/ml HSA at the potential range

Various parameters, including the concentration of coating HSA, Ab–Au/TMC/Fe₃ O₄ NPs suspension, incubation time and temperature, were also characterised in order to establish the optimal conditions for determination of analyte (HSA). To this end, first, the effects of different quantities of Ab–Au/TMC/Fe₃ O₄ NPs suspension (10, 15, 20 and 25 μl) and coated antigen (100, 200, 400, 500 and 800 ng) were investigated. The results revealed that with an increase in the amount of Ab–Au/TMC/Fe₃ O₄ NPs suspension and the coated antigen concentration, the signal response increased and reached the maximum level at 20 μl and 500 ng, respectively.

In addition, other factors such as incubation time and temperature influencing the immunoreaction of the assay were studied by CV method. In this way, different periods of time from 20 to 60 min and temperature (4, 25 and 37 °C) were set to obtain the optimum condition. The maximum generated currents achieved at 30 min incubation at 37 °C. In the present paper, the total time taken to complete the assay was the same as our previous reports [8], ∼80 min, whereas the other approaches needed longer time [23, 34].

3.4 Performance of the immunosensor

3.4.1 Sensitivity and detection range of the immunosensor

Performance of the proposed immunosensor was monitored by detection of HSA in standard samples, based on the developed protocol under the optimal conditions. Serial concentrations of standard solutions were prepared by diluting the known concentrations of human albumin with normal urine samples to obtain the final concentrations of 0.25–1000 pg/ml. Then, competition analysis experiments were performed in triplicates for each concentration by DPV method. Fig. 7 a illustrates the expected competitive mechanism, through which the intensity of cathodic peak currents would decrease with increasing possibility of interaction between free HSA in standard or real samples with conjugated Ab–Au/TMC/Fe₃ O₄, compared with the immobilised antigen and vice versa. In the present paper, significant technical advantages obtained from concurrent oxidation of AuNPs at high potential and denaturation of the biomolecules in highly acidic condition. First, the detachment of the possible biocomposite surface presented such a large electro active area for the oxidised Au ions to reduce back effectively through the DPV method. Second, the loss of the oxidised Au ions by diffusion can be prevented as a result of the negative charge of the chelated compounds with the high amount of chloride ions in the acidic medium. Finally, the negatively charged AuNPs were attracted by constant application of highly positive voltage, resulted in the promotion of electrode position on the PVA modified electrode.

Fig. 7.

Sensitivity and detection range of the immunosensor

a Differential pulse voltammograms for the electrochemical assay of HSA on serial dilutions of antigen ((a) 0, (b) 0.25, (c) 0.5, (d) 1, (e) 25, (f) 50, (g) 75, (h) 100, (i) 125, (j) 200, (k) 500 and (l) 1000 pg/ml) in 1 M HCl at scan rate of 50 mV/s

b Calibration curve of the different concentrations of HSA target in urine. Each data peak is the average of three replicates

Amplified current signal generated by the reduction of the Ab–Au/TMC/Fe₃ O₄ NPs tagged was recorded as an electrochemical response. The dose–response curve obtained in the range of 0.25–1000 pg/ml indicated the quantitative measurement of HSA by the DPV signals (Fig. 7 b).

The limit of detection at a signal‐to‐noise ratio of 3σ was 0.2 pg/ml (where σ is the standard deviation of the signal in a blank solution). Additionally, an analytical performance compared between the studied immunosensor and some other similar immunosensors based on the different signal amplification strategies is shown in Table 2.

Table 2.

Comparing study between the present immunosensor with other similar immunosensors based on the different signal amplification strategies by using NPs as tracing tag

NPs	Detection range	Detection limit	References
CNS/AuNPs	0.01–10 ng/ml	0.009 ng/ml	[13]
AuNRs	0.5–500 pg/ml	0.032 pg/ml	[11]
Pt‐Fe3 O4 NPs	0.05–18 ng/ml	15.3 pg/ml	[12]
AuNPs	2.5–200 µg/ml	25 ng/ml	[8]
AuNPs	0.5–200 µg/ml	1 ng/ml	[9]
core/shell Fe3 O4/ AuNPs	0.5–5 ng/ml	0.05 ng/ml	[4]
Au/TMC/Fe3 O4 NPs	0.25–1000 pg/ml	0.2 pg/ml	present work

3.4.2 Specificity, reproducibility and stability of the immunosensor

To estimate the specificity of the developed immunosensor, the influence of possible interference from other molecules on the response of the developed immunosensor was also investigated. A solution of 500 pg/ml HSA containing urea, creatinine, human immunoglobulin, glucose, tetracycline and acetaminophen was prepared. The immunosensor was then separately incubated with all of the prepared compounds at a concentration well over the maximum levels that might be encountered in the clinical samples. The peak current values of 500 pg/ml HSA solution with and without interference substances were similar (I _p = 240 ± 5), indicating the specificity of the immunosensor (Fig. 8 a).

Fig. 8.

Specificity, reproducibility and stability of the immunosensor

a Differences in signal intensities toward various types of interference compounds, which were generated by oxidation of bound Au/TMC/Fe₃ O₄ and subsequent reduction of AuCl⁻⁴ in DPV mode. Interference study was performed using a solution of HSA (500 pg/ml) containing (a) urea (0.33 mol/l), (b) creatinine (3.5 gr/l), (c) human haemoglobin (3.5 mg/l), (d) glucose (7 gr/l), (e) tetracycline (67 mg/l), (f) acetaminophen (67 mg/l)

b Thermal stability of the immunosensor

The reproducibility of the immunosensor was investigated by five intra‐assays of the HSA at a concentration of 0.5 and 1000 pg/ml. All electrodes showed close electrochemical responses; and the figures for relative standard deviations (RSDs) were 4.5 and 1.6% representing acceptable reproducibility of the proposed immunosensor. The thermal stability of the immunosensor was examined over a period of 15 days, based on our previously reported protocol (Fig. 8 b) [8, 35]. No obvious changes occurred after storage for 4 days at 37 °C. However, the stability of the sensor started to decrease slowly, with 91% of biosensor activity being presented on day 15. This could be possibly justified by the fact that the PVA polymer and Au/TMC/Fe₃ O₄ NPs provide a suitable microenvironment for the immobilisation of HSA and Ab molecules to keep their activity.

3.4.3 Response characteristics of the immunosensor

The reliability of the immunosensor was also achieved by analysing ten unknown urine samples and comparing the results with those obtained via a homemade chemiluminometric immunoassay (CLIA) method. The RSDs estimated from three repeated measurements for diabetic and control samples were <6.2% (Table 3). The results indicated that the proposed immunosensor can be used for detection of biomarkers in clinical samples.

Table 3.

Comparison of HSA concentrations in samples measured by our immunosensor and CLIA method

Sample	Methods		RSD, %
	Immunosensor, pg/ml	CLIA, pg/ml
1	979	984	0.3
2	861	855	0.4
3	652	660	0.8
4	424	445	3.4
5	256	261	1.3
6	156	162	2.6
7	76	83	6.2
8	0.43	≤10−5	–
9	0.35	≤10−5	–
10	0.3	≤10−5	–

CLIA. Data are presented as the means of three replicates. The samples were diluted by buffered solutions in order to obtain pg/ml of the samples.

3.4.4 Accuracy

The accuracy parameter was determined by the recovery test, which consisted of adding known amounts of HSA into the sample solutions. This test was conducted by three different concentrations (0.5, 500 and 1000 pg/ml) of test sample in three replicate sample preparations, and the per cent recoveries (mean ± %RSD of three replicates) of HSA in urine sample were calculated. The recovery results are listed in Table 4, which shows acceptable results with RSD values ranging between 1.04 and 6.7%. The HSA recoveries are between 98.08 and 110%, which clearly indicate the potentiality of the present immunoassay for analytes detection in real samples.

Table 4.

Spike and recovery results (n = 3) obtained from HSA immunosensor in urine sample

HSA added, pg/ml	HSA founded, pg/ml	RSD, %	Recovery
0.5	0.55 ± 0.09	6.73	110
500	490.4 ± 7.9	1.37	98.08
1000	985.3 ± 12.17	1.04	98.53

4 Conclusion

This paper presents a new competitive electrochemical immunoassay, combining a disposable chip with the Au/TMC/Fe₃ O₄ NPs tracing tag to further enhancements in signal amplification, magnetism separation as well as background signal diminution which is another critical point in the development of biosensing devices. Chemical modification of chitosan was adopted to create functionalised TMC with membrane‐forming ability for Fe₃ O₄ NPs, providing more space with positive charge for the attachment of AuNPs on the surface of TMC/Fe₃ O₄ NPs. This can be a convincing reason for the increased current response of Au/TMC/Fe₃ O₄ NPs in electrochemical sensing. More significantly, the newly designed label presented a high reductive peak current, compared with our previous label and other reported electrochemical labels. The immunosensor demonstrated a high dynamic range and sensitivity within femtogram/ml levels.