Carcinoembryonic antigen (CEA), alpha-fetoprotein (AFP), carbohydrate antigen 125 (CA 125), carbohydrate antigen 19–9 (CA 19–9), and fetal bovine serum were purchased from Beijing KeyGen Biotech Co., Ltd. (Beijing, China). Amino-modified CEA aptamer (5′-ATACAGCTTCAATT-NH₂-3′) (Yan et al., 2023) was purchased from Sangon Biotechnology Co., Ltd (Shanghai, China). Prostate-specific antigen (PSA) was procured from Beijing Biodragon Immunotechnologies Co., Ltd. (Beijing, China). Tetraethyl orthosilicate (TEOS, 98%), cetyltrimethylammonium bromide (CTAB), bovine serum albumin (BSA), glutaraldehyde (GA), sodium dihydrogen phosphate dihydrate (NaH₂PO₄·2H₂O), disodium hydrogen phosphate dodecahydrate (Na₂HPO₄·12H₂O), sodium hydroxide (NaOH), potassium ferricyanide [K₃Fe(CN)₆], and potassium ferrocyanide [K₄Fe(CN)₆] were purchased from Aladdin Bio-Chem Technology Co., Ltd. (Shanghai, China). 3-Aminopropyltriethoxysilane (APTES) and potassium hydrogen phthalate (KHP) were purchased from Shanghai McLean Reagent Co., Ltd (Shanghai, China). Sulfuric acid, acetone, anhydrous ethanol (99.8%), and concentrated hydrochloric acid (HCl, 36%–38%) were obtained from Shuanglin Reagent Co., Ltd. (Hangzhou, China). Sodium nitrate (NaNO₃) was ordered from Hangzhou Gaojing Fine Chemical Co., Ltd. (Hangzhou, China). Screen-printed carbon electrodes (SPCEs) were purchased from Metrohm (Bern, Switzerland).

In brief, SPCEs contain three integrated electrodes, namely, a working electrode (4 mm diameter), a counter electrode made up of conductive graphite paste, and an Ag reference electrode comprising of conductive silver paste. Phosphate-buffered saline (PBS, 0.01 M, pH = 7.4) was prepared using NaH₂PO₄·2H₂O and Na₂HPO₄·12H₂O. All the aqueous solutions used here were prepared using ultrapure water (18.2 MΩ cm) from Milli-Q Systems (Millipore Inc., Massachusetts, America). All chemical reagents were of analytical grade.

The morphology and thickness of NH₂-VMSF were characterized using transmission electron microscopy (TEM, model HT7700, Hitachi, Tokyo, Japan). To prepare TEM samples, the NH₂-VMSF layer was carefully scraped off the electrode using a scalpel and dispersed in anhydrous ethanol with subsequent ultrasonic dispersion. Then, the resulting dispersion was drop-cast onto a copper grid. Before morphology characterization under 200 kV, the sample was dried naturally. All electrochemical experiments, including cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS), and differential pulse voltammetry (DPV), were conducted on an Autolab electrochemical workstation (model PGSTAT302N, Metrohm Autolab, Switzerland). The frequency range for EIS measurements was from 0.1 Hz to 100 kHz, with a perturbation amplitude of 5 mV.

NH₂-VMSF was grown on the surface of a p-SPCE electrode using the EASA method, as reported in the literature (Ma et al., 2022b). Bare SPCE was electrochemically polished by CV scanning in diluted H₂SO₄ (0.05 M) 10 times at a potential of 0.4 V–1.0 V. Then, the electrode was thoroughly washed with ultrapure water and dried with nitrogen. After that, the cleaned SPCE was subjected to electrochemical polarization. Specifically, a constant potential of +1.8 V was applied to SPCE for 300 s to perform anodic oxidation, followed by cathodic polarization in PBS (0.1 M, pH = 5) scanning from −1.3 V to +1.25 V. The resulting electrode is called p-SPCE.

To grow NH₂-VMSF on the p-SPCE, a mixture of 20 mL ethanol, 20 mL NaNO₃ solution (0.1 M, pH = 2.36), CTAB (1.585 g), and APTES (318 μL) was prepared. The pH of the solution was adjusted to 2.97 using HCl before TEOS (2,732 μL) was added. Then, the solution was vigorously stirred and reacted for 2.5 h to obtain the precursor solution. Subsequently, the p-SPCE electrode was put into the precursor solution and subjected to the constant current (current density: −0.74 mA/cm², duration: 10 s) to grow NH₂-VMSF on the p-SPCE surface. The resulting electrode was washed with ultrapure water, dried with nitrogen, and aged at 80°C overnight to obtain the p-SPCE modified with a hybrid film consisting of surfactant micelles (SMs) and NH₂-VMSF, which was named SM@NH₂-VMSF/p-SPCE. Finally, the SM@NH₂-VMSF/p-SPCE electrode was immersed in an HCl-ethanol solution (0.1 M) and stirred for 5 min to remove SMs from silica nanochannels, yielding NH₂-VMSF/p-SPCE with open nanochannels.

NH₂-VMSF/p-SPCE was used to construct an electrochemical aptasensor for CEA detection by using GA as a cross-linking agent, finally generating the amino groups on the outer surface of NH₂-VMSF and further immobilizing CEA-specific aptamer (Apt). Specifically, to modify the GA only on the outer surface of NH₂-VMSF, the amino groups on the outer surface were first derivatized with aldehyde groups before removing the SMs. A 1% GA solution was dropped onto the SM@NH₂-VMSF/p-SPCE and incubated at 37°C in the dark for 20 min. After thorough washing, the electrode was soaked in a 0.1 M HCl-ethanol solution and stirred for 5 min to remove SMs. The resulting electrode was denoted as the GA/NH₂-VMSF/p-SPCE. Next, a 100 mM Apt solution in 0.01 M PBS (pH 7.4) was dropped onto the GA/NH₂-VMSF/p-SPCE carrying aldehyde moieties and incubated at 4°C for 90 min. The electrode was then washed with 0.01 M PBS (pH 7.4) to remove any unbound Apt to obtain Apt/GA/NH₂-VMSF/p-SPCE. A 0.1 wt% solution of BSA in 0.01 M PBS (pH 7.4) was used to incubate the Apt/GA/NH₂-VMSF/p-SPCE at 4°C for 30 min to block any nonspecific binding sites. The resulting aptasensor was denoted as BSA/Apt/GA/NH₂-VMSF/p-SPCE.

The BSA/Apt/GA/NH₂-VMSF/p-SPCE aptasensor was incubated with different concentrations of CEA at 4°C for 60 min. The detection solution was a 0.1 M KCl solution containing 1.25 mM Fe(CN)₆ ^3−/4−. Electrochemical signals of Fe(CN)₆ ^3−/4− were determined by the BSA/Apt/GA/NH₂-VMSF/p-SPCE aptasensor before and after CEA binding using DPV. The standard addition and recovery methods were used to analyze the real samples of human serum from healthy adults without complicated pretreatments. The received human serum provided by healthy volunteers was diluted 50 times with PBS (0.01 M, pH = 7.4), and then various known amounts of CEA were added in turn to obtain real samples for detection.

Combining the advantages of carbon electrodes and screen-printed electrodes, SPCE has the advantages of good chemical stability, a wide potential window, low cost, and easy mass production. As for the conductive property of SPCE, NH₂-VMSF can be grown on SPCE using the EASA method but lacks stability. Therefore, prior to the fabrication of NH₂-VMSF, bare SPCE undergoes an electrochemical pre-activation procedure (Scheme 1). As displayed, oxygen-containing functional groups appear at the p-SPCE and can form chemical bonds with silanol groups of NH₂-VMSF to effectively increase the stability of NH₂-VMSF on the p-SPCE. In addition, abundant defects presented in the pre-activation operation can boost the electrochemical activity of SPCE. NH₂-VMSF directly obtained from the EASA method remains on the surfactant micelles (SMs) within the silica nanochannels and is employed to incubate with glutaraldehyde (GA) cross-linking agent to specifically generate aldehyde groups on the outer surface of NH₂-VMSF. Note that SMs confined in the nanochannels of NH₂-VMSF effectively prohibit the access and modification of GA on the inner nanochannels. After further covalent binding with CEA-specific aptamer (Apt) and blockage of nonspecific binding sites by bovine serum albumin (BSA), SMs are excluded using 0.1 M HCl-ethanol solution to obtain BSA/Apt/GA/NH₂-VMSF/p-SPCE. The detection mechanism for CEA relies on the impeded diffusion of electrochemical probes [Fe(CN)₆ ^3−/4−] to the underlying p-SPCE upon the recognition of CEA. Based on the above principle, the quantitative relation between the decreased electrochemical current signals of Fe(CN)₆ ^3−/4− and CEA concentration can be used to realize the detection of CEA.

Considering the good electrocatalytic ability of p-SPCE, CV curves of bare SPCE and p-SPCE in 0.1 M PBS (pH 5.0) are compared. As seen in Figure 1A, compared with that of bare SPCE, apparent enhancement of changing current is observed at the p-SPCE, suggesting the increased electroactive area of p-SPCE after pre-activation treatment. Moreover, a pair of redox peaks near 0 V appeared at the p-SPCE and were assigned to the electrochemical reaction between hydroquinone and quinone. Figure 1B displays the CV responses of bare SPCE and p-SPCE to 1.25 mM Fe(CN)₆ ^3−/4−. Increased electrochemical redox currents are shown at the p-SPCE, confirming the improved electron transport property of p-SPCE and further allowing the design of highly sensitive electroanalytical sensors.

NH₂-VMSF scraped from p-SPCE was observed by TEM and the results are shown in Figures 2A, B. The top-view TEM image of NH₂-VMSF exhibits many nanopores as bright spots with a pore diameter of 2–3 nm (Figure 2A). The cross-sectional TEM image of NH₂-VMSF shows several nanochannels parallel to each other, with a uniform thickness of 120 nm (Figure 2B). Subsequently, different electrodes including bare SPCE, p-SPCE, SM@NH₂-VMSF/p-SPCE, and NH₂-VMSF/p-SPCE were employed to detect charged 0.5 mM Fe(CN)₆ ³⁻ and Ru(NH₃)₆ ³⁺, and CV curves are shown in Figures 2C, D. As displayed, electrochemical pre-activation of SPCE can result in the increased charging currents and Faradic currents for these two probes, suggesting the increased electroactive area and accelerated electron transport ability of p-SPCE. Only charging currents for two probes are shown at the SM@NH₂-VMSF/p-SPCE, indicating that the SMs inside the nanochannels can block the access of probes and the as-prepared NH₂-VMSF on the p-SPCE is intact. NH₂-VMSF has many open nanochannels, and its inner surface is rich in amino groups. Due to the protonation of amino groups, NH₂-VMSF/p-SPCE exhibits a positively charged surface and electrostatic selectivity for the above two charged probes. By comparing the Faradic currents obtained at the p-SPCE, the Faradic currents for negatively charged Fe(CN)₆ ³⁻ probe are significantly enhanced, and the signals for positively charged Ru(NH₃)₆ ³⁺ probe are decreased. This result indicates that the NH₂-VMSF/p-SPCE with electrostatic effect has the ability to amplify Fe(CN)₆ ³⁻ signals in the following quantitative analysis study. The above results confirm the successful fabrication of NH₂-VMSF on the p-SPCE and are similar to the previous reports (Etienne et al., 2009).

NH₂-VMSF/p-SPCE is a good electrode interface that can support the development of developing highly sensitive and disposable biosensors. To examine the ability of BSA/Apt/GA/NH₂-VMSF/p-SPCE, CEA was used as a model and its corresponding CEA-specific aptamer as a recognition element was immobilized on the outer surface of NH₂-VMSF/p-SPCE. The resulting BSA/Apt/GA/NH₂-VMSF/p-SPCE aptasensor is achieved by stepwise modification of GA, Apt, and BSA (Scheme 1) and is characterized by CV and EIS techniques. As shown in Figure 3, consequent modification of GA, Apt, and BSA leads to decreased redox peak currents and enhanced semicircle diameter corresponding to the charge transfer resistance, which is due to the hindered diffusion of Fe(CN)₆ ^3−/4− toward the electrode. When BSA/Apt/GA/NH₂-VMSF/p-SPCE is incubated with 100 pg/mL CEA, further hindered diffusion for Fe(CN)₆ ^3−/4− occurs and suggests the potential of our proposed aptasensor for quantitative determination of CEA.

Experimental conditions for the construction of BSA/Apt/GA/NH₂-VMSF/p-SPCE can influence the analytical performance of CEA. Therefore, incubation times for the CEA-specific aptamer and target CEA were studied, and the results are shown in Figure 4. As exhibited, anodic peak currents at the BSA/Apt/GA/NH₂-VMSF/p-SPCE aptasensor decrease with the increasing incubation time at the beginning and subsequently tend to approach a plateau. The equilibrium times for the CEA-specific aptamer and target CEA of 90 min and 60 min, respectively, are used in the following study.

The analytical performance of the BSA/Apt/GA/NH₂-VMSF/p-SPCE sensor for CEA detection was evaluated by the DPV method. Several concentrations of CEA were incubated at the BSA/Apt/GA/NH₂-VMSF/p-SPCE surface and tested in a 0.1 M KCl solution containing 1.25 mM Fe(CN)₆ ^3−/4−. As revealed in Figure 5A, anodic peak currents of Fe(CN)₆ ^3−/4− gradually decline with an increase in CEA concentration. By plotting the anodic peak current ratio (△I/I ₀, where △I refers to I − I ₀, and I and I ₀ denote the anodic peak current in the presence and absence of CEA, respectively.) against the logarithm of the CEA concentration, a wide linear range of 100 fg/mL to 100 ng/mL is obtained. The corresponding linear regression equation is △I/I ₀ = 0.086 log C _CEA (pg/mL) + 0.1674 (R ² = 0.997). The smallest discernible DPV signal is obtained by subtracting the standard deviation from the average anodic peak current of three-time blank signals, which is named I, and is substituted into the above linear regression equation to calculate the limit of detection (LOD). The LOD for CEA is estimated to be 24 fg/mL, which is lower than those of electrochemical sensors shown in Table 1 and demonstrates an excellent disposable and sensitive aptasensor.

The selectivity, reproducibility, and stability of our fabricated BSA/Apt/GA/NH₂-VMSF/p-SPCE sensor were also verified, and the data are shown in Figures 5B–D. As presented in Figure 5B, PSA, CA19-9, CA125, and AFP were tested by a BSA/Apt/GA/NH₂-VMSF/p-SPCE sensor in a 0.1 M KCl solution containing 1.25 mM Fe(CN)₆ ^3−/4−, respectively. The anodic peak current variations were compared with those of CEA and a mixture consisting of CEA and these four potential interfering species. The results indicate that the magnitude of anodic peak currents obtained at the BSA/Apt/GA/NH₂-VMSF/p-SPCE remains unchanged in the presence of these four interfering species, implying that the proposed aptasensor enables the selective detection of CEA. Seven parallel BSA/Apt/GA/NH₂-VMSF/p-SPCE aptasensors were used to detect 0.1 ng/mL CEA, displaying comparable electrochemical current variation and showing a good reproducibility of the as-prepared aptasensor (Figure 5C). The prepared BSA/Apt/GA/NH₂-VMSF/p-SPCE without any solution was placed at 4°C and used to detect 0.1 ng/mL CEA on different days. Figure 5D shows that the BSA/Apt/GA/NH₂-VMSF/p-SPCE aptasensor had excellent stability for the detection of 0.1 ng/mL CEA within 7 days.

The fabricated BSA/Apt/GA/NH₂-VMSF/p-SPCE aptasensor was applied to detect CEA in human serum samples. After simple dilution treatment and the addition of a series of CEA samples with known concentrations, artificial human serum samples were obtained and measured by our BSA/Apt/GA/NH₂-VMSF/p-SPCE aptasensor. As shown in Table 2, the detected concentrations (found) by BSA/Apt/GA/NH₂-VMSF/p-SPCE aptasensor are comparable to those known spiked concentrations (added), exhibiting satisfactory recoveries and relative standard deviation (RSD) values.