Quantum-biological interface in biosensor design: detecting proteins with electrochemical aptasensor

Electrochemical biosensors are promising tools for clinical diagnostics, yet challenges remain in extending sensitivity, linear range, and stability, particularly in complex biological matrices. Here, we report an electrochemical aptasensor for dengue NS1 protein detection based on a self-assembled monolayer (SAM) of DNA aptamers, 6-mercapto-1-hexanol, and 6-ferrocenyl-hexanethiol, characterized using electrochemical capacitance spectroscopy (ECS). The aptamer:thiol ratio was optimized, with the 1:50 condition providing the best analytical performance. The platform achieved sensitivities of 0.18% ± 0.02% per decade in PBS and 0.21% ± 0.01% per decade in commercial human serum, within a linear range of 0.01–1,000 ng/mL. Limits of detection were 24.9 ng/mL in PBS and 25.8 ng/mL in serum. Although long-term stability decreased after 7–14 days, the sensor demonstrated robustness in both simple and complex medium. These results confirm the viability of aptamer-based ECS platforms for clinically relevant NS1 detection and represent a step toward integrating quantum-scale concepts into bioelectrochemical sensing.

The study and development of biosensors is relatively new in the history of science. In 1916, Griffin and Nelson demonstrated the immobilization of the enzyme invertase in aluminum hydroxide and charcoal (Kansas et al., 1916). In 1956, Leland C. Clark, considered the “father of biosensors,” developed the first true biosensor to detect oxygen, and in 1962, he developed an amperometric enzyme electrode for glucose detection (Heineman et al., 2006). Then, in 1969, Guilbaut and Montalvo discovered the first potentiometric biosensor for urea detection (George and Moltalvo, 1969).

Since those days, several analytical techniques have been developed, enabling a wide variety of biosensors, focus on acoustic, electrical, optical, magnetic, thermal, and mechanical principles (White, 1987). In particular, electrochemical platforms are versatile (Thévenot et al., 1999) and one of the most widely used types of transducers for analytical applications, allowing the use of conductometric (Akhtarian et al., 2023; Saiapina et al., 2024), impedimetric (Štukovnik and Bren, 2022; Silva et al., 2024), amperometric (Guilbault and Lubrano, 1973; Rehwald et al., 1984; Jiinsson and Gorton, 1985; Mizutani and Yabuki, 1997) e and potentiometric (Walker et al., 2021) techniques.

The impedimetric technique is based on applying a small amplitude AC potential and measuring the resulting current, which can be classified as Faradaic or non-Faradaic. In Faradaic process, there is a reaction on the electrode surface in which an ion is reduced or oxidized, with the reagent in solution and the reaction products not retained on the electrode surface, returning to the solution (Biesheuvel et al., 2021). In non-Faradaic current, the charge is progressively stored, and the reactant species that are oxidized/reduced is retained on the electrode surface, becoming part of the electrode structure itself, so the ions that interact with the surface are not released back to the solution bulk (Biesheuvel et al., 2021).

These phenomena can be studied using electrochemical impedance spectroscopy (EIS) and electrochemical capacitance spectroscopy (ECS) techniques. While EIS measures complex impedance, ECS measures how the capacitive component of the electrode/electrolyte interface varies with frequency and potential, with the redox pair coupled on the surface (Cecchetto et al., 2017), eliminating the need for analytical redox solutions.

With ongoing advancements in the field, there is a clear trend toward the development of more sensitive devices, featuring broader linear ranges and lower detection limits that enable the identification of minute quantities of analytes or the quantification of highly specific biological phenomena. As illustrated in the timeline (Figure 1), early biosensors focused on detecting relatively simple molecules, such as glucose, at millimolar concentrations (10 −3 M) (Guilbault and Lubrano, 1973). Over time, the scope of detectable analytes has expanded significantly, supporting a wide range of diagnostic applications. Today, some biosensors can achieve detection at the femtomolar level (10 −15 M) even for complex samples (Zhang and He, 2024).

When detection scales are reduced to low levels, quantum phenomena may become more relevant to explain behaviors observed in physical-chemical systems. Thus, researchers are currently working on hypotheses and simulations in order to reinterpret systems related to the accumulation or blocking of charges (Serra, 2020), and determine the impact of stochastic phenomena, quantization of energy levels of interfaces and molecules (Grall et al., 2025) in the context of particle interaction and applications in biosensors, also proposing formulations for new concepts, such as quantum capacitances (Sánchez et al., 2025), and seeking to prove and characterize them with practical experimental and computational methods.

There are many studies that use capacitive analysis in diagnostic sensors, exploring different transduction strategies and, especially, different models for processing electrochemical data. For example, some studies employ capacitive sensors for the detection of protein biomarkers, in which the analyte-bioreceptor interaction promotes changes in the dielectric properties of the electrode/solution interface, and the analytical signal is treated as the relative variation in capacitance as a function of target concentration, allowing quantitative interpretation based on surface coverage and classical adsorption models (Le et al., 2020). There are also studies that use more complex capacitive architectures, in which the sensor response is extracted from capacitive parameters derived from electrochemical spectroscopy, with the data often normalized or transformed to increase sensitivity and robustness against biological matrices, while maintaining a quantitative character in correlation with the analyte of interest (Cecchetto et al., 2017). In contrast to classical treatment, some studies demonstrate that capacitive systems can be treated from a perspective based on quantum characteristics of electronic transport, in which biomolecular interaction modulates parameters associated with state density and effective response rates of the sensor material, so that the spectroscopic response is reinterpreted in terms of derived quantities that reflect the quantum regime of the system, allowing the diagnostic signal to be analyzed by statistical metrics and classification criteria, rather than relying exclusively on absolute capacitive parameters (Lucas Garrote et al., 2022).

In this context, the present study seeks to develop an aptasensor platform to detect Dengue NS1 using capacitance techniques based on a self-assembled monolayer (SAM) composed of DNA aptamers, 6-mercapto-1-hexanol (MCH), and 6-ferrocenyl-hexanethiol (FCH) to form a redox monolayer. Electrochemical capacitance spectroscopy demonstrated the aptamer-protein interaction, allowing the determination of an optimal ratio between the aptamer and thiol in the SAM to improve analytical parameters such as sensitivity, limit of detection (LoD), limit of quantification (LoQ), linear range, and stability, focus on future high-sensitivity biosensors.

Materials

The DNA aptamer sequence (5′- HS(CH 2) 6 – TTT​TAG​CGG​ATC​CGA​TGG​GTG​GGG​GGG​TGG​GTA​GG ATCCGCG - 3′) HPLC-purified was purchased from Exxtend in lyophilized form (Mok et al., 2021). NS1 protein, serotype 4, was obtained from Abcam (ab181957). Tris-EDTA, TE buffer (10 mM Tris-HCl, 1 mM disodium EDTA, pH 8.0), magnesium chloride (MgCl 2, ≥98%), bovine serum albumin (BSA), dibasic sodium phosphate (Na 2 HPO 4, ≥99%), monobasic potassium phosphate (KH 2 PO 4, ≥99%), 6-mercapto-1-hexanol (97%), 6-ferrocenyl-hexanethiol, phosphate-buffered saline tablets and commercial human serum (male AB plasma, US origin, sterile-filtered) were purchased from Sigma-Aldrich. All aqueous solutions were prepared using ultrapure water (18.2 MΩ cm, Merck Millipore Direct-Q 5UV) with a Biopak Polisher filter (Merck Millipore, CDUFB-001). Sulfuric acid (H 2 SO 4, ≥95%) and sodium chloride (NaCl, ≥99%) were purchased from Synth, sodium hydroxide (NaOH, ≥97%) from Dinamica, and potassium chloride (KCl, ≥99%) from Cinetica. The gold surface electrodes were purchased from CH Instruments, Inc. (CHI101) and have a diameter of 2 mm.

Electrode functionalization

The electrode surface was functionalized through co-immobilization of DNA aptamer sequences with a spacer molecule, 6-(ferrocenyl)hexanethiol (FCH), using an immobilization buffer with 0.8 M PB, 1.0 M NaCl, 5 mM MgCl 2, and 1 mM EDTA (pH 7.0) (Keighley et al., 2008). The electrodes were incubated in a humid chamber for 16 h at 4 °C. Following this step, the electrodes were rinsed and incubated for 1 h in MCH (10% ethanol) to ensure complete surface blocking (Bachour Junior et al., 2021). Finally, they were washed with PBS and stabilized for 1 h in the same solution.

The study to optimize the aptamer-protein ratio was performed in quadruplicate. The analytical curve was obtained in sextuplicate for each target protein concentration, and temporal stability was analyzed based on triplicate for each protein concentration.

Statistical analysis was performed by analyzing the means of the data sets analyzed and their associated uncertainties with the sum of the squares of the deviations of data points from the sample mean.

Aptamer and NS1 preparation

Lyophilized aptamers were reconstituted using TE buffer (Tris-HCl + EDTA), aliquoted, and stored at −40 °C. Before experiments, aptamers were heated at 95 °C for 10 min, followed by gradual cooling to room temperature over 1 hour (Mok et al., 2021). Non-structural 1 (NS1) protein was diluted in PBS, aliquoted and stored at −40 °C. Protein samples were prepared using PBS or commercial human serum in different concentrations (Bachour Junior et al., 2021).

Electrochemical measurements

All electrochemical experiments were performed using a potentiostat (Metrohm/Autolab PGSTAT302), shown in Figure 2a, with FRA module controlled by NOVA 2.1 program and a three-electrode electrochemical cell configuration, shown in Figure 2b, with Ag/AgCl as reference electrode and platinum as counter electrode. Electrochemical capacitance spectroscopy was carried out using EIS protocol with a fix potential corresponding to the maximum phase angle and frequencies range from 100 kHz to 100 mHz (Bachour Junior, Batistuti Sawazaki and Mulato, 2024). Experiments are conducted in PBS and commercial human serum before and after interaction with NS1, as well as at each step of surface modification.

The capacitance of the system is evaluated based on the complex capacitance (Jolly et al., 2016), defined as: Where Z′ and Z″ are the real and imaginary components, respectively, of the impedance measurements, ω = 2πf is the angular frequency of the measurement. The real and imaginary components of capacitance are represented by C′ and C″. From this, Cole-Cole capacitance plots, which are graphs of C′ vs. -C″, are obtained, so that the capacitance of the system is obtained from the diameter of the semicircle formed (Jolly et al., 2016).

As the functionalization of the monolayer is based on a mechanism of reducing the capacitance of the system as the platform interacts with the target molecule, the convention is to perform sensitivity analysis and other characterizations based on −ΔC, so that a reduction in the capacitance measurement results in an increase in the detection signal.

The sensitivity of the platform is obtained based on the calibration curve obtained for the capacitive response in relation to the variation in concentration of the target analyte, thus being represented by the angular coefficient given by the fitting of the data set and its error given by the standard deviation of this adjustment, also provided by the analysis software.

Platform optimization

The occurrence of aptamer-protein interaction was verified based on ECS measurements, relating the capacitance reduction as a function of the target analyte concentration increase (Figure 3).

As shown in Figure 3b, the frequency with the highest phase angle was ∼1.6 Hz, corresponding to 87°. This frequency was used for the following analyses, since the capacitive signal changes as a function of frequency (Singh et al., 2018) and this indicates the point of highest capacitive response of the platform.

The surface density was optimized using aptamer:thiol ratio immobilized on the electrode surface by altering the volume fraction of aptamer and MCH in the immobilization solutions. Based on the percentage variations in capacitance, − Δ C (%), shown in Figure 4, the optimized ratio was defined as 1:50, with an increase from approximately 0.35256% ± 0.00006%–0.575% ± 0.008% for the variation from 10 to 1,000 ng/mL.

Aptasensor performance and analytical parameters

The optimized aptasensor platform (1:50 ratio) was employed to evaluate the analytical parameters. Calibration curves were constructed using protein concentrations from 0.01 to 1,000 ng/mL with increase in decade, prepared in PBS and commercial human serum. For assays in PBS, the SAM surface was blocked with BSA diluted in PBS, whereas for assays in serum, undiluted serum containing HSA was applied to block the surface and minimize non-specific interactions.

Figure 5 presents the analytical curve obtained for this aptasensor. In PBS the interaction rate ranged from −0.143% ± 0.004%–0.99% ± 0.02% across NS1 concentrations from 0.01 ng/mL to 1,000 ng/mL. In commercial human serum, interaction rates varied from 0.297% ± 0.008%–1.29% ± 0.01% for the same concentration range. The platform in PBS exhibited a sensitivity of 0.18% ± 0.02% per decade (R 2 = 0.94), whereas in serum the sensitivity was 0.21% ± 0.01% per decade (R 2 = 0.99).

The limit of detection (LoD) is defined as the lowest concentration of analyte in a sample that can be detected, though not necessarily quantified, under the specified experimental conditions (Agência Nacional de Vigilância Sanitária - ANVISA, 2017). It is determined by multiplying the standard deviation by 3.3 and dividing the result by the slope of the calibration curve. The LoD values obtained were 24.9 ng/mL for the protein diluted in PBS and 25.8 ng/mL for the protein in undiluted serum.

The limit of quantification (LoQ) refers to the lowest analyte concentration that can be reliably measured with acceptable precision and accuracy under the established experimental conditions (Agência Nacional de Vigilância Sanitária - ANVISA, 2017). It is calculated by multiplying the standard deviation by 10 and dividing by the slope of the calibration curve. The LoQ values determined were 75.6 ng/mL for the protein diluted in PBS and 78.2 ng/mL for the protein in commercial human serum.

Platform long-term stability was evaluated following the established protocol: after backfilling with MCH, the electrodes were immersed in PBS buffer and stored at 4 °C. Measurements were taken in the first day after SAM immobilization (standard procedure), after 7 and 14 days. Figure 6 presents the results obtained for three NS1 concentrations across these intervals.

At 10 ng/mL, the responses were 0.107% ± 0.005% after 1 day, −0.2651% ± 0.0005% after 7 days, and −0.5624% ± 0.0003% after 14 days. For 100 ng/mL, the values were 0.224% ± 0.005%, −0.2223% ± 0.0004%, and −0.424% ± 0.002% at the same time points. At 1,000 ng/mL, the responses recorded were 0.68% ± 0.03%, −0.1481% ± 0.0002%, and −0.165% ± 0.004% for 1, 7, and 14 days, respectively.

Platform response and optimization

The aptamer-NS1 interaction shift the redox center of molecular film close to the electrode surface, reducing capacitance, as observed in Figure 3c and facilitating the electron transfer between redox species (ferrocene) in the SAM and the electrode. This process is still viewed mainly in a classical perspective, being a niche that requires further computational simulation studies involving calculations of energy levels and wave functions of molecules such as proteins and bioreceptors, seeking to better understand how characteristics of such structures, such as size, charge, and composition, influence electron transfer bioreceptor-target interaction, and which mechanisms act in this process at quantum levels.

The optimized aptamer:thiol ratio, 1:50, was considered the optimal ratio, as it presents a linear increase in signal, consistent with NS1 concentration increase and with the greater response to the lowest concentration of the target protein. The results show that at ratios of 1:100, 1:200, and 1:500, the low density of aptamers on the surface increases non-specific interactions, leading to higher signal variability, including negative values.

Analytical parameters

Figure 5 present the analytical curve where the protein diluted in commercial human serum produced a higher capacitive response compared to NS1 diluted in PBS. This behavior can be related to the greater complexity of human serum, which has higher ionic strength and an increased likelihood of non-specific adsorption on the detection layer. Nevertheless, the sensitivity of the detection monolayer remained consistent, showing a comparable signal variation per decade under the tested conditions. These findings highlight the robustness of the platform, demonstrating that the optimized ratio ensures a target-selective surface capable of maintaining high sensitivity even in complex matrices. Furthermore, ECS measurements revealed a linear response for both mediums throughout the entire concentration interval, encompassing clinically relevant levels used in diagnostic tests (Alcon et al., 2002).

The stability of the detection platform showed an increase in capacitance in tests performed after 7 and 14 days. This behavior may indicate a loss of function of the detection monolayer, representing a degradation of the aptamers and oxidation of immobilized thiolate elements (Love et al., 2005), generating an accumulation of charges on the biosensor surface and facilitating the occurrence of non-specific interactions, given the disorganization of SAM due to the loss of aptamers, MCH, and FCH.

The use of other storage methods would be an option to extend the biosensor’s shelf life. Methods such as vacuum packaging or the use of modified atmospheres, with the use of inert gases, can help slow down the oxidative processes (Vermeulen et al., 2024) that degrade the detection monolayer, allowing detections for a longer period after storage.

The future of analysis and the impact of quantum phenomena

The Faradaic and non-Faradaic processes are the main model and method of analyzing the surface dynamics of various biosensors, representing the occurrence of charge transfer and charge accumulation processes on the sensor surface. However, these models are still based only on classical phenomena, and therefore represent an area of study to be developed, with a view to enabling the validation of hypotheses and the creation of new system modeling tools that take into account the occurrence of quantum effects.

Faradaic processes are modeled through electrochemical equivalent circuit, as Randle circuit (Cecchetto et al., 2017). However, Terra emphasis in his thesis that traditionally, Coulomb blockade phenomena have been described using semiclassical or equivalent circuit models, in which “experimental results concerning these phenomena have been modelled with either semiclassical or circuit-like models, which derive electrical parameters such as capacitances and resistances in order to reproduce the measured currents trends” (Serra, 2020). In contrast, the methodology from this thesis is based on first-principles modeling, solving the Schrödinger equation using the Finite Element Method (FEM) and, mainly, the Non-Equilibrium Green’s Function (NEGF) formalism to directly simulate electronic transport. Thus, while classical models represented Coulomb blockade effects through equivalent capacitances, this work proposes that such effects emerge from the discretization of energy levels and the dynamics of electron tunneling in nanostructures, constituting the physical origin of capacitive behavior. Although the simulations do not yet fully reproduce the experimental results, the methodology constitutes a solid first step toward modeling biosensor (Serra, 2020).

In the other hand, non-Faradaic current from electron transfer at interfaces modified with redox monolayers, as biosensor surfaces can be described using quantum capacitance (Cq) as a fundamental parameter (Sánchez et al., 2025). Unlike classical capacitance, which is linked to the accumulation of charges in an electric double layer, Cq is directly associated with the density of electronic states (DOS) of the interface. In this context, each redox center of the molecular film functions as a discrete energy level, capable of mediating electron transfer between free redox species in the electrolyte and the electrode, and each molecule in the monolayer functions as a discrete quantum state, whose occupation depends on the energy alignment with the free redox species in the electrolyte and follows the statistical laws of quantum mechanics (Fermi–Dirac statistics). Thus, the quantum capacitance theory shifts the classical understanding of a “geometric and cumulative” property to an electronic and dynamic property, closely connected with the quantization of available states and the active role of molecules in mediating charge transfer processes.

Furthermore, a new computational modeling platform for studying bioelectrochemical processes at quantum scale was also developed (Grall et al., 2025). The methodology combines the formalism from point stochastic processes with descriptions of electronic dynamics, allowing the electron transfer simulation events at the individual level. This approach was implemented in proprietary software (QBIOL), designed to integrate statistical, kinetic, and quantum methods into a single simulation framework. The main results showed that QBIOL is capable of reproducing not only the average kinetics of bioelectrochemical reactions, but also the stochastic fluctuations associated with electron transport. This is particularly relevant in low-dimensional systems, such as sensors based on nanoparticles or molecular monolayers, in which the experimental response can be highly nonlinear and dependent on confinement effects, so that traditional modeling based only on deterministic differential equations cannot capture these behaviors, whereas the stochastic-quantum approach can provide more realistic predictions for nano- and bioelectronic systems.

In this way, it is expected that even more sensitive sensors with lower detection limits will be able to be established reliably and with good reproducibility, once methods for modeling the phenomena that interfere with the dynamics of their surface are known.

It is concluded that the use of the aptamer sequence as a bioreceptor in SAM for the detection of NS1 protein from dengue proved to be feasible, with analytical parameters properly characterized and optimized. At a 1:50 aptamer:thiol ratio, 0.18% ± 0.02% sensitivities per decade in PBS and 0.21% ± 0.01% in commercial human serum were obtained, with a linear range from 0.01 to 1,000 ng/mL. The platform demonstrated robustness against variations in NS1 protein concentration, maintaining a consistent response even in complex biological matrices, such as human serum. Long-term stability studies, however, indicate the need to implement additional strategies to extend the sensor’s shelf life. Thus, the performance obtained reinforces the potential of the aptasensor for application in high-sensitivity platforms, operating within the clinical diagnostic range and representing a step forward toward the detection of events on a quantum scale.