Rapid detection of carcinoembryonic antigen by means of an electrochemical aptasensor

Summary

Carcinoembryonic antigen (CEA) is a critical biomarker for identifying colon cancer. This work presents an electrochemical impedance spectroscopy (EIS) based aptasensor for detecting CEA, utilizing a single-stranded DNA (ssDNA) aptamer previously selected and characterized by our research group. The surface of an interdigitated gold electrode (IDE) was successfully functionalized with an 18-HEG-modified aptamer sequence. The developed aptasensor demonstrated high specificity and sensitivity with detection limits of 2.4 pg/mL and 3.8 pg/mL for CEA in buffer and human serum samples, respectively. The optimal incubation time for the target protein was 20 min, and EIS measurements took less than 3 min. Atomic force microscopy (AFM) micrographs supported the EIS data, demonstrating a change in IDE surface roughness after each modification step, confirming the successful capture of the target. The potential of this developed EIS aptasensor in detecting CEA in complex samples holds promise.

Graphical abstract

Highlights

• •A label-free EIS aptasensor for rapid and sensitive detection of CEA was developed
• •Detection limits of 2.4 pg/mL in buffer and 3.8 pg/mL in human serum were attained
• •Incubation time was 20 min, while the EIS measurements took less than 3 min
• •AFM validated successful target protein capture

Bio-electrochemistry; Applied sciences; Sensor system

Introduction

Colorectal cancer (CRC), commonly known as colon and/or rectal cancer, is a severe public health problem since it is the third most common and second most fatal disease in the world.¹ Global CRC incidence is estimated to rise by 60% by 2035, with 2.2 million new cases and 1.1 million deaths occurring yearly.^2,3 CRC is a condition that affects the colon or rectum and is caused by the abnormal growth of glandular epithelial cells in the colon.⁴ Among the most often utilized screening treatments for CRC are endoscopic exams of the large intestine (particularly flexible sigmoidoscopy and colonoscopy) and stool tests such as fecal occult blood test (FOBT). However, the aforementioned screening technologies have tangible shortcomings, such as invasiveness, low specificity and sensitivity, the need for trained personnel, and high cost.⁵ The use of tumor biomarkers, such as carcinoembryonic antigen (CEA), not only aids in cancer detection but also plays a crucial role in cancer screening, prognosis, and early detection of recurrence or spread, helping to overcome this burden. CEA is a cell adhesion glycoprotein with a molecular weight of 70 kDa that increases to 180 kDa when glycosylated. It can be identified in various body samples, with widely used tumor tissue and blood and urine samples predominantly noninvasive.^6,7 Modern advances in electronics, sensor technology, and microfluidics applications have enabled device miniaturization and testing of various analytes.⁸ Thus, CEA can be directly tested using a small device known as an electrochemical biosensor, which can respond continuously, reversibly, and without causing any disruption to a sample. Electrochemical biosensors combine the analytical power of electrochemical techniques with the specificity of biological recognition processes, relying on selective interactions between a target compound and a recognition element (enzyme, antibody, aptamer, tissue, or other biomolecules) to create an electrical signal proportional to the concentration of an analyte.^9,10,11 Most biosensors proposed for CEA detection today predominantly rely on antibodies as biorecognition elements.¹² Aptamers are synthetic single-stranded deoxyribonucleic (ssDNA) or ribonucleic acid (RNA)¹³ molecules used as target identifiers in sensing platforms as well.¹⁴ They offer advantages over the antibodies, such as thermal stability, ease of synthesis and modification, and cost-effectiveness.^15,16,17

Electrochemical impedance spectroscopy (EIS) is an established method for measuring an electrical system’s impedance characteristics, typically sinusoidal, at various frequencies of an applied disturbance.^18,19 The minimal and non-interfering nature of the stimulus sinusoidal voltage with most biorecognition layers is a significant benefit of EIS biosensors.²⁰ EIS is a label-free and sensitive detection technique that allows downsizing²¹ and integrating interdigitated electrodes (IDEs) that consist of two separate arrays of microelectrodes.²² The primary characteristic of an IDE is a gap (distance) between two individual microelectrodes arranged in an interdigitated structure; that is, the gap is less than 50 μm, which is less than the diffusion layer formed on the anode and cathode during the redox reaction. Consequently, the response at the counter electrode will maintain the decreased mediator concentration on the working electrode (WE), which is almost constant if every single microelectrode of an IDE is utilized as both a working and a counter electrode. As a result, the current will reach a steady state right away, and the sensor will display a high electric current.²² They are one of the most used transducers in technical applications, particularly in biological and chemical sensors, because of their low cost, simplicity of manufacture, and great sensitivity.²³

One example of an electrochemical aptasensor is demonstrated in a study,²⁴ where graphene ink and graphene and poly (3,4-ethylenedioxythiophene): poly(styrenesulfonate (PEDOT:PSS) were gradually applied to a paper substrate to create a conductive composite paper electrode. An electrochemical aptasensor for the detection of CEA was developed with a limit of detection (LOD) of 0.45 ng/mL in a buffer and 1.06 ng/mL in a serum. The fabricated paper-based device validated the sensitivity and specificity of electrochemical measurements for CEA in serum samples, employing aptamers for immobilization. Shekari et al.²⁵ developed a sensitive sandwich-type electrochemical aptasensor for CEA detection using a hemin-G4-based signal amplification technique. The electrode surface was modified with nitrogen-doped graphene (NG), gold nanoparticles (AuNPs), and graphene quantum dots (GQDs) that allowed the detection of CEA in human serum samples with an LOD for CEA in a buffer that was found to be 3.2 fg/mL. These studies suggest that enhancing the sensitivity of aptasensors and exploring the development of a label-free EIS aptasensor could significantly reduce detection time, providing a cost-effective approach.

Here, we developed a novel label-free EIS-based aptasensor for detecting CEA in a buffer and human serum. Figure 1 illustrates the workflow of the EIS aptasensor fabrication and the subsequent CEA detection process. The bare IDE surface was first modified using a thiolated CEA aptamer (6) sequence, which our research team previously selected and characterized.²⁶ In the current study, the ssDNA aptamer sequence (6) with a total length of 39 nucleotides underwent modification with the addition of a linker in the form of HS(CH₆)₆-OP(O)₂O-(CH₂CH₂O)₆-5′-TTTTT- aptamer (6) -3′. Following this step, the IDE surface was backfilled with 6-mercapto-1-hexanol (MCH) to act as a co-immobilizing agent, thereby reducing non-specific binding. Using an IDE in this study was primarily due to its integrated electrochemical setup, eliminating the need for additional electrodes. This novel aptasensing technique that is quick, label-free, and user-friendly incorporates a newly designed CEA aptamer and successfully detects CEA in both buffer and serum samples. Validation was also carried out using atomic force microscopy (AFM) and cyclic voltammetry (CV), and the results aligned with the EIS findings.

Figure 1.

A schematic overview of the EIS aptasensor for CEA detection

(A) An IDE’s overall structure consists of gold, two working electrodes, an auxiliary electrode, and a reference electrode, all made of the same material on a glass substrate.

(B) IDE surface functionalization steps and CEA target incubation.

(C) Measurements of EIS signals after each IDE surface modification step and the detection of the target CEA using a potentiostat and a portable computer.

Results and discussion

Electrochemical characterization

After each modification step, the CV and EIS measurements were carried out to characterize the electrode’s interface properties.²⁷ Figure 2 illustrates the EIS and CV results for the IDE at different surface modification stages. Both EIS and CV measurements were performed in the presence of a dissolved 2 mM ferro/ferricyanide [Fe (CN)₆]^3-/4- redox pair, which served as the redox mediator. Figure 2A presents a typical Nyquist plot of IDE surface functionalization steps. The Nyquist plots collect and display the generated solution resistance (R_s), charge transfer resistance (R_ct), and Warburg impedance (W).²⁸ Each point on the Nyquist plot represents an impedance value at a particular frequency. The diameter of the semicircular Nyquist plot often serves as an approximation for the R_ct.²⁹ At the x-axis, impedance was measured at low frequencies on the right side of the plot, while higher frequencies corresponded to impedances plotted on the left.²⁸ CV is a well-known and versatile electrochemical method commonly used to evaluate the redox status of molecular species.^30,31 It measures the current response to a voltage applied to the sample using a WE.³¹

Figure 2.

Characterization of an IDE surface with EIS and CV

(A) A typical Nyquist plot and (B) a cyclic modification voltammogram of a bare IDE surface, a 4 h aptamer incubation, backfilling with MCH, and the target CEA incubation (2 ng/mL). The inset in (A) shows the Randles equivalent circuit employed to model the EIS data, where R_s is the solution resistance, R_ct is the charge transfer resistance, C_dl is the double-layer capacitance, and W is the Warburg element. The experiment was conducted in three biological replicates. All data are shown as the means of ± SEMs.

As can be seen from Figure 2A, the bare IDE surface exhibited a small semicircle (R_ct = 857 Ω). Additionally, in the voltammogram provided in Figure 2B, the highest peak current value was recorded due to the absence of an electron transfer-resistant material. Then, the semicircle steadily increased when CEA aptamer (6) was immobilized at concertation of 5 μM (R_ct = 1716 Ω) as the result of the redox buffer’s negative electroactive ions rejecting the negatively charged phosphate groups of the DNA aptamer, confirming its attachment to the surface,^32,33,34 which is consistent with the CV data, where the cathode peak and anode peak decreased after fixing the aptamer because the negatively charged phosphoric acid backbones of aptamer impede the electron transfer of the redox couple.^27,35 Further blocking with MCH solution resulted in the charge transfer resistance change (R_ct = 5333 Ω), following CV peak currents decrease because MCH could form an additional barrier on the surface of the IDE and confirm the filling of free surface on the IDE surface. Finally, the R_ct increased to 19,457 Ω, when incubated with a 2 ng/mL CEA target protein for 20 min. The peak current depicted in the voltammogram decreased with the CEA immobilization since the hydrophobic layer of the protein could greatly hinder the conductivity.³⁶ Randles equivalent circuit model, as illustrated in Figure 2A inset, was chosen to fit the experimental data. In this model, R_s was connected in series with the double-layer capacitance C_dl and parallel with the R_ct of the surface, along with the inclusion of W to account for diffusion.

EIS aptasensor optimization studies

To ensure the sensitivity and accuracy of the EIS aptasensor for detecting CEA and to conserve reagents, it was necessary to optimize various parameters, including aptamer concentration and protein incubation time. All experimental conditions were the same as described in Section Fabrication of the EIS aptasensor and CEA detection, except aptamer concentration and protein incubation time. The results of the aptamer concentration optimization study are presented in Figure 3A. The response values of R_ct change increased with the concentration of CEA aptamer. Aptamer concentrations of 5 μM or higher resulted in a considerably greater R_ct change response (>20%) compared to 2 μM and 4 μM (0.01 < p < 0.05). However, there was no statistically significant difference between 5 μM and 8 μM concentrations, suggesting that the R_ct change is saturated at 5 μM. Therefore, the most effective aptamer concentration for target protein detection in the present study was determined to be 5 μM. Further, the incubation time for the target CEA was optimized by varying the incubation period while maintaining a constant CEA value (2 pg/mL), as shown in Figure 3B. The data also revealed that the R_ct response is saturated for any incubation duration of 20 min or more. The results showed that a 20-min incubation period was the optimal incubation period for the target protein to achieve an R_ct change response value > 20% (0.001 < p < 0.01). Thus, for all subsequent experiments conducted throughout this work, an aptamer concentration of 5 μM and an incubation period of 20 min were used.

Figure 3.

Optimization of experimental parameters

(A) CEA aptamer concentration optimization study. ∗p ≤ 0.05, ∗∗ 0.001 < p < 0.01.

(B) Target CEA incubation time optimization study, where target protein concentration was 2 pg/mL ∗p ≤ 0.05, ∗∗ 0.001 < p < 0.01. Experiments were conducted in at least three biological replicates. All data are shown as the means ± SEMs.

Analytical performance of the EIS aptasensor

Following the optimization of parameters, target CEA was detected as described in Section Fabrication of the EIS aptasensor and CEA detection. The results of the sensitivity study for CEA diluted in a buffer (phosphate-buffered saline (PBS), pH 7.6, 10 mM) are illustrated in Figure 4A. An increase in the R_ct change was observed upon increasing CEA concentrations (one sample t test; p = 0.0014). The lowest concentration, 0.002 pg/mL, exhibited a 9.34% change in R_ct. At the highest CEA concentration of 2 ng/mL, there was a 45% change in R_ct. Based on the collected data, the R_ct change values showed a linear correlation (R² = 91%) with the concentration of CEA in the range from 0.002 ng/mL to 2 ng/mL, described by the regression equation y = 5.9899x + 4.7077, where x represents CEA concentration, and y represents the R_ct value (Figure 4A inset). All R_ct values were subtracted from the background (PBS, pH 7.6, 10 mM). The developed aptasensor showed a sensitivity with an LOD of 2.4 pg/mL within a detection range of 2 pg/mL to 2 ng/mL in a buffer. The high sensitivity and specificity of the EIS aptasensor developed in this study were primarily attained through the utilization of the CEA aptamer sequence (6), selected and characterized in our previous study.²⁶ Adding linkers and five thymine residues to the aptamer created space between it and the surface, facilitating its folding and enhancing its ability to bind to the protein.^37,38 Ethylene oxide (CH₂CH₂O), a hydrophilic component of the linker, was crucial in extending the aptamer away from the surface monolayer, thereby preventing steric interference with protein binding.^37,38 The IDE geometry, including factors like electrode gap width, could have also affected the sensitivity of the developed impedimetric aptasensor, contributing to its improved sensitivity.

Figure 4.

Analytical performance of the developed EIS aptasensor

(A) Concentration-dependent analysis of the EIS aptasensor for CEA detection in a buffer and serum. The insets show the calibration curve from 0.002 to 2000 pg/mL for buffer and from 2 to 2000 pg/mL for serum. The experiment was conducted in nine biological replicates.

(B) Specificity of the EIS aptasensor. Target CEA and non-target proteins, IL-6, HSA, and S glycoprotein of SARS CoV-2, were tested at 7 pg/mL. The experiment was conducted in three biological replicates. Data are shown as means ± SEMs (∗ 0.01 < p < 0.05). All EIS measurements were recorded in a 2 mM ferro/ferricyanide [Fe (CN)₆]^3-/4- redox couple.

It was crucial to assess the functionality of the aptasensor under conditions that closely resemble real-world scenarios, as the device is intended for serum-based applications. The experimental procedure for detecting CEA spiked in serum mirrored the procedure used for buffer, with the only difference being the substitution of PBS for serum in the CEA dissolution process (Figure 4A). The R_ct change exhibited a linear increase (R² = 96%) (one sample t test; p = 0.0333) across a concentration range from 2 pg/mL to 2 ng/mL (Figure 4A inset), characterized by the regression equation of y = 7.308x–0.2715, where x represents the CEA concentration, and y represents the R_ct change value. The LOD for the EIS aptasensor in detecting CEA in spiked serum samples was determined to be 3.8 pg/mL, achieved through affinity, specificity, and specific modifications incorporated into the aptamer (6) as described earlier for the LOD in the detection of CEA in the buffer. The presence of impurities and other proteins in serum samples could impact the binding of the target protein to the aptamer, resulting in lower R_ct change values. The normal range for CEA in the blood is 0–2.5 ng/mL,^12,39 with levels above 10 ng/mL suggesting a severe disease and levels exceeding 20 ng/mL indicating potential malignancy development.^40,41 Summarizing the data in Table S1, our aptasensor’s LOD aligns with or exceeds that of already published aptasensors with improved detection time (less than 30 min). Therefore, our current approach demonstrates the aptitude to detect CEA within the clinically relevant ranges while being rapid.

In this research, we also evaluated the response of the developed EIS aptasensor to non-target proteins, including human serum albumin (HSA), interleukin-6 (IL-6), and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spike (S) glycoprotein, using the developed EIS aptasensor. The specificity test results are shown in Figure 4B. Unlike the three interfering proteins, the EIS aptasensor exhibited a significant R_ct change response only to CEA. The R_ct change for CEA equaled 12.4%, whereas the R_ct change values for non-target proteins were negative. The experimental procedure was identical to that used for CEA, except for the non-target proteins. We selected these non-target proteins for specific reasons. HSA, a liver-produced globular protein with a molecular weight of 66.5 kDa,^42,43 stands as the most abundant plasma protein (35–50 mg/mL) in blood, maintaining stable blood pH and osmotic pressure.⁴⁴ IL-6, a cytokine that promotes inflammation and plays a role in the growth and development of human cells, also triggers the release of several proteins causing acute inflammation.⁴⁵ Additionally, we included S glycoprotein for specificity testing, which is responsible for binding to the human angiotensin-converting enzyme 2 (ACE2) and initiating viral entry into host cells.^46,47,48,49

Characterization of an IDE surface morphology

AFM is a relatively recent technology for studying local surface features at length scales ranging from submicron to nanoscale.⁵⁰ A cantilever with a molecularly sharp probe at its end, several micrometers in length, is used to trace the topography of the sample and measure forces between the probe and sample with piconewton sensitivity.⁵¹ Because of its ability to image at such tiny scales, compact size, and ease of use, AFM is one of the few instruments capable of characterizing surface characteristics around extremely small structures.⁵⁰ Figures 5A–5E shows the surface morphology of all functionalization phases of an IDE surface, including the bare electrode. Figure 5F depicts the root-mean-square roughness of the IDE surface during all functionalization and detection stages. As can be seen, the surface of the bare IDE in Figure 5A is relatively smooth, with a measured roughness of 1.62 ± 0.04 nm. The roughness increased to 2.16 ± 0.05 nm following aptamer treatment (Figure 5B). However, when the aptamer-treated surface was functionalized with the blocking agent, MCH, the roughness dropped to a value close to the bare electrode, measuring 1.74 ± 0.05 nm (Figure 5C). The reduction was due to MCH filling in the valleys created by aptamers, which also lowers the height difference between the highest and the lowest points on the surface.

Figure 5.

Analysis of surface morphology for bare and modified electrodes

Representative 3D images of a 1 μm² scanned electrode surface for (A) bare electrode, (B) treatment with the aptamer, (C) blocking with MCH, (D) treatment with CEA in PBS, and (E) treatment of CEA in serum. (F) Comparison of roughness for all functionalization steps (N ≥ 30). Micelle area (G) and circularity (H) are statistically compared after each functionalization step (∗ 0.01 < p < 0.05, ∗∗ 0.001 < p < 0.01, ∗∗∗p < 0.001).

Subsequent treatment with CEA in PBS (pH 7.6, 10 mM) (Figure 5D) and serum (Figure 5E), the roughness increased to 2.67 ± 0.07 nm and 3.31 ± 0.16 nm, respectively. These results confirm the successful attachment of protein molecules to the aptamer. Surprisingly, the surface of the CEA in the serum step had a higher roughness, suggesting either that the CEA in serum forms aggregates or that other components in serum bind to the surface-bound aptamers. To validate this hypothesis, a custom-written MATLAB code was used to quantify the size and shape of micelles formed on the electrode surface for each functionalization step, as illustrated in Figure S1. Figure 5G shows the micelle cross-sectional area for all steps, with the CEA in serum having the largest micelle area. Figure 5H demonstrates that micelles from most stages, except for CEA in serum, are somewhat circular because their circularity (C = 4πA/P², where A is the area and P is the perimeter) values are close to 1.

To sum up, we successfully developed a novel, label-free EIS-based aptasensor for rapid and sensitive detection of CEA. Our sensing strategy achieved a 2.4 pg/mL detection limit for CEA within the detection range of 2 pg/mL to 2 ng/mL in a buffer. Furthermore, when applied to human serum, our aptasensor successfully detected CEA with a detection limit of 3.8 pg/mL within a short incubation time of 20 min and a detection time of 2.5 min, highlighting its diagnostic potential. Moreover, the aptasensor demonstrated specific binding to CEA compared to non-target proteins, emphasizing the achieved selectivity resulting from the specificity of the aptamer sequence (6) previously selected by our group and the IDE surface chemistry. AFM analysis of the IDE surface validated our findings. Our aptasensor’s detection limit, as shown in Table S1, is consistent with previously reported aptasensors and offers a promising alternative for detecting colon cancer. Future work will evaluate the aptasensor’s analytical performance in clinical samples and optimize the blocking procedure by employing different solutions.

Limitations of the study

When testing spiked human serum samples, there is a potential for cross-reaction with other proteins. In such cases, gold nanoparticles (AuNPs) can enhance the detection signal. Another approach is to utilize a sandwich assay with our selected aptamers and increase detection sensitivity.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Chemicals, peptides, and recombinant proteins
CEA human	Sigma-Aldrich	Cat# C4835
Nuclease-free water	Sigma-Aldrich	Cat# W4502-1L
PBS	Sigma-Aldrich	Cat# P4417-100TAB
6-mercapto-1-hexanol (MCH)	Sigma-Aldrich	Cat# 725226-1G
Potassium hexacyanoferrate II	Sigma-Aldrich	Cat# P3289
Potassium hexacyanoferrate III	Sigma-Aldrich	Cat# 244023
Tris(2-carboxyethyl) phosphine (TCEP)	Sigma-Aldrich	Cat# C4706-2G
Human serum	Sigma-Aldrich	Cat# H4522
IL-6	Sigma-Aldrich	Cat# SRP3096
HSA	Sigma-Aldrich	Cat# SRP6182-1MG
SARS-CoV-2 (2019-nCoV) Spike RBD recombinant	Sino Biological	Cat# 40592-VNAH
Oligonucleotides
CEA aptamer (6): HS-(CH6)6-OP(O)2O-(CH2CH2O)6-5′-TTTTT-GCCAGCGAG TTTTGACCGTTTTTCTCTCTTTTC CGCCTA-3′	Previously selected and characterized by Yunussova et al.26 and synthesized by Eurogentec	Custom synthesized
Software and algorithms
Microsoft Excel 2010 v.16.52	Microsoft Office	https://www.microsoft.com/en-us/microsoft-365/previous-versions/microsoft-excel-2010
EIS Spectrum Analyzer v.1.0	Research Institute for Physical-Chemical Problems, Belarusian State University	http://www.abc.chemistry.bsu.by/vi/analyser/
Origin Pro 2016 v.b9.3.2.303	OriginLab Corporation	https://www.originlab.com/index.aspx?go=Support&pid=3224
PSTrace	Metrohm DropSens	N/A
GraphPad Prism v.9.1.0	Prism	https://www.graphpad.com/features
JPK NanoWizard 4XP	Bruker	https://www.bruker.com/en/products-and-solutions/microscopes/bioafm/jpk-nanowizard-4-xp-bioscience.html
Gwyddion	Nečas and Klapetek52	http://gwyddion.net
MATLAB	Mathworks	https://www.mathworks.com
MATLAB code to calculate micelle area and micelle circularity	This paper
Other
Super sharp high-resolution silicon AFM cantilevers	TipsNano	Cat# NSG30_SS
PalmSens 3 impedance analyzer	PalmSens BV	https://www.palmsens.com/app/uploads/2016/12/PalmSens3-description.pdf
Cable connector for an IDE	Metrohm DropSens	Cat# CACIDEMEA
Block heater, dual control	Stuart, UK	Cat# SBH130DC
UV/ozone cleaner	Bioforce Nanosciences	N/A
Leica DM500 microscope	Leica	N/A
JPK NanoWizard 4XP, an atomic force microscope	Bruker Instruments	N/A

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Damira Kanayeva (dkanayeva@nu.edu.kz).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• •All data reported in this paper will be shared by the lead contact upon request.
• •The MATLAB code used to calculate the micelle area and micelle circularity is available in this paper’s supplemental information.
• •Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Method details

Fabrication of the EIS aptasensor and CEA detection

The surface of an IDE was checked for integrity and the absence of any scratches using the Leica DM500 microscope (Leica, Germany). Then, it was thoroughly cleaned with 96% ethanol before being subjected to a UV/ozone for 20 min. Following that, the IDE was washed again with ethanol and left to air-dry before measuring the EIS signal in the redox couple buffer containing 2 mM ferro/ferricyanide [Fe (CN)₆]^3−/4− redox couple (potassium hexacyanoferrate II/III) in a 10 mM PBS (pH 7.6). The protocol was adapted from⁵³ with slight modifications. In brief, a 100 μM of HS-(CH₆)₆-OP(O)₂O-(CH₂CH₂O)₆-5′-TTTTT- CEA aptamer (6) -3′ was dissolved with a reduction buffer (TCEP) at a 1:2 volume ratio for 1 h to reduce the 3′ ends of the aptamer. The solution was subsequently diluted with 10 mM PBS (pH 7.6) to yield 5 μM aptamer final concentration. The aptamer solution was heated for 5 min at 95°C before being placed on ice for 10 min and allowed to cool to room temperature for another 5–7 min. Finally, for the immobilization of aptamer onto the IDE surface, IDE was dipped in an Eppendorf tube containing a 500 μl aptamer solution and incubated at room temperature for 4 h. Following the aptamer incubation, the IDE surface was washed with 10 mM PBS (pH 7.6), and an EIS signal was measured. After the aptamer incubation, the IDE was incubated for 16 h at 4°C with 500 μl of 3 mM MCH diluted in 10 mM PBS (pH 7.6). The stock MCH solution was made with 98% ethanol to a concentration of 10 mM and kept at −20°C until further use. Target CEA (0.002 pg/ml to 2 ng/ml) diluted in a 10 mM PBS (pH 7.6) in a volume of 50 μl was incubated on the IDE surface for 20 min at room temperature. Finally, the electrodes were rinsed in a washing buffer (PBS, 10 mM, pH 7.6). The signal was detected in the redox couple buffer containing 2 mM ferro/ferricyanide [Fe (CN)₆]^3−/4− redox couple (potassium hexacyanoferrate II/III) in a 10 mM PBS (pH 7.6). The experimental setup of the method is illustrated in Figure 1.

Electrochemical measurements

Electrochemical measurements were recorded using the PalmSens 3 impedance analyzer equipped with the PSTrace 5.8 software. An IDE was composed of two working electrodes (52 bands, with a band gap and a band width of 10 μm, 4 mm² surface area), an auxiliary electrode, and a reference electrode, all manufactured in the same material (gold), on a glass substrate by optical lithography technology. A cable connector was used to connect an IDE to the potentiostat for the EIS measurement (0.1 Hz to 50 kHz with 56 frequencies and 10 mV a.c.) in the redox couple buffer containing 2 mM ferro/ferricyanide [Fe (CN)₆]^3−/4− redox couple (potassium hexacyanoferrate II/III) in a 10 mM PBS (pH 7.6). Nyquist plots were fitted to Randle’s equivalent circuit using the EIS spectrum analyzer software. CV was performed by scanning the potential from −0.4 to 0.7 V at a scan rate of 0.1 V/s.

Fitting errors of less than 2% were taken for data analysis. All measurements were performed at room temperature inside an in-house-made Faraday cage. All measurements were carried out in triplicate, and the mean value of replicates, standard deviations, and standard errors from the mean were used to report the results.

Optimization studies

An aptamer concentration and a protein incubation time were optimized for the EIS aptasensor development. Detection of 2 pg/ml of CEA in the measurement buffer (10 mM PBS, pH 7.6) was conducted while evaluating the effect of different concentrations (2, 4, 5, and 8 μM) of the CEA aptamer (6). The impact of an incubation time (5, 10, 15, 20, and 25 min) for 2 pg/ml of the target CEA was evaluated, where the IDE surface was functionalized with a 5 μM CEA aptamer (6). All steps were the same as described in Section Fabrication of the EIS aptasensor and CEA detection except for conditions specified in this section.

Specificity study

CEA and non-target proteins such as IL-6, HSA, and SARS-CoV-2 S glycoprotein in a concentration of 7 pg/ml were diluted in a 10 mM PBS (pH 7.6) and were further incubated on the IDE surface for 20 minutes at room temperature. Finally, the electrodes were rinsed in a washing buffer (PBS, 10 mM, pH 7.6), and the signal was detected in the redox couple buffer. IDE surface functionalization steps were the same as described in Section Fabrication of the EIS aptasensor and CEA detection. 10 mM PBS (pH 7.6) was used as a background. All R_ct values were subtracted from the background.

Detection of CEA in serum

Commercially available human serum was diluted 100 times with 100 mM PBS (pH 7.6), and the following CEA concentrations were added into each serum solution: 0.002, 0.02, 0.2, 2, 20, 200, and 2000 pg/ml. After incubating CEA in serum (50 μl) for 20 minutes, the IDE surface was rinsed with 10 mM PBS (pH 7.6), and the electrochemical signals were measured in the redox couple buffer. R_ct values of the sample were subtracted from the background that served as serum diluted 100 times in 100 mM PBS (pH 7.6).

AFM study

The surface of electrodes after each treatment stage was imaged using the JPK NanoWizard 4XP, an atomic force microscope with a super sharp probe NSG30_SS. The sensing tip had nominal spring constant K = 40 N/m, resonant frequency f = 320 kHz, and tip radius r = 2 nm. All images were acquired in the air at room temperature, operating in Quantitative imaging (QI™) mode at 1 μm² scan size with a 5 nm/pixel resolution. The peak force or set point was 10 nN, and the z-speed was 50 μm/s. JPK-Data processing application software was used to analyze all the acquired images, and histogram operation was applied to obtain root mean square (RMS) roughness. Gwyddion software was used to obtain the representative 3D images.⁵² Three independent samples were prepared and imaged for all stages, including the bare IDE, each with at least 15 different sites for statistical analysis. The size and shape of the micelles that were formed on the electrode surface were measured using a custom-written MATLAB code.

Quantification and statistical analysis

For each concentration, the percentage in charge transfer resistance was determined using the formula [(R_conc-n - R_PBS)/R_PBS) × 100] according to the R_ct values obtained with the fitting error values less than 2%. All measurements were carried out at least in triplicate, and the mean value of replicates, standard deviations, and standard errors from the mean were used to report the results. p < 0.05 values were accepted as significant. The statistical significance of the AFM data obtained was evaluated by the Kruskal-Wallis test, one-way ANOVA, Mann-Whitney test, or unpaired t-test, depending on the test type and the features of each dataset. LOD was calculated using the formula presented in a study.⁵⁴

Published: March 28, 2024