Beyond the Gold–Thiol Paradigm: Exploring Alternative Interfaces for Electrochemical Nucleic Acid-Based Sensing

Abstract

Nucleic acid-based electrochemical sensors (NBEs) use oligonucleotides as affinity reagents for the detection of a variety of targets, ranging from small-molecule therapeutics to whole viruses. Because of their versatility in molecular sensing, NBEs are being developed broadly for diagnostic and biomedical research applications. Benchmark NBEs are fabricated via self-assembly of thiol-based monolayers on gold. Although robust for rapid prototyping, thiol monolayers suffer from limitations in terms of stability under voltage modulation and in the face of competitive ligands such as thiolated molecules naturally occurring in biofluids. Additionally, gold cannot be deployed as an NBE substrate for all biomedical applications, such as in cases where molecular measurements coupled to real-time, under-the-sensor tissue imaging is needed. Seeking to overcome these limitations, the field of NBEs is pursuing alternative ligands and electrode surfaces. In this perspective, I discuss new interface fabrication strategies that have successfully achieved NBE sensing, or that have the potential to allow NBE sensing on conductive surfaces other than gold. I hope this perspective will provide the reader with a fresh view of how future NBE interfaces could be constructed and will serve as inspiration for the pursuit of collaborative developments in the field of NBEs.

The 21st century is witnessing a surge of technological advances enabled by sensor science. Ever more sophisticated sensors control airplanes and privately funded space rockets, make possible the existence of self-driving cars, and play key roles in the electric vehicle revolution. Sensors have also changed the broad human experience of war, exposing society to real-time, fluid confrontations observable via battle-engaged drones and warfighter-worn monitors.^1,2 Unquestionably, sensors will also empower artificial intelligence, potentially taking us to a future that could only be described as science fiction becoming real life. Such societal transformations are also infusing changes in medical care. New healthcare paradigms seek to leverage sensor technologies to achieve real-time health monitoring on the go for high-precision, individualized therapy. Examples of such new paradigms include wearable watches that continuously check for heart arrythmias,^3,4 continuous glucose monitors integrated with insulin pumps for feedback-controlled glucose management,⁵ and wearable rings that track circadian behavior.⁶ Thus, regardless of application, it seems clear that time-resolved sensing platforms will continue to permeate many aspects of the human experience.

While most industrial applications listed above rely on physical sensors, continuous molecular monitoring is critical for biomedical research and medical applications.⁷ To enable the tracking of specific molecules in the body, sensor platforms use biorecognition elements such as enzymes, antibodies, or nucleic acids. Such affinity reagents provide molecular specificity that can be coupled to sensitive signaling schemes. This perspective focuses on nucleic acid-based electrochemical sensors (NBEs), a group of technologies that leverage the sequence- and structure-specific affinity of nucleic acids for molecular recognition, and their folding dynamics, secondary and tertiary structures for signal transduction. Although many names or acronyms can be found in the literature to refer to variants of these platforms, here I use the term NBEs so that the concepts discussed can be taken as structure, application, and sensing mechanism agnostic. Also, please note that many of the concepts and strategies discussed here may also apply to sensors based on other affinity receptors such as antibodies,⁸ nanobodies,⁹ peptides,¹⁰ or templated polymers.¹¹

To enable signaling, NBEs often rely on reversible target binding-induced changes in their structural conformation. These sensors are designed to have, ideally, two structural states: a target-bound state that transfers electrons at a rate k₁, and a target free state that transfers electrons at a rate k₂, where k₁ ≠ k₂ by at least 1 order of magnitude. The differences in magnitude between k₁ and k₂ are introduced via careful structural design of the nucleic acid sequences (Figure 1). Here I highlight three example NBE technologies that leverage this signaling strategy: (1) electrochemical, aptamer-based sensors (E-ABs, Figure 1A),¹² in which nucleic acid aptamers undergo binding-induced folding that often increases the rate of electron transfer between a reporter and the electrode surface, (2) electrochemical DNA sensors (E-DNAs, Figure 1B),¹³ in which hybridization of target oligonucleotides (oligos) with an electrode-bound nucleic acid strand often causes a decrease in electron transfer rate, and (3) molecular pendulum sensors (MPs, Figure 1C),¹⁴ which are actuated by reversibly stepping the voltage of the electrode between values opposite relative to the sensor’s potential of zero charge. When the sensor is negatively polarized, electrode-bound double-stranded oligos are repelled away from its surface; when positively polarized, they are attracted. This voltage-induced oligo actuation drives electron transfer at a rate that changes in magnitude when the sensors are exposed to target, because target binding changes the drag forces that affect the field-induced oscillatory movement of the oligos.

Figure 1.

Example architectures of nucleic acid-based electrochemical sensors (NBEs). These technologies achieve signaling by leveraging target binding-induced changes in the structural conformation of electrode-bound nucleic acids. Here I highlight three examples. (A) Electrochemical, aptamer-based sensors (E-ABs). In this example, k₂ ≫ k₁. (B) Electrochemical DNA sensors (E-DNAs). Here, k₁ ≫ k₂. (C) Molecular pendulum sensors, in which k₁ ≫ k₂. In all illustrations, k₁ represents the electron transfer rate of the unbound state and k₂ the rate of the bound state.

Although the aformentioned technologies are based on different signaling strategies, it is important to note they share a common architecture: an electrical conductor gold surface is coated with a self-assembled monolayer consisting of a mixture of blocking alkylthiols and functional alkylthiol-modified oligos. The oligos are additionally modified with a redox reporter able to transfer electrons to the underlying gold electrode. This architecture is beautiful for its simplicity and for the ease with which it allows rapid sensor prototyping. Usually, NBEs are fabricated via a series of passive incubation steps^13,14 in which clean gold surfaces are first immersed in solutions containing the alkylthiol-modified oligonucleotides for a set period; then, immersed in solutions containing the blocking alkylthiols, usually overnight. The resulting sensor interfaces can be interrogated in a variety of media shortly after.

Unfortunately, the NBE sensing architecture holds some critical limitations. The thiol–gold bonds (S–Au) it uses are prone to competitive displacement by thiols naturally occurring in biological fluids (e.g., cysteine, glutathione).¹⁵ Additionally, thiol-based monolayers often have imperfections (i.e., gaps, pinholes),¹⁶ which leave room for undesired electrochemical reactions to occur in the background.¹⁷ The reactions that most immediately affect sensor stability are surface gold oxidation at positive voltages¹⁸ and the formation of reactive oxygen species on gold at negative voltages.¹⁹ Both cause decay of NBEs and can occur simultaneously if sensor interrogation is not carefully tuned.

To illustrate the impact of the above processes on sensor decay, Figure 2 shows experiments recently conducted in my laboratory. E-ABs made with a methylene blue-modified aptamer (the aptamer’s target is not relevant to this discussion) were immersed in phosphate-buffered saline (Figure 2A) or the same solution containing physiological levels of cysteine (∼250 μM, Figure 2B). The sensors were interrogated every 10 s via voltammetry, going negative to a voltage where methylene blue and oxygen reduction to peroxide can occur (<−0.2 V vs Ag|AgCl). Under these conditions, cyclic voltammograms measured at t = 0 h and at t = 16 h after continuous interrogation (once cycle every 10 s) show linear disappearance of the methylene blue redox current down to 50% for NBEs deployed in phosphate-buffered saline (Figure 2C). However, in the presence of cysteine the methylene blue redox current exponentially disappeared down to <20% over the same period (Figure 2D). These current losses can be directly correlated to loss of oligonucleotides from the sensor via surface chronocoulometric titrations using ruthenium hexamine,¹⁵ confirming that loss of oligonucleotides, not methylene blue, drives the decay of these NBEs.

Figure 2.

Effect of competitive displacement on NBEs signaling. Cyclic voltammograms measured in (A) phosphate-buffered saline or (B) the same solution containing 250 μM of cysteine (approximately the physiological levels found in blood). The voltammograms correspond to E-ABs using a methylene blue-modified aptamer and mercaptohexanol backfilling, at 25 °C. Voltammetric scanning was conducted serially every 10 s for 16 h across the voltage window shown. (C) Without cysteine, the methylene blue redox current decayed linearly via voltage-induced desorption of thiol-modified aptamers. (D) In the presence of cysteine, the decay becomes exponential because of competitive displacement by cysteine. Solid lines represent the mean of 8 sensors, shaded areas their standard deviation.

Other degradation pathways for NBEs exist, including for example the lability of thiol monolayers at increased temperatures (e.g., sensors to be deployed in the body must operate at 37 °C). Enzymatic degradation of sensor oligonucleotides by nucleases is also a concern over long periods of deployment in biofluids. Finally, nonspecific protein binding also affects the signaling of NBEs. However, such decay mechanisms are additive to voltage-induced desorption (Figure 1A) and competitive displacement (Figure 2B), which occur with faster time constants. Thus, to keep the focus of this perspective narrow, I only focus on discussing the latter.

Motivated to overcome the problem of competitive displacement of thiols in biofluids, my laboratory and others have pursued the investigation of alternative electrode coating chemistries not prone to ligand exchange in vivo. This motivation has been further driven by the realization that gold is not an ideal surface for certain biomedically valuable technology applications. For example, monitoring tumors or infections in the brain would require the coupling of continuous sensing with magnetic resonance imaging (MRI) to regularly probe tumor/plaque secretions and size. Unfortunately, NBEs made on gold would not be MRI transparent; therefore, other coatings may be required for such applications.

The following sections discuss efforts made on the quest to expand the scope of coating chemistries and surfaces for NBE sensing beyond thiols on gold. This discussion does not include strategies based on composite materials (the often-unjustified mixing or serial casting of various elements that abounds in the literature). Instead, the strategies discussed below were carefully selected because they aim at mimicking the interfacial architecture of NBEs as illustrated in Figure 1: specifically, strategies that sought to create an electrode blocking layer, functionalize this layer with oligonucleotides, or/and interact with the conformation-switching characteristics of the oligonucleotides via redox reporters.

Strategies Based on Self-Assembly

Thiol-on-gold sensing interfaces are attractive for NBEs prototyping in part because of how easily thiol monolayers can be formed via self-assembly on gold. This characteristic raises the question: could other self-assembling ligands be leveraged for the fabrication of operationally stable NBEs on gold and other surfaces? The immediate answer is potentially yes, with the caveat that thiol–gold monolayers have been investigated and developed for over four decades;²⁰ thus, alternative ligands may require extensive development to catch up with the prevalence of established alkanethiols. Nevertheless, below I discuss promising strategies that may pave the way for future expansion of NBEs to interfaces other than thiols on gold.

The first promising strategy consists of self-assembling monolayers of N-heterocyclic carbenes (NHCs). NHCs form strong σ-bonds with transition metals.^22,23 Additionally, in 2013 it was demonstrated that NHCs can self-assemble onto monolayers.²⁴⁻²⁸ This discovery was exciting because, unlike thiols, NHC monolayers are not susceptible to competitive displacement by thiols naturally occurring in biological media or by other adsorbates.²⁹⁻³⁵ Although extensive research has been dedicated to demonstrating the assembly and structural characteristics of NHC monolayers, only a couple of studies have investigated their long-term stability under the continuous voltage cycling used to interrogate NBEs. The first publication, by Pellitero et al.,²¹ demonstrated that NHC monolayers can passivate gold electrodes to the same extent as benchmark thiol monolayers (see capacitive currents in Figure 3A, left vs right), and achieve comparable stability over 7 days of continuous voltage cycling (Figure 3B). Similar observations were made in a following publication by Dominique et al.,³⁶ in which the stability of NHC monolayers on gold under continuous voltage cycling was investigated via surface-enhanced raman spectroscopy (SERS, Figure 3C). This second study demonstrates that NHC monolayers on gold can tolerate voltage modulation for over 11,000 cycles. The two manuscripts noted that modifications to the NHC monolayer backbone can affect monolayer stability under voltage cycling; thus, further investigations into NHC structure–stability relationships, and how they affect sensing performance, are needed. Additionally, at the time of writing this perspective only one publication has reported the successful coupling of nucleic acid oligonucleotides to NHCs.³⁷ However, this report focuses on therapeutic applications of NHCs and not on NHC monolayers for biosensing. Given the limited extent to which NHC-supported biosensing monolayers have been investigated, and their promise as alternative monolayer forming ligands, NHCs are ripe for aggressive exploration in the field of NBEs.

Figure 3.

N-heterocyclic carbene monolayers on gold as a potential substitute for thiols. (A) Top panels showing cyclic voltammograms of an NHC-ester (left) and mercaptohexanol (right) monolayers on polycrystalline gold electrodes measured at t = 0 h and t = 168 h (7 days) of incubation in phosphate-buffered saline at 25 °C. Note that capacitive currents for both monolayers are virtually identical after 7 days. (B) Bottom panels showing voltammetric charging currents sampled at 0.15 V every 10 s for the NHC-ester (left) and mercaptohexanol (right) monolayers. While NHC reorganization leads to better electrode passivation and, therefore, lower charging currents, the mercaptohexanol monolayer is structurally stable until it slowly starts to desorb by the end of the measurement period. Solid lines represent the average of 4 electrodes. Shaded areas represent their standard deviation. All voltammograms measured at a rate of 100 mV s^–1. Adapted with permission from ref (21). Copyright 2023 American Chemical Society. (C) SERS spectra of an NHC monolayer on gold acquired before and after performing cyclic voltammetry at 100 mV s^–1. Spectroscopic signatures were unchanged up until 11,250 voltammetric cycles. Adapted with permission from ref (21). Copyright 2023 Wiley and Sons, Inc.

A second strategy based on self-assembly involves the tethering of nucleic acids to carbon electrodes via π–π stacking. Carbon is an attractive electrode material for NBEs because of its biocompatibility and broad voltage window in aqueous environments (∼2 V). Carbon probes are already in use for electrophysiological³⁸ and neurotransmitter measurements in the brain,³⁹ but have not been extensively used for affinity-based continuous molecular monitoring. Here, I highlight the use of the anchoring group pyrene to self-assemble oligonucleotides on carbon electrodes (Figure 4A).^40,41 Specifically, pyrene and its analogues can π-stack onto graphene and carbon nanotubes.⁴² This approach was first leveraged to create self-assembled layers of nucleic acid aptamers onto fluorescence resonance energy transfer-based sensors (FRETs).⁴³ The resulting sensors achieved continuous molecular monitoring inside a flow chamber over measurement periods of minutes (Figure 2B). Unfortunately, such pyrene-supported sensors were not exposed to voltage modulation, leaving open the question of whether this strategy can support continuous interrogation via voltage cycling. However, in a 2023 preprint by Wu et al.,⁴⁴ pyrene-modified oligonucleotides were employed for continuous in vivo monitoring of evoked dopamine release in mouse brains. This preprint reports functional NBEs in field effect transistor (FET) format (Figure 4C). Much like in Figure 1A, target binding to the aptamers causes a conformational change in surface-bound oligos. However, instead of changing the electron transfer rate of the system, in this case the binding event causes a change in the number of whole carriers within a graphene-supported channel. The experimental result is similar: a change in current at a given voltage. Such FET-based NBEs achieve dopamine monitoring for multiple days in vivo, albeit with observable decreases in signaling output over time (Figure 4D). At this stage, it is not clear what causes the sensor signal to decrease. Their NBEs are coated with an antifouling layer; therefore, their sensing interface is not naked to the in vivo environment like in the examples shown in Figure 1. Thus, although pyrene-based tethering of oligonucleotides to graphene electrodes seems to be a promising approach for continuous NBE sensing, several key questions remain unanswered. First, π-stacking of polycyclic aromatic compounds like pyrene on graphene has been determined to be physisorption,⁴⁵ leaving open the question of how stable these interactions are under continuous voltage modulation. Second, the resistance of this bonding mechanism in the face of other absorbates (competitive displacement) has not been directly investigated. Third, carbon electrodes are known to erode under extensive voltage cycling (depending on voltage magnitude, of course), potentially limiting the operational life of carbon electrodes.

Figure 4.

NBEs using pyrene-modified oligonucleotides for self-assembly on graphene. (A) Example of a pyrene-modified oligonucleotide. The structure shown was synthesized by coupling pyrene to the oligo via Click reaction. Adapted with permission from ref (40). Copyright 2020 Mary Ann Liebert, Inc. (B) Fluorescence emission profile of graphene-supported NBEs under continuous flow and in the face of changing concentrations of an aptamer’s target. Adapted with permission from ref (43). Copyright 2016 American Chemical Society. (C) Illustration of pyrene-modified PEG5-alcohol polymers and nucleic acid oligos inside a protective hydrogel for continuous in vivo dopamine sensing. (D) Graphene-based NBEs monitoring dopamine release in mouse brains at day one of measurements (top) and after 5 days (bottom), showing ∼50% signal loss. Panels (C) and (D) adapted from ref (44). Copyright 2023 under Creative Commons 4 license.

Strategies Based on Electrografting

An alternative to monolayer formation based on molecular self-assembly is the grafting of ligands via electrode-driven oxidation–reduction reactions.⁴⁶ One of the most investigated strategies in this sense is the electrografting of aryldiazonium ions (Figure 5A), which results in organic layers that can be further functionalized post deposition (for example, via orthogonal amide or Click coupling reactions). However, a critical challenge with this family of reagents is that they tend to form multilayers during the grafting process.⁴⁷ To overcome this issue, careful design of the diazonium ring substituents and the electrolyte used during electrografting, as well as of the deposition period,^48,49 is critical. For example, approximately full monolayer formation has been achieved via use of aryldiazonium ions with (i) cleavable protection groups^50,51 or (ii) leveraging steric constraints,⁵² via electrografting in the presence of (iii) a radical scavenger⁵³ (Figure 5B) or solvated redox mediators,⁵⁴ and (iv) from ionic liquids.⁵⁵ These cited works report electrode surface coatings between 0.9 and 1.6 nm (close to single-monolayer thickness). For reference, a monolayer of mercaptohexanol on gold is ∼0.9 nm thick (as measured via force microscopy).⁵⁶ Because radical scavengers can be deployed on commercially available aryldiazonium salts and do not require significant synthetic knowhow, this approach⁵³ may be the most pragmatic for future NBEs development. However, comprehensive works characterizing monolayer density, reactivity, and efficiency for coupling oligonucleotides, and extended layer stability under continuous voltage cycling, are still required.

Figure 5.

Electrografting as a route to fabricate NBEs on carbon. (A) Grafting of aryldiazonium ions onto carbon typically yields multilayer-thick electrode coatings. (B) However, the use of radical scavengers like 2,2-diphenyl-1-picrylhydrazyl (DPPH) can significantly limit the polymerization of ions in favor of the formation of monolayer coatings. The data shown here correspond to mass measurements using an electrochemical quartz crystal microbalance (EQCM). Adapted with permission from ref (53). Copyright 2013 American Chemical Society. (C) Similar to the grafting of aryldiazonium ions, it is possible to electrochemically graft primary aliphatic amines onto carbon electrodes. (D) Relative to thiol–gold (S–Au) bonds, the resulting nitrogen–carbon (N–C) bonds are more stable at positive voltages (up to 0.7 V), resulting in monolayers that do not desorb for days under continuous voltage cycling. Adapted with permission from ref (57). Copyright 2022 under Creative Commons 4 license.

Like with aryldiazonium ions, it is also possible to form monolayers on electrodes via electrografting of aliphatic amines (Figure 5C). Pellitero et al. investigated this strategy to create biosensing interfaces that emulated the architecture of NBEs on carbon electrodes.⁵⁷ To determine the extent of electrode surface coating from various aliphatic amines, the authors employed electrode surface characterizations based on capacitive and Faradaic measurements. Additionally, the authors successfully created mixed electrografted layers that were conducive to the fabrication of NBEs. Unfortunately, through this work it became clear that nitrogen–carbon (N–C) bonds formed via electrografting can be electrochemically reduced at voltages more negative than −0.1 V vs Ag|AgCl, making this sensor fabrication strategy incompatible with the use of methylene blue as the NBEs redox reporter. However, the study also highlights that, at voltages more positive than 0.3 V, monolayers of aliphatic amines on carbon are significantly more stable than thiol monolayers on gold (Figure 5D). Therefore, these results point to the needed development of additional chemically stable and electrochemically reversible reporters with redox potentials more positive than that of methylene blue (i.e., −0.2 V vs Ag|AgCl). Learning from the extensive published work on aryldiazonium ions, future research on the fabrication of NBEs via electrografting of aliphatic amines should investigate the use of radical scavengers or protecting groups to increase monolayer density and suppress multilayer formation. If successful, the electrografting of aliphatic amines investigated by Pellitero at al.⁵⁷ could lead to a new class of carbon-supported NBEs with multiday stability under continuous cycling across positive (>0.3 V) voltage ranges.

Electrode-Agnostic Strategies

Although electrografting is technically an electrode-agnostic approach (i.e., electrografting can occur on carbon and metal surfaces), control over the structural characteristics of the resulting layer may vary significantly from electrode material to electrode material given, for example, differences in electrocatalytic activity. Therefore, the development of one common recipe to graft aryldiazonium ions or aliphatic amines electrochemically across an array of materials may not be possible. However, other strategies exist that offer truly electrode-agnostic monolayer formation. These strategies involve the physical deposition of conductive (or semiconductive) layers that (1) fully passivate the underlying electrode to prevent the occurrence of undesired electrochemical reactions and (2) offer a scaffold for the efficient coupling of oligonucleotides to create NBEs. Here I explore two approaches that address these requirements: conductive oxides and conductive polymers. However, several other strategies exist, each with pros and cons, which are discussed in detail elsewhere.^58,59

Conductive oxides are an attractive strategy for the fabrication of NBEs because they can be deposited as stable films on biocompatible plastic, glass, and metallic surfaces.⁶⁰ The prototypical examples of this class of materials are indium- (ITO) and fluorine-doped tin oxides (FTO), which are semitransparent and ideal for applications that require NBE sensing coupled to optical or spectroscopic imaging.⁶¹ Conductive oxides can be functionalized via self-assembly of organic layers based on an array of functional alkyl groups, including silanes, phosphonates, carboxylates, amines, catechols, and others.⁵⁹ Then, NBEs can be fabricated on the resulting monolayers via orthogonal coupling of oligonucleotides.

Although several studies have been published on the fabrication of oligonucleotide-modified conductive oxides, for example in refs (62 and 63), my laboratory reported one of the first studies specifically investigating the stability of NBEs formed on conductive oxides under continuous voltage modulation.⁶⁴ In this study, we created monolayers of alkylphospohonic acids on ITO, which we then functionalized with various oligonucleotide sequences to create NBEs (Figure 6A). Two critical challenges were highlighted by this study. First, the extent of surface coverage by the monolayer of alkyl phosphonates on ITO was low compared to that of benchmark thiols on gold, as judged by the large capacitive currents observed by voltammetry and the low Faradaic currents from reporter-modified oligonucleotides (Figure 6B, black arrows vs red arrow). The ability to better pack such monolayers and, therefore, better functionalize them with oligos likely depends on the structure of the underlying conductive oxide. Therefore, better sensing interfaces than those reported by our study may be attainable via careful engineering of the conductive oxide layer (the investigations in this study were conducted on generic ITO substrates purchased via Amazon). Second, ITO undergoes voltage-induced leaching of indium when in physiological buffers (as measured via inductively coupled plasma-mass spectrometry, Figure 6C, top), making this material not ideal for long-term sensing (Figure 6D). However, FTO may overcome this stability issue^65,66 (Figure 6C, bottom) and could be explored as a functional alternative to ITO for continuous NBE sensing.

Figure 6.

Electrode material-agnostic strategies to fabricate NBEs. (A) Alkylphosphonic acids can be leveraged to create monolayers on conductive oxides such as indium–tin oxide (ITO). Such monolayers can be modified in a subsequent step with oligonucleotides to create NBEs. Adapted from ref (64). Copyright 2023 Creative Commons 4. (B) Cyclic voltammograms of ITO-supported NBEs reveal large capacitive currents and small redox currents from methylene blue-modified oligonucleotides, which are indicative of worse electrode surface coverage relative to analogous thiol monolayers on gold. Adapted from ref (64). Copyright 2023 Creative Commons 4. (C) Top: voltammetric measurements coupled to inductively coupled plasma-mass spectrometry (ICP-MS) reveal a strong dependence of ITO stability on voltage. However, even without voltage bias, indium and tin ions from ITO can be detected in buffered solutions, an indication that ITO electrodes are not stable. Bottom: in contrast, FTO has a much broader window of stability relative to ITO. Red lines represent current density. Adapted with permission from ref (66). Copyright 2017 Creative Commons 4. (D) Because of the inherent lability of ITO, NBEs fabricated on ITO undergo faster signal loss relative to analogous NBEs fabricated on gold. Adapted from ref (64). Copyright 2023 Creative Commons 4. (E) A second electrode-agnostic strategy for NBEs fabrication involves the electrodeposition of conductive polymers, which can act as scaffolds for the covalent conjugation of oligonucleotides. Adapted from ref (67). Copyright 2023 Creative Commons 4. (F) Unfortunately, the conductivity of such polymers can be lost at negative voltages, adding the requirement of new redox reporters with positive reduction potentials. In this example, a tetrathiafulvalene derivative is shown, which achieves effective coupling to the conductive polymer surface by quickly decaying under repetitive voltage interrogation. Adapted from ref (67). Copyright 2023 Creative Commons 4.

A second strategy to create electrode agnostic sensing layers relies on the deposition of conductive polymers as a scaffold to which reporter-modified oligonucleotides can be appended for NBE sensing. Theoretically, conductive polymers can be electrodeposited, spin coated, or drop cast onto any electrode surface, with the structure of the polymer being somewhat independent of the electrode’s reactivity. Unfortunately, the vastness of the field of conductive polymers is such that a meaningful summary of research integrating these materials/interfaces into NBEs cannot be provided in this short perspective. However, in 2023 we provided a detailed account of the effect of monomer structure and class on the passivation of platinum electrodes via electrodeposition of conductive polymers.⁶⁷ This work discusses the possibility of using various polymeric interfaces for the tethering of redox reporters and redox reporter-modified oligonucleotides to platinum surfaces (Figure 6E). Additionally, this publication underscores important challenges that must be considered when planning the use of conductive polymers for NBEs development. For example, one important challenge involves controlling the polymer deposition protocol to prevent the formation of films that are too dense to allow functional coupling of oligonucleotides. This is the case of polymeric films made from aromatic amines such as phenylenediamine, which can be readily functionalized with small-molecule redox reporters but tend to be too dense to allow functional integration of oligonucleotides into the bulk of the film during orthogonal coupling. A second challenge involves determining the voltage window available for the interrogation of the polymers; for example, thiophenes allow continuous interrogation over voltages more positive than 0 V vs Ag|AgCl but quickly lose conductivity and, therefore, NBE sensing ability at more negative voltages. This issue led the authors to pursue the investigation of redox reporters with positive reduction potentials, like a tetrathiafulvalene derivative (Figure 6F). Although this reporter could be successfully coupled to the polymer film, it was not stable in chloride-containing buffers and, consequently, its redox currents quickly decayed under continuous interrogation (from first to fifth scan in Figure 6F).

Our work on conductive polymers could not successfully establish a strategy to develop NBEs for continuous molecular monitoring. However, it created a valuable experimental framework that should inspire future developments in this area. For example, the structure of the thiophene-based polymers considered in our work could be modified to allow the use of negative voltages without compromising the conductivity of the polymer. Additionally, the study did not test all possible monomer structures that lead to conductive polymers, and only considered a single electrode material, platinum. Given the vast scope of molecular structures that can be implemented to tune the conductivity and stability of the polymeric films, there is tremendous opportunity for future investigations in the context of long-term, continuous NBE sensing on conductive polymers.

Summary and Outlook

In this perspective I have discussed various strategies to develop NBEs outside of the benchmark standard of utilizing thiol monolayers on gold. Such strategies are important because thiol monolayers are chemically labile in biological environments and do not support all possible applications of NBE sensing, including cases where molecular monitoring coupled to imaging is needed in real time. Although the studies discussed here only offer preliminary demonstrations of NBE sensing on surfaces other than gold, as an ensemble they reflect a growing drive to translate NBEs to alternative sensing interfaces. I hope that our work in this area will stimulate the community in the fields of inorganic chemistry, organic synthesis, polymer sciences, and materials engineering to join our efforts and help us advance the field of NBEs.

I often get asked by students if I see NBEs being deployed in medical devices and diagnostics at a commercial level soon. From the perspective of my personal interactions with various startup and established companies, I respond that deployment of NBEs at a visible level across various human experiences should be occurring in the next 5 to 10 years. I think this is a conservative estimate. However, even if I am wrong, by inviting you to approach the challenges discussed in this work from your own creative perspective, together we can ensure that NBEs will play a role in biomedical research and medicine soon.

The author declares no competing financial interest.