Fabrication of an Electrochemical Aptasensor Composed of Multifunctional DNA Three-Way Junction on Au Microgap Electrode for Interferon Gamma Detection in Human Serum

Interferon gamma (IFN-γ) is an important cytokine with antiviral, antibacterial, and immunosuppressive properties. It has been used as a biomarker for the early detection of several diseases, including cancer, human immunodeficiency virus (HIV), tuberculosis, and paratuberculosis. In this study, we developed an electrochemical biosensor composed of multifunctional DNA 3WJ to detect IFN-γ level with high sensitivity. Each multifunctional triple-stranded aptamer (MF-3WJ) was designed to have an IFN-γ aptamer sequence, anchoring region (thiol group), and 4C–C (cytosine–cytosine) mismatch sequence (signal generation), which could introduce silver ions. To generate the electrochemical signal, four Ag+ ions were intercalated (3wj b-3wj c) in the 4C–C mismatch sequence. MF-3WJ was assembled through the annealing step, and the assembly of MF-3WJ was confirmed by 8% tris–boric–EDTA native polyacrylamide gel electrophoresis. The Au microgap electrode was manufactured to load sample volumes of 5 µL. The reliability of electrochemical biosensor measurement was established by enabling the measurement of seven samples from one Au microgap electrode. MF-3WJ was immobilized on the Au microgap electrode. Then, cyclic voltammetry and electrochemical impedance spectroscopy were performed to confirm the electrochemical properties of MF-3WJ. To test the electrochemical biosensor’s ability to detect IFN-γ, the limit of detection (LOD) and selectivity tests were performed by square wave voltammetry. A linear region was observed in the concentration range of 1 pg/mL–10 ng/mL of IFN-γ. The LOD of the fabricated electrochemical biosensor was 0.67 pg/mL. In addition, for the clinical test, the LOD test was carried out for IFN-γ diluted in 10% human serum samples in the concentration range of 1 pg/mL–10 ng/mL, and the LOD was obtained at 0.42 pg/mL.


Tuberculosis (TB) is an infectious disease caused by several types of Mycobacterium.
Although TB is as old as human history, it remains an unresolved problem that threatens public health [1,2]. According to the World Health Organization (WHO), in 2019 alone, 1.4 million people died from TB, and about 10 million people were infected with TB [3,4]. The number of people infected with TB is declining by about 2% per year, but the number of new patients is still increasing due to the growing population [5]. The most effective way to manage TB risk is to quickly diagnose active TB patients and provide appropriate treatment to prevent the spread of TB bacteria, thereby preventing the occurrence of new TB patients.
For TB diagnosis, various methods, such as chest radiography [6], Mycobacterium tuberculosis smear [7], M. tuberculosis molecular diagnosis [8], susceptibility to isoniazid and rifampin [9], and tuberculin test (TST), have been carried out [10]. TST, which is used Biomedicines 2021, 9,692 2 of 10 to investigate infection with M. tuberculosis, may produce false-positive results due to the Bacillus Calmette-Guérin vaccine. The interferon gamma (IFN-γ) test was introduced to address this issue [11,12]. The IFN-γ test involves an antigen that stimulates T cells [13] sensitized by M. tuberculosis, and is used to check for infection with M. tuberculosis through an enzyme-linked immunosorbent assay (ELISA) [14,15]. Apart from resolving the falsepositive problem of TST, the IFN-γ test is in the spotlight as a test method that can be used to diagnose latent tuberculosis. While the ELISA-based technique is a commonly used method capable of quantifying IFN-γ in a sensitive and specific manner, production of an antibody is time-consuming and costly. Other disadvantages associated with this method include several washing steps, temperature sensitivity, and difficulty in miniaturization and multiplexing owing to the use of several reagents. An aptamer is a single-stranded nucleic acid that has many advantages over antibodies, such as temperature stability and low manufacturing cost. It is composed of a relatively simple nucleic acid sequence that is easy to modify and has excellent convenience and flexibility in structural design [16][17][18]. In this study, we attempted to insert a redox label for electrochemical detection into an aptamer by utilizing its excellent flexibility.
Electrochemical biosensor [19] is a powerful analysis tool due to its portability, fast response time, and high sensitivity. The miniaturization of modern microelectronics is made possible by constructing microelectrodes that are suitable for detecting small samples [20,21]. The large-scale production and low cost of electronic devices are important reasons that make the electrochemical approach more attractive for high-throughput analysis.
As shown in Figure 1, the detection strategy proposed in this study involved the self-assembly of a multifunctional triple-stranded aptamer (MF-3WJ) on a Au microgap electrode. The change in redox current was quantified using square wave voltammetry (SWV), and it was confirmed that this change was dependent on the concentration of IFN-γ. MF-3WJ consisted of three arms, namely, an IFN-γ aptamer sequence, a thiol group, and a C-C mismatch sequence. Ag + was inserted between the C-C mismatched sequences to stabilize the double strand, and Ag + was used as a redox label [22]. The reliability of the electrochemical biosensor was secured by measuring seven samples on one Au microgap electrode designed to minimize the use of IFN-γ samples.

Assembly of the Multifunctional DNA 3WJ Structure
The MF-3WJ structure consisted of three fragments with each functioning as IFN-γ aptamer sequence, anchoring region using thiol functional group, and electrochemical signal reporter through C-C mismatch sequence. Ag + was inserted into the C-C mismatch sequence and used as an oxidation-reduction label [24]. MF-3WJ was assembled in TMS buffer (50 mM tris, 10 mM MgCl 2 , 100 mM NaCl) by annealing. The annealing process included heating the same molar ratio ssDNA at 80 • C for 5 min and then cooling to 4 • C at a rate of 1 • C/min. The assembled MF-3WJ structure was confirmed using tris-boric-EDTA native polyacrylamide gel electrophoresis (TBE native PAGE). The electrophoresis status was examined using 100 bp DNA ladder (Bioneer Inc., Daejeon, South Korea). Each DNA sample was electrophoresed at 70 V for 60 min [25].

Fabrication/Preparation of the Au Microgap Electrode
The Au microgap electrode [26] was designed by the Korea Advanced Nano Fab Center (Daejon, Korea). Metal deposition of Cr (2 nm) and Au (50 nm) on the SiO 2 was carried out using an electron beam evaporator. The Au microgap electrode was 10 mm wide and 15 mm long and had an area of 150 mm 2 . The working electrode area was 0.09 mm 2 , and the gap between the working electrodes and the counter electrode was 10 µm. For pretreatment of the electrodes, the electrodes were performed for 15 min with ultrasonication in acetone solution. Then, the electrode was sequentially washed with ethanol and distilled water and dried using nitrogen gas. For electrochemical measurements, polydimethylsiloxane (PDMS) was attached to the electrode and used as a chamber. PDMS was prepared by reacting sylgard 184A (DOW, Midland, MI, USA) and sylgard 184B (DOW, Midland, MI, USA) in a 10:1 ratio and drying in an oven at 70 • C for 1.5 h. Figure 2 shows the electrode drawing and preparation process [27]. sonication in acetone solution. Then, the electrode was sequentially washed with ethanol and distilled water and dried using nitrogen gas. For electrochemical measurements, polydimethylsiloxane (PDMS) was attached to the electrode and used as a chamber. PDMS was prepared by reacting sylgard 184A (DOW, Midland, MI, USA) and sylgard 184B (DOW, Midland, MI, USA) in a 10:1 ratio and drying in an oven at 70 °C for 1.5 h. Figure  2 shows the electrode drawing and preparation process [27].

Electrochemical Analysis
All electrochemical experiments were performed on a three-electrode system using an electrochemical workstation (CHI 760E, CH Instruments, Austin, TX, USA). The threeelectrode system consisted of a Ag/AgCl reference electrode (CH Instruments, USA) and a Au microgap electrode as the counter electrode and working electrode. Cyclic voltammetry (CV) was performed in 10 mM HEPES (pH 7.03) and 5 mM [Fe (CN)6] 3−/4− at a scan rate of 0.03 V/s and a voltage range of −0.3 to 0.6 V [28]. Electrochemical impedance spectroscopy (EIS) was performed with a voltage of 0.25 V in the frequency range of 1 Hz to 100 kHz in the same buffer. Square wave voltammetry (SWV) was performed in 10 mM

Electrochemical Analysis
All electrochemical experiments were performed on a three-electrode system using an electrochemical workstation (CHI 760E, CH Instruments, Austin, TX, USA). The threeelectrode system consisted of a Ag/AgCl reference electrode (CH Instruments, USA) and a Au microgap electrode as the counter electrode and working electrode. Cyclic voltammetry (CV) was performed in 10 mM HEPES (pH 7.03) and 5 mM [Fe (CN) 6 ] 3−/4− at a scan rate of 0.03 V/s and a voltage range of −0.3 to 0.6 V [28]. Electrochemical impedance spectroscopy (EIS) was performed with a voltage of 0.25 V in the frequency range of 1 Hz to 100 kHz in the same buffer. Square wave voltammetry (SWV) was performed in 10 mM HEPES (pH 7.03) in the potential range of −0.3 to 0.6 V and at a frequency of 15 Hz and an amplitude of 0.025 V.

Multifunctional DNA 3WJ Feasibility Evaluation
Multifunctional DNA consisted of three different ssDNA sequences. IFN-γ-Apt-3wj-a was an aptamer sequence that specifically bound to IFN-γ, SH-3wj-c was modified with a thiol group to immobilize on the substrate, and 3wj-b/3wj-c had four consecutive cytosine sequences to introduce Ag + . Ag + stabilized the duplex DNA strand and was used as an oxidation-reduction label. Figure 3A shows the expected two-dimensional structure of MF-3WJ. The assembly of the MF-3WJ structure was confirmed using 8% TBE PAGE ( Figure 3B) [25]. The bands in lanes 2, 3, and 4 showed single-stranded DNA of 3wj-a, 3wj-b, and 3wj-c, respectively. Lane 5 showed an assembled MF-3WJ structural band. In lane 5, a new band that was not observed in lanes 2-4 formed on top, indicating that MF-3WJ was assembled.
an oxidation-reduction label. Figure 3A shows the expected two-dimensional structure of MF-3WJ. The assembly of the MF-3WJ structure was confirmed using 8% TBE PAGE (Figure 3B) [25]. The bands in lanes 2, 3, and 4 showed single-stranded DNA of 3wj-a, 3wj-b, and 3wj-c, respectively. Lane 5 showed an assembled MF-3WJ structural band. In lane 5, a new band that was not observed in lanes 2-4 formed on top, indicating that MF-3WJ was assembled.

Comparison of the Electrochemical Performances of the Au Microgap Electrode and Au Substrate
MF-3WJ was immobilized to the prepared Au microgap electrode using a self-assembly process (gold-sulfur bond). The electrochemical capability of the prepared Au microgap electrodes was compared with a Au substrate (2 cm × 0.8 cm, National NanoFab Center, Daejeon, South Korea). Au substrate samples were prepared in the same way as Au microgap electrodes. The electrochemical performance was evaluated through CV and EIS. In the presence of [Fe(CN)6] 3−4− , the following redox reaction occurred between the exposed electrode surface and the electrolyte. Figure 4A,B shows the electrochemical performance of a Au substrate, while Figure  4C,D shows the electrochemical performance of a Au microgap electrode. Current peaks of the redox species [Fe(CN)6] 3−⁄4− were confirmed at 160 and 320 mV in the CV of each

Comparison of the Electrochemical Performances of the Au Microgap Electrode and Au Substrate
MF-3WJ was immobilized to the prepared Au microgap electrode using a self-assembly process (gold-sulfur bond). The electrochemical capability of the prepared Au microgap electrodes was compared with a Au substrate (2 cm × 0.8 cm, National NanoFab Center, Daejeon, South Korea). Au substrate samples were prepared in the same way as Au microgap electrodes. The electrochemical performance was evaluated through CV and EIS. In the presence of [Fe(CN) 6 ] 3−4− , the following redox reaction occurred between the exposed electrode surface and the electrolyte. Figure 4A,B shows the electrochemical performance of a Au substrate, while Figure 4C,D shows the electrochemical performance of a Au microgap electrode. Current peaks of the redox species [Fe(CN) 6 ] 3−⁄4− were confirmed at 160 and 320 mV in the CV of each process ( Figure 4A,C). After immobilization of the MF-3WJ on the substrate, the redox peak current value decreased remarkably. The negative charge of the phosphate groups contained in the DNA backbone interfered with the redox reaction between the electrode surface and the electrolyte, blocking the current. An additional pair of redox peaks were confirmed on the substrate modified with Ag ion. This redox peak occurred when Ag ions intercalated in MF-3WJ exchanged electrons between Ag + and Ag 2+ . EIS is an electrochemical impedance spectroscopy method that has been widely used as an effective and fast method to measure the impedance value of the electrode surface during the frequency change process. In the Nyquist plot, the half-circle diameter refers to the charge transfer resistance (R ct ) of the working electrode surface. Figure 4B,D depicts the Nyquist plots of a Au microgap electrode and Au substrate. EIS was performed in the same buffer as the CV. When comparing bare and modified electrodes with MF-3WJ, the change in R ct was higher in the modified electrodes. This finding confirmed that MF-3WJ DNA was immobilized on the surface. The negatively charged phosphate backbone of MF-3WJ induced larger R ct because it hindered the movement of [Fe(CN) 6 ] 3−⁄4− to the electrode surface. In the substrate modified with Ag ions, the R ct was markedly reduced. In MF-3WJ, it was confirmed that the intercalation of Ag ions can significantly enhance electron transfer, providing more active regions than bare Au, which is known as a good conductive material. resistance (Rct) of the working electrode surface. Figure 4B,D depicts the Nyquist plots a Au microgap electrode and Au substrate. EIS was performed in the same buffer as t CV. When comparing bare and modified electrodes with MF-3WJ, the change in Rct w higher in the modified electrodes. This finding confirmed that MF-3WJ DNA was imm bilized on the surface. The negatively charged phosphate backbone of MF-3WJ induce larger Rct because it hindered the movement of [Fe(CN)6] 3−⁄4− to the electrode surface. the substrate modified with Ag ions, the Rct was markedly reduced. In MF-3WJ, it w confirmed that the intercalation of Ag ions can significantly enhance electron transfe providing more active regions than bare Au, which is known as a good conductive mat rial.  Each process was verified on Au substrates and Au microgap electrodes through CV and EIS. Table 1 shows the comparison of the electrochemical performance of the Au substrate and the Au microgap electrode. ∆ is the difference between the signal in bare and the substrate modified with 3wj+ag. ∆Ipc/area and ∆R ct /area were calculated as signals generated per unit area of the working electrode. When comparing the electrochemical performances of the Au substrate and the Au microgap electrode, the Au microgap electrode showed superior performance.

Electrochemical Biosensor Performance
To evaluate the performance of the fabricated electrochemical biosensor, SWV was performed in 10 mM HEPES. Different concentrations of IFN-γ (10 ng/mL to 1 pg/mL) were used as samples, and Ag 1/2+ intercalated into MF-3WJ was used as an electron mediator in SWV. The SWV signal of the MF-3WJ/Au microelectrode cultured with different concentrations of IFN-γ is shown in Figure 5A. This signal was obtained using artificial IFN-γ samples diluted with PBS. As the concentration of IFN-γ increased, the corresponding peak current gradually decreased. This was because more antigen-antibody immune complexes that act as insulating layers were formed to reduce the signal. Figure 5B shows the linear regression curve with an intercept of 8.4 × 10 −7 ± 1.5 × 10 −7 and a slope of 1.9 × 10 −7 ± 1.2 × 10 −8 (R 2 = −0.98). The linear regression curve was obtained using the peak current value. The limit of detection (LOD) of the electrochemical biosensor using this regression curve was 0.67 pg/mL. In addition, IFN-γ samples diluted in 10% human serum were prepared, and SWV was performed in the same concentration range to confirm that they could be used in artificial IFN-γ samples and actual clinical samples. Figure 5C shows the SWV signal in clinical IFN-γ samples. It was observed that in clinical IFN-γ samples, the corresponding peak current decreased as the concentration of the samples increased. Figure 5D shows the linear regression curve with an intercept of 9.5 × 10 −7 ± 7.2 × 10 −8 and a slope (R 2 = 0.99) of 1.5 × 10 −7 ± 7.4 × 10 −9 . The LOD of the electrochemical biosensor using this regression curve was 0.42 pg/mL. This suggested the possibility of detecting IFN-γ in actual clinical samples. To evaluate the manufactured electrochemical biosensor's selectivity, SWV was performed after reacting different proteins (myoglobin, hemoglobin, BSA, IFN-γ, CRP, TNF-α) at the same concentration of 0.25 µg/mL ( Figure 5E). Results from the selectivity experiment showed a remarkably lower signal for IFN-γ compared with other proteins. Therefore, it was confirmed that the fabricated electrochemical biosensor had high selectivity. Table 2 presents a comparison of the performances of IFN-γ sensors reported in the literature and the electrochemical biosensor proposed in this study.

Conclusions
In this study, we developed an electrochemical DNA aptasensor with a microgap electrode for IFN-γ detection. For minimizing the detection process, the present electrochemical biosensor comprised a multifunctional DNA aptamer as a bioprobe. Moreover, a Au microgap electrode provided minimal sample loading volumes and enabled the seven sample measurements in one electrode. Using the Ag + inserted between MF-3WJs as the redox label, the change in the electrochemical signal of IFN-γ was confirmed through SWV. It was observed that IFN-γ diluted with PBS showed a dynamic range of 1 pg/mL-10 ng/mL, and the LOD was 0.67 pg/mL. IFN-γ diluted with 10% human serum also showed the same dynamic range, and the LOD was 0.42 pg/mL. The proposed electrochemical biosensor has several advantages compared with those reported in previous studies. For example, it could react directly with IFN-γ without the need for multiple labeling and washing processes. In addition, it increases reliability as it can be measured seven times with one electrode. Furthermore, it has a low detection limit and a wide dynamic range. In the future, this electrochemical biosensor platform may be used not only for IFN-γ detection and molecular diagnosis with the low sample volume, but also for the detection of several biomarkers in immunological and cancer research.