- Email: [email protected]
Toward multimarker and functional assays from crude cell lysates: controlling spacing and signal amplification in DNA-CT–based bioelectrochemical devices
Available online at www.sciencedirect.com
ScienceDirect
Current Opinion in
Electrochemistry
Review Article
Toward multimarker and functional assays from crude cell lysates: controlling spacing and signal amplification in DNA-CT–based bioelectrochemical devices Ariel L. Furst1, Natalie B. Muren2 and Michael G. Hill2 Abstract
DNA-based electrochemical biosensors that rely on charge transport through DNA (DNA-CT) to detect biomarkers of interest have shown great promise in proof-of-principle studies because of their specificity, sensitivity, and low cost. However, this approach has not translated successfully into real clinical applications. One significant barrier has been the need to measure biomarkers in unpurified, complex cell and tissue lysates, which is difficult with DNA-modified electrodes because of crowding and nonspecific adhesion of biomolecules to the surface. Here, we discuss recent achievements in the control of DNA monolayer formation and the amplification of DNA CT signals that help to facilitate the detection of meaningful biomarkers from unprocessed, clinically relevant lysate samples. Addresses 1 Department of Chemistry, University of California, Berkeley, Berkeley, CA, 94720-1460, United States 2 Department of Chemistry and Chemical Biology, Occidental College, Los Angeles, CA, 90041, United States Corresponding author: Hill, Michael.G. ([email protected])
Current Opinion in Electrochemistry 2019, 14:104–112 This review comes from a themed issue on Bioelectrochemistry Edited by Elena Ferapontova and Miroslav Fotja For a complete overview see the Issue and the Editorial
present in low concentrations; clinically useful assays for these biomarkers would be compatible with unprocessed complex biological samples (e.g., whole blood, saliva, or crude cellular lysates). Currently, few assays of any kind have advanced to pointof-care use, with notable exceptions including pregnancy tests and portable electronic glucose sensors [1]. These tests share the same reasons for their success: (i) the single target that they detect (human chorionic gonadotropin and glucose, respectively) carries an unambiguous meaning that dictates patient care and (ii) the binary presence or absence of that target, along with its quantification, provides a clinically actionable result. Far more commonly, however, tests based on quantifying a single biomarker often fail to yield clinically useful results [2]. Indeed, there is an incipient drive toward diagnostic targets comprising entire panels of biomarkers, as these collections can provide more precision in characterizing an individual’s true disease state. Such panels now exist for conditions ranging from pancreatic cancer [3], to Alzheimer’s disease [4], to sepsis [5,6]. Perhaps, an even more promising tack is to develop clinical tests based on functional assays of key biomarkers versus simple detection and quantification. This latter approach has garnered recent interest from both bench-top scientists and clinicians [7,8].
Available online 11 January 2019 https://doi.org/10.1016/j.coelec.2018.12.008 2451-9103/© 2019 Elsevier B.V. All rights reserved.
Keywords DNA electrochemistry, Cell lysate, Activity assay, Charge transfer.
Introduction With the ever-growing body of ‘eomics’ data available to researchers and clinicians, there is a push to develop medical treatments that are tailored to the unique genomic, proteomic, metabolomic, and epigenomic composition of individuals. Within this context, a central diagnostic goal is to develop high-throughput, point-ofcare assays to report accurately on the identity and progress of any disease states. Traditionally, this has meant analyzing for specific molecular biomarkers, often Current Opinion in Electrochemistry 2019, 14:104–112
The added complexities of multitarget and/or functional assays, especially within the constraints of an inexpensive point-of-care application, present a singular challenge in clinical diagnostics design. Owing to their ease of use, high sensitivity, and low cost, electrochemical-based platforms have attracted considerable interest toward this goal [9,10]. In particular, DNA-modified electrodes have been widely examined, as the inherent specificity of DNA for biomoleculebased recognition allows it to function as an effective sensing element [11]. There are two general strategies to use DNA as a recognition element (Figure 1). The most common approach relies on nucleic acid hybridization, either direct capture of a complementary nucleic acid target or the use of redox-labeled DNA hairpin constructs or www.sciencedirect.com
DNA-CT-based electrochemical sensors Furst et al.
105
Figure 1
Current Opinion in Electrochemistry
DNA-based electrochemical detection of biomolecule targets. Common strategies involving DNA hybridization (left column) and DNA CT (right column) are illustrated. Hybridization: electrodes are covalently modified with a single-stranded probe sequence, and changes in hybridization upon target binding produces a change in the electrochemical signal. Variations of this strategy include the following: (a) Single-stranded DNA capture; upon hybridization of target DNA, electroactive molecules (blue spheres) are recruited to the duplex and produce an electrochemical signal; (b) Labeled aptamer DNA; upon binding by target proteins or small molecules (purple crescent), the electroactive molecule is brought closer to the electrode surface and produces a signal; (c) Labeled hairpin DNA; hybridization of complementary target DNA causes the hairpin to unfold, separating the electroactive molecule from the surface and turning the signal off. DNA CT: electrodes are modified with single-stranded or duplex DNA, and redox-active reporter molecules that are electronically coupled to the DNA p-stack allow for the detection of intervening biomolecules which either complete (signal on) or perturb (signal off) this stack. Common strategies include the following: (d) Capture of complementary nucleic acids which complete the DNA duplex and produce an electrochemical signal; (e) abasic sites and other DNA lesions; or (f) Certain DNA binding proteins, both of which can disrupt the DNA p-stack and attenuate the electrochemical signal.
aptamers. In these latter constructs, target binding causes a conformational change that alters the electrochemical signal. An alternative signaling strategy relies on the unique property of double-stranded DNA to mediate long-range charge transport (DNA-CT) [12]. By modifying electrodes with DNA duplexes that feature electronically coupled, redox-active reporter molecules, a DNA-mediated electrochemical signal that depends on the integrity of the intervening DNA bases can be measured [13]. Because DNA CT is interrupted by DNA lesions, mismatches, and bound proteins that perturb the continuous p-stack, this is an extremely sensitive strategy for detecting biomarkers that bind to a specific DNA sequence [9]. Although DNA-based electrochemical approaches have shown promise in laboratory proof-of-principle studies, they have not translated into successful clinical applications. A major obstacle has been developing assays compatible with unpurified, complex biological samples: excessive sample processing can result in unintentional elimination of biomarker targets and precludes assessing www.sciencedirect.com
protein (or other biomolecule) activity in a native biological matrix. This task is particularly difficult using DNA-modified electrodes because of the propensity of biomolecules to adhere nonspecifically to electrode surfaces. Practically, overcoming this challenge involves assembling electrode surfaces with enough space between the probe DNA sequences to provide access for target binding in the constrained surface environment. Moreover, with the decreased number of DNA recognition elements and the use of dilute samples, there is an additional need to amplify the correspondingly lower electrochemical signals. Here, we focus on DNA CTe based detection schemes, describing recent work that we and others have performed to address these issues to move DNA-modified electrode detection closer to clinical utility.
Interhelical spacing in DNA films DNA-modified electrodes are typically formed via selfassembly of thiolated duplexes onto gold, followed by passivation of any uncovered surface with a small organic thiol such as mercaptohexanol (Figure 2a) [14]. As Current Opinion in Electrochemistry 2019, 14:104–112
106 Bioelectrochemistry
interhelical interactions of the DNA dominate the selfassembly process, even ‘low-density’ films tend to appear as islands of close-packed helices within a sea of passivating reagent. This inhomogeneity decreases both the sensitivity and reproducibility of detection. In recent years, three main strategies for controlling probe spacing and surface homogeneity have emerged that are as follows: manipulating the underlying electrode morphology [15e17]; developing next-generation passivation methods for DNA self-assembly [18]; and chemically controlling DNA attachment to a preformed monolayer [19,20,21]. Owing to their simplicity, consistency, and high level of control, we focus here on chemical (as opposed to physical) strategies to control DNA self-assembly. Improving self-assembled mixed monolayers
The Ferapontova group has significantly advanced the detection capabilities of self-assembled DNA monolayers by manipulating the nature of the adjuvant underlying passivating film [18]. Using a polyethylene glycol (PEGylated-) thiol (Figure 2b), they were able to suppress nonspecific adsorption of serum proteins onto dilute DNA monolayers, while still maintaining extremely efficient passivation against the DNAbinding electrochemical reporter molecules (methylene blue [MB] and ferricyanide), thus preventing false signals by direct electrochemical reduction. Using this PEGylated passivating agent and a DNA probe that featured an aptamer sequence specific for the cancer biomarker Her-2/neu, they successfully detected picomolar concentrations of the protein in serumd conditions amenable for clinical applications. Using click chemistry to assemble DNA
Enhancing target capture from complex mixtures also has been achieved by controlling the spacing and density of DNA duplexes on electrode surfaces via orthogonal coupling based on click chemistry (Figure 2c). Here, a mixed alkanethiol monolayer doped with a small fraction of molecules that possess a reactive head group is first self-assembled onto a gold electrode. Then, an appropriately labeled DNA duplex is coupled to the surface. Changing the concentration of the reactive head group in the underlying monolayer allows for homogeneous surfaces with controlled DNA probe duplex loadings. One strategy for this type of coupling involves surface attachment based on catalyst-free [3 þ 2] cycloaddition between an azide head group and a DNA duplex modified with a strained cyclooctyne [19]. The resulting density of DNA on the electrode surface is directly proportional to the mole fraction of azide in the underlying monolayer. Importantly, atomic force microscopy reveals homogenous films with minimal DNA aggregation, and these monolayers display key characteristics Current Opinion in Electrochemistry 2019, 14:104–112
of DNA-mediated electrochemistry, including signal decrease upon incorporation of a mismatch. Direct electrochemistry of reporter molecules intercalated into the DNA helices is extremely sensitive to proteinbinding events on this surface, with the ability to detect transcription factors well below the solution KD of the protein. Activatable catalysis for click chemistry
Cu(I)-catalyzed Huisgen 1,3-dipolar cycloaddition has been widely used to prepare functionalized monolayers [22]. In these systems, the Cu(I) is typically generated in situ using either a chemical reductant or by electrochemical reduction of a Cu(II) precursor. Unfortunately, electrochemical reduction of common water-soluble Cu(II) precursors often results in coppercontaining deposits on the electrode surface, impeding subsequent DNA CT to DNA-bound reporter molecules. One solution to this problem is to use a twoelectrode DNA-grafting platform in which active Cu(I) is generated at an auxiliary microelectrode [20,21]. In this system (Figure 2d), the microelectrode is positioned above a mixed monolayer doped with terminal azide groups: in the presence of alkyne-terminated DNA, electrochemical reduction of Cu(II) triggers DNA coupling to the underlying film in a spatially controlled manner. Any electrochemically induced copper deposits occur at the auxiliary electrode, shielding the DNA monolayer from damage. Direct electrochemistry of intercalated reporter molecules at a patterned array featuring different DNA sequences allowed simultaneous detection at physiological concentrations of multiple transcription factors from a complex mixture. Indeed, with appropriate signal amplification [23], these surfaces are sufficiently selective to allow detection of methyltransferase activity directly from human tumor tissue via DNA CT [24,25]. Reagentless oxidative coupling
By combining features of chemical-based cyclooctyne surface coupling with electrochemical catalyst activation, a reagentless method was recently developed to control DNA coupling to gold surfaces from a single pot [26]. This technique (Figure 2e) relies on a suite of oxidative coupling reactions developed in the Francis group for site-specific bioconjugations [27]. The key coupling reaction between substituted aniline and catechol is typically initiated by a mild chemical oxidant (either ferricyanide or periodate). However, in many cases, chemical oxidants are incompatible with sensitive biomolecules. Therefore, electrode-surface coupling of DNA was accomplished by direct electrochemical oxidation of catechol. Using this technology, aniline, alkyl thiols, and amine-modified DNA were all successfully coupled to catechol-modified electrodes. Interestingly, DNA-modified electrodes prepared in this www.sciencedirect.com
DNA-CT-based electrochemical sensors Furst et al.
107
Figure 2
Current Opinion in Electrochemistry
Strategies for controlling the spacing and homogeneity of DNA-modified electrodes. (a) In conventional monolayer formation, DNA sequences modified with a terminal alkanethiol self-assemble onto gold electrodes, followed by treatment with small thiol-containing molecules, such as mercaptohexanol, to cover and passivate any remaining electrode surface. (b) After the self-assembly of alkanethiol-modified DNA, a small molecule PEG-thiol can also be used as a more effective surface passivation agent for biological samples. (c) The spacing of individual DNA duplexes can be controlled by first forming a mixed-alkylthiol monolayer doped with chemically active head groups that can be later coupled to DNA. (d) When a terminal azide is used as the active www.sciencedirect.com
Current Opinion in Electrochemistry 2019, 14:104–112
108 Bioelectrochemistry
Figure 3
Current Opinion in Electrochemistry
Immobilization of electroactive cells onto an electrode allows for DNA CT–dependent current generation. Cells modified with single-stranded DNA are directed to adhere specifically to an electrode modified with the complementary sequence through DNA hybridization. When fully complementary DNA sequences are used, the cells generate more current (WM DNA) than either free cells or cells immobilized with mismatched (MM) DNA sequences, indicating a DNA CT–mediated pathway.
fashion were used recently to immobilize electron transfereproficient cells onto electrode surfaces (Figure 3). Thus, in addition to providing a route to whole-cell displays, these constructs enable the expansion of DNA CT to green energy technology [28].
Generation and amplification of electrochemical signals With straightforward (and scalable) strategies available for preparing DNA-modified electrodes to capture biomarkers from complex mixtures, there is still the challenge of generating sufficiently large signaling currents for clinical use. Recent work has focused on three main strategies: direct electrochemical readout; electrocatalysis; and collectoregenerator amplification using a bipotentiostat. Direct readout
In the simplest case, electrochemical readout can be accomplished by comparing the electrochemical signal
of a redox-active reporter molecule (e.g., MB, Nile blue, or daunomycin [9]) associated into DNA before versus after a signaling event (e.g., capture of a small-molecule target, binding of a protein, etc.). For helix-dependent DNA-CT assays, it is important that the reporter molecules intercalate into the p-stack [13]. This type of readout was recently used by Kahanda et al [29] in a functional assay for the activity of a chemotherapeutic agent in raw cell lysate. b-Lapachone is an orthoquinone anticancer drug that acts as a substrate for NAD(P)H: quinone oxidoreductase 1 (NQO1)dan NAD(P)H-driven enzyme overexpressed in a number of aggressive solid tumors. Upon reduction by NQO1, the resulting hydroquinone is rapidly reoxidized to b-lapachone, with concomitant generation of H2O2 and depletion of NAD(P)H. This redox cycling results in the buildup of reactive oxygen species, leading to the formation of 8-oxoguanine. Repair enzymes subsequently excise the damaged bases, leading to strand scissions and abasic sites within the DNA.
head group in this mixed monolayer, DNA can be patterned onto specific regions of an electrode by using a secondary electrode to activate an inert precatalyst and attach oligonucleotides onto particular locations via click chemistry. (e) A one-pot, electrochemical coupling method between a catecholterminated monolayer and aniline-modified DNA enables precise control over the final surface density of DNA. PEG, polyethylene glycol. Current Opinion in Electrochemistry 2019, 14:104–112
www.sciencedirect.com
DNA-CT-based electrochemical sensors Furst et al.
109
Figure 4
Current Opinion in Electrochemistry
Amplification strategies for DNA-CT detection. (a) Conventional electrocatalysis (left) versus collector-generator amplification (right). Electrocatalysis (left) involves amplification of the DNA-mediated redox signal of a DNA-intercalating probe (methylene blue, MB) with a small signaling molecule (ferricyanide) in solution. MB is reduced to leucomethylene blue (LB) by DNA CT, but is rapidly reoxidized back to MB by ferricyanide. If the individual electron-transfer rates are sufficiently large, the resulting current is limited only by ferricyanide diffusion, providing a substantially larger signal compared to direct electrochemistry. Incorporation of a secondary electrode (right) in collector-generator mode allows for the regeneration of ferricyanide at rates that exceed diffusion. Given the efficiency of long-range DNA-CT, this detection strategy generates significantly more current than conventional electrocatalysis www.sciencedirect.com
Current Opinion in Electrochemistry 2019, 14:104–112
110 Bioelectrochemistry
In an elegant assay, DNA duplexes featuring a covalent redox reporter molecule were self-assembled onto an electrode array to serve as a stand-in for cellular DNA. After treating these surfaces with cell lysate from tumor samples, the electrochemical signal arising from DNACT to the redox probe was then monitored upon addition of b-lapachone. The more damage to the individual helices within the film caused by futile b-lapachone redox cycling, the more attenuation in the electrochemical signal. Using a pulsed voltammetric readout technique (Osteryoung square-wave voltammetry) helped isolate the relatively small Faradaic currents from charging current background. This assay nicely exploits DNA-CT, as the end products of b-lapachone activity are charge transportedisrupting lesions of the p-stack. Notably, because formation of DNA lesions depends also on the activity of downstream repair proteins and other factors specific to the unique cell biology of the cancer cell, drug-induced DNA damage can deviate markedly from NQO1 expression levels. Thus, this DNA-CTe based functional assay provides critical clinical data that would be inaccessible using conventional biomarker analyses that simply quantify the expression level of a single component of a multifactorial pathway. Electrocatalysis
The finite surface concentrations of DNA-bound redox reporter molecules (determined by the electrode area and DNA surface density) ultimately limit the inherent sensitivity of direct electrochemical readout, highlighting the need for selective Faradaic signal amplification. This is commonly accomplished by an electrocatalytic cycle in which a DNA-bound reporter (e.g., MB) mediates the reduction of a soluble redox substrate (e.g., ferricyanide). As long as the substrate is blocked from interacting with the electrode surface, its reduction occurs only upon DNA-CT to the reporter. Because the ultimate electrochemical substrate is in the solution phase, the current is no longer limited by the amount of redox-probe molecule that can be adsorbed onto the electrode surface. In addition to electrocatalytic readout of low concentrations of circulating blood markers [18], non-DNAemediated electrocatalytic cycles (e.g., electrostatically bound Ru(NH3)3þ 6 /ferricyanide pairs) have been widely used for an impressive host of biomarker detection assays [10,11,30]. Collector/generator amplification
Although electrocatalysis has enabled detection of a broader range of analytes, often at lower concentrations,
the currents remain limited by the diffusion of the soluble redox signaling substrate. Moreover, under practical operating conditions, there is typically background current owing to some direct substrate reduction because of incomplete electrode passivation. Thus, while detection can be identified in a controlled laboratory setting by monitoring the ‘percent difference’ between electrocatalytic versus background signals with and without an analyte present, the absolute change in current may be quite small compared with the overall baseline currents. One approach to enhance the signal-to-noise ratio is to use a bipotentiostat to drive a two-electrode system in collector-generator mode, Figure 4 [31], Combined with DNA-CT, this alignment virtually eliminates backgrounds, as the signaling reporter is generated only via DNA-mediated processes; moreover, orienting the collector and generator electrodes at sufficiently small spacings allows recycling of the signaling reporter at rates that exceed the diffusion limit (cf. positive feedback in scanning electrochemical microscopy) [32]. We have applied this collector-generator concept to macroscopic platforms [20,21]. This amplification platform has proven useful in the development of an activity assay for the human methyltransferase DNMT1 from crude tumor cell lysate (Figure 4b). DNMT1 plays a critical role in the epigenetic regulation of gene expression, and aberrant DNA methylation by DNMT1 has been associated with tumorigenesis in a wide variety of cancers [33]. This electrochemical assay [34] is based on a two-step procedure in which electrode-bound DNA substrates are first treated with DNMT1, followed by incubation with a restriction enzyme that is incompetent toward fully methylated duplexes. DNMT1 activity, therefore, provides protection from excision by the restriction enzyme, and methyltransferase activity is reflected in the magnitude of the electrochemical signal that remains after restriction enzyme treatment. By carefully controlling interhelical DNA spacing during the selfassembly process and switching to a two-electrode detection platform, this assay was adapted to provide robust and high-throughput measurement of human DNMT1 activity in the crude lysate of both cultured cells [24] and patient tumor samples [25]. Most notably, the measured DNMT1 activity of tumor tissue did not correlate with DNMT1 expression or abundance, as measured by reverse-transcription polymerase chain reaction (RT-PCR) and Western blot, yet that activity was the most consistent indicator of tumor tissue
(center). (b) Electrochemical assay to detect methyltransferase activity on a two-electrode, electrocatalytic platform. Electrodes are prepared with DNA that contains overlapping recognition sites for the human methyltransferase DNMT1 and a restriction enzyme. Treatment of the electrodes with a crude cell lysate allows for methylation of the electrode-bound DNA by active DNMT1 in the lysate. Following this, the electrode is treated with a restriction enzyme and the electrocatalytic signal is measured again. As methylation protects the DNA from being cut from the surface by the restriction enzyme, the level of methyltransferase activity in the cell lysate sample is reflected by the magnitude of the electrochemical signal that remains after restriction enzyme treatment. The presence of active DNMT1 in the sample results in maintenance of the ‘on’ electrochemical signal (top) while the absence of active DNMT1 in the sample causes the electrochemical signal to be turned off (bottom). Current Opinion in Electrochemistry 2019, 14:104–112
www.sciencedirect.com
DNA-CT-based electrochemical sensors Furst et al.
[20]. This finding again highlights the importance of functional assays, conducted within a native matrix, to provide a handle for monitoring disease states that is more reliable than biomarkers which are present further upstream in a pathological process.
Conclusions It is becoming increasingly clear that routine diagnostic evaluation of unprocessed biological samples represents a great opportunity for improving the clinical management of complicated disease processes. Yet the complexity of these samples has hampered the development of high-throughput devices suitable for pointof-care analyses. In working toward next-generation DNA-based electrochemical biosensors, recent advances in controlling the surface morphology of modified electrodes, coupled with enhanced amplification methods, provide new design elements for creating diagnostic strategies tailored for specific biomarker panels and/or functional assays of multicomponent biological pathways. Indeed, DNA-CT activity assays on crude lysates have begun to yield useful information relevant to therapeutic-drug efficacy as well as disease etiology. Leveraging progress in platform design with transformative advances in chemical biology opens the door for an increasingly important role for DNA-based electrochemical devices in translating basic research findings into practical clinical tools.
Conflict of interest A.L.F. was supported by the A. O. Beckman Postdoctoral Fellowship. NBM has no conflict of interest. M.G.H. thanks the John Stauffer Charitable Trust and Fletcher Jones Foundation for support.
References Papers of particular interest, published within the period of review, have been highlighted as: of special interest of outstanding interest 1.
da Silva ETSG, Souto DEP, Barragan JTC, Giarola J de F, de Moraes ACM, Kubota LT: Electrochemical biosensors in pointof-care devices: recent advances and future trends. ChemElectroChem 2017, 4:778–794, https://doi.org/10.1002/ celc.201600758.
2.
Rifai N, Gillette MA, Carr SA: Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nat Biotechnol 2006, 24:971 – 983, https://doi.org/ 10.1038/nbt1235.
3.
Liu J, Gao J, Du Y, Li Z, Ren Y, Gu J, Wang X, Gong Y, Wang W, Kong X: Combination of plasma microRNAs with serum CA19-9 for early detection of pancreatic cancer. Int. J. Cancer 2012, 131:683–691, https://doi.org/10.1002/ijc.26422.
4.
Frisoni GB, Boccardi M, Barkhof F, Blennow K, Cappa S, Chiotis K, Démonet JF, Garibotto V, Giannakopoulos P, Gietl A, Hansson O, Herholz K, Jack CR, Nobili F, Nordberg A, Snyder HM, Ten Kate M, Varrone A, Albanese E, Becker S, Bossuyt P, Carrillo MC, Cerami C, Dubois B, Gallo V, Giacobini E, Gold G, Hurst S, Lönneborg A, Lovblad KO, Mattsson N, Molinuevo JL, Monsch AU, Mosimann U, Padovani A, Picco A, Porteri C, Ratib O, Saint-Aubert L, Scerri C, Scheltens P,
www.sciencedirect.com
111
Schott JM, Sonni I, Teipel S, Vineis P, Visser PJ, Yasui Y, Winblad B: Strategic roadmap for an early diagnosis of Alzheimer’s disease based on biomarkers. Lancet Neurol 2017, 16:661–676, https://doi.org/10.1016/S1474-4422(17) 30159-X. 5.
Pierrakos C, Vincent JL: Sepsis biomarkers: a review. Crit Care 2010, 14:R15, https://doi.org/10.1186/cc8872.
6.
Dolin HH, Papadimos TJ, Stepkowski S, Chen X, Pan ZK: A novel combination of biomarkers to herald the onset of sepsis prior to the manifestation of symptoms. Shock 2018, 49:364–370, https://doi.org/10.1097/SHK.0000000000001010.
7.
Senft D, Leiserson MDM, Ruppin E, Ronai ZA: Precision oncology: the road ahead. Trends Mol Med 2017, 23:874–898, https://doi.org/10.1016/j.molmed.2017.08.003.
[8].
Furst AL, Hoepker AC, Francis MB: Quantifying hormone disruptors with an engineered bacterial biosensor. ACS Cent Sci 2017, 3:110–116, https://doi.org/10.1021/ acscentsci.6b00322. This manuscript describes a novel electrochemical strategy for detecting a broad spectrum of endocrine-disrupting toxins based upon binding to a native protein receptor displayed on the surface of engineered E. coli. By reporting total estrogenic activity, this approach enables detection of endocrine-disrupting chemicals without requiring knowledge their structure. It represents a fundamental shift from standard assays, in which a specific capture element is constructed for a single biomarker. 9.
Drummond TG, Hill MG, Barton JK: Electrochemical DNA sensors. Nat Biotechnol 2003, 21:1192–1199, https://doi.org/ 10.1038/nbt873.
10. Boon EM, Ceres DM, Drummond TG, Hill MG, Barton JK: Mutation detection by electrocatalysis at DNA-modified electrodes. Nat Biotechnol 2000, 18:1096–1100, https://doi.org/ 10.1038/80301. [11]. Labib M, Sargent EH, Kelley SO: Electrochemical methods for the analysis of clinically relevant biomolecules. Chem Rev 2016, 116:9001–9090, https://doi.org/10.1021/ acs.chemrev.6b00220. This comprehensive review catalogs the diverse targets and electrochemical strategies under development for biomolecule detection. [12]. Beratan DN, Naaman R, Waldeck DH: Charge and spin transport through nucleic acids. Curr Opin Electrochem 2017, 4:175–181, https://doi.org/10.1016/J.COELEC.2017.08.017. This paper summarizes some of the recent developments in theoretical efforts to understand the phenomena of long-range charge transport through the double helix. [13]. Kékedy-Nagy L, Ferapontova EE: Sequence-specific electron transfer mediated by DNA duplexes attached to gold through the alkanethiol linker. J Phys Chem B 2018, 122: 10077–10085, https://doi.org/10.1021/acs.jpcb.8b07505. This study finds that the heterogeneous electron-transfer rate of MB reduction is significantly faster at DNA-modified surfaces featuring ATversus GC-duplexes. The rates correlate with the different binding modes of MB at these two surfaces (groove binding at AT versus intercalation at GC), emphasizing the importance of sequence context when designing DNA-CT assays that depend on intercalated probe molecules. 14. Levicky R, Herne TM, Tarlov MJ, Satija SK: Using self-assembly to control the structure of DNA monolayers on gold: a neutron reflectivity study. J Am Chem Soc 1998, 120: 9787–9792, https://doi.org/10.1021/ja981897r. 15. Elliott Joanne M, Birkin Peter R, Bartlett Philip N, Attard GS: Platinum microelectrodes with unique high surface areas. 1999, https://doi.org/10.1021/LA9908945. 16. Mahshid SS, Vallée-Bélisle A, Kelley SO: Biomolecular steric hindrance effects are enhanced on nanostructured microelectrodes. Anal Chem 2017, 89:9751–9757, https://doi.org/ 10.1021/acs.analchem.7b01595. 17. Arroyo-Currás N, Scida K, Ploense KL, Kippin TE, Plaxco KW: High surface area electrodes generated via electrochemical roughening improve the signaling of electrochemical aptamer-based biosensors. Anal Chem 2017, 89: 12185–12191, https://doi.org/10.1021/acs.analchem.7b02830.
Current Opinion in Electrochemistry 2019, 14:104–112
112 Bioelectrochemistry
[18]. Salimian R, Kékedy-Nagy L, Ferapontova EE: Specific pico molar detection of a breast cancer biomarker HER-2/neu protein in serum: electrocatalytically amplified electroanalysis by the aptamer/PEG-modified electrode. ChemElectroChem 2017, 4:872–879, https://doi.org/10.1002/ celc.201700025. Incorporating an aptamer sequence within an electrode-bound DNA films, a simple chemical blocking stratey for non-specific protein adsorption, and an electrocatalytic amplification strategy, this platform allows for screening of an important cancer biomarker from circulating blood, potentially obviating unnecessary and invasive biopsies. 19. Furst AL, Hill MG, Barton JK: DNA-modified electrodes fabricated using copper-free click chemistry for enhanced protein detection. Langmuir 2013, 29:16141–16149, https://doi.org/ 10.1021/la403262v. 20. Furst A, Landefeld S, Hill MG, Barton JK: Electrochemical patterning and detection of DNA arrays on a two-electrode platform. J Am Chem Soc 2013, 135:19099–19102, https:// doi.org/10.1021/ja410902j. 21. Furst AL, Hill MG, Barton JK: A multiplexed, two-electrode platform for biosensing based on DNA-mediated charge transport. Langmuir 2015, 31:6554–6562, https://doi.org/ 10.1021/acs.langmuir.5b00829. 22. Orain C, Le Poul N, Gomila A, Kerbaol J-M, Cosquer N, Reinaud O, Conan F, Le Mest Y: A generic platform for the addressable functionalisation of electrode surfaces through self-induced “electroclick. Chem - A Eur J 2012, 18:594–602, https://doi.org/10.1002/chem.201102620. 23. Furst AL, Hill MG, Barton JK: Electrocatalysis in DNA sensors. Polyhedron 2014, 84:150–159, https://doi.org/10.1016/ j.poly.2014.07.005. 24. Furst AL, Muren NB, Hill MG, Barton JK: Label-free electrochemical detection of human methyltransferase from tumors. Proc Natl Acad Sci Unit States Am 2014, 111:14985–14989, https://doi.org/10.1073/pnas.1417351111. [25]. Furst AL, Barton JK: DNA electrochemistry shows DNMT1 methyltransferase hyperactivity in colorectal tumors. Chem Biol 2015, 22:938–945, https://doi.org/10.1016/ j.chembiol.2015.05.019. In this study, the DNMT1 detection platform discribed in reference 24 was appied to human colorectal-tumor specimens. Significantly, while the electrochemical results in crude cell lysate correlate closely with the much more time-consuming (and expensive) standard tritium-labeling activity assay, they vary substantially from DNMT1 expression levels as measured by RT-PCR and western blotting. This finding underscores the importance of a functional approach for clinically relevant diagnostics involving complex biological pathways. 26. Furst AL, Smith MJ, Francis MB: Direct electrochemical bioconjugation on metal surfaces. J Am Chem Soc 2017, 139: 12610–12616, https://doi.org/10.1021/jacs.7b06385.
Current Opinion in Electrochemistry 2019, 14:104–112
27. Elsohly AM, Francis MB: Development of oxidative coupling strategies for site-selective protein modification. Acc Chem Res 2015, 48:1971–1978, https://doi.org/10.1021/ acs.accounts.5b00139. [28]. Furst AL, Smith MJ, Lee MC, Francis MB: DNA hybridization to interface current-producing cells with electrode surfaces. ACS Cent Sci 2018, 4:880–884, https://doi.org/10.1021/ acscentsci.8b00255. Shewanella oneidensis is often used as a component of bioelectrochemical cells, but gives inconsistent currents as electron injection normally depends upon biofilm formation. As highlighted in this study, wiring cells through DNA duplexes using controlled monolayer assembly highlights the ability of DNA CT to be used as both a conduit and a method to read out current production. [29]. Kahanda D, Singh N, Boothman DA, Slinker JD: Following anticancer drug activity in cell lysates with DNA devices. Biosens Bioelectron 2018, 119:1–9, https://doi.org/10.1016/ J.BIOS.2018.07.059. As described in the text, this work describes a functional electrochemical assay for a small-molecule chemotherapeutic. Operating via drug-induced enzymatic NAD(P)H reduction of dioxygen to reactive oxygen species (peroxide, superoxide, etc.), the drug is recycled many times thus providing inherent signal amplification. Because the downstream effect of this redox cycling is DNA damage that perturbs p stacking, DNA-CT is particularly well suited as a detection strategy. It is noteworthing that the degree of DNA damage does not necessarily correlate with target-enzyme expresion levels, stressing the importance of a simple activity assay performed under cell-specific conditions. 30. Fang Z, Kelley SO: Direct electrocatalytic mRNA detection using PNA-nanowire sensors. Anal Chem 2009, 81:612–617. 31. McClintock SA, Purdy WC: A bipotentiostat for four-electrode electrochemical detectors. Anal Lett 1981, 14:791–798, https:// doi.org/10.1080/00032718108055483. 32. Gorodetsky AA, Hammond WJ, Hill MG, Slowinski K, Barton JK: Scanning electrochemical microscopy of DNA monolayers modified with nile blue. Langmuir 2008, 24:14282–14288, https://doi.org/10.1021/la8029243. 33. Zhang W, Xu J: DNA methyltransferases and their roles in tumorigenesis. Biomark Res 2017, 5:1, https://doi.org/10.1186/ s40364-017-0081-z. [34]. Muren NB, Barton JK: Electrochemical assay for the signal on detection of human DNA methyltransferase activity. J Am Chem Soc 2013, 135:16632–16640, https://doi.org/10.1021/ ja4085918. This assay allowed for the first measurements of human DNMT1 using an electrochemical platform. Notably, when tested with crude lysates, cellular debris from the lysates adhered nonspecifically to the electrode, preventing access of methyltransferases to the electrode-bound DNA.
www.sciencedirect.com
ScienceDirect
Current Opinion in
Electrochemistry
Review Article
Toward multimarker and functional assays from crude cell lysates: controlling spacing and signal amplification in DNA-CT–based bioelectrochemical devices Ariel L. Furst1, Natalie B. Muren2 and Michael G. Hill2 Abstract
DNA-based electrochemical biosensors that rely on charge transport through DNA (DNA-CT) to detect biomarkers of interest have shown great promise in proof-of-principle studies because of their specificity, sensitivity, and low cost. However, this approach has not translated successfully into real clinical applications. One significant barrier has been the need to measure biomarkers in unpurified, complex cell and tissue lysates, which is difficult with DNA-modified electrodes because of crowding and nonspecific adhesion of biomolecules to the surface. Here, we discuss recent achievements in the control of DNA monolayer formation and the amplification of DNA CT signals that help to facilitate the detection of meaningful biomarkers from unprocessed, clinically relevant lysate samples. Addresses 1 Department of Chemistry, University of California, Berkeley, Berkeley, CA, 94720-1460, United States 2 Department of Chemistry and Chemical Biology, Occidental College, Los Angeles, CA, 90041, United States Corresponding author: Hill, Michael.G. ([email protected])
Current Opinion in Electrochemistry 2019, 14:104–112 This review comes from a themed issue on Bioelectrochemistry Edited by Elena Ferapontova and Miroslav Fotja For a complete overview see the Issue and the Editorial
present in low concentrations; clinically useful assays for these biomarkers would be compatible with unprocessed complex biological samples (e.g., whole blood, saliva, or crude cellular lysates). Currently, few assays of any kind have advanced to pointof-care use, with notable exceptions including pregnancy tests and portable electronic glucose sensors [1]. These tests share the same reasons for their success: (i) the single target that they detect (human chorionic gonadotropin and glucose, respectively) carries an unambiguous meaning that dictates patient care and (ii) the binary presence or absence of that target, along with its quantification, provides a clinically actionable result. Far more commonly, however, tests based on quantifying a single biomarker often fail to yield clinically useful results [2]. Indeed, there is an incipient drive toward diagnostic targets comprising entire panels of biomarkers, as these collections can provide more precision in characterizing an individual’s true disease state. Such panels now exist for conditions ranging from pancreatic cancer [3], to Alzheimer’s disease [4], to sepsis [5,6]. Perhaps, an even more promising tack is to develop clinical tests based on functional assays of key biomarkers versus simple detection and quantification. This latter approach has garnered recent interest from both bench-top scientists and clinicians [7,8].
Available online 11 January 2019 https://doi.org/10.1016/j.coelec.2018.12.008 2451-9103/© 2019 Elsevier B.V. All rights reserved.
Keywords DNA electrochemistry, Cell lysate, Activity assay, Charge transfer.
Introduction With the ever-growing body of ‘eomics’ data available to researchers and clinicians, there is a push to develop medical treatments that are tailored to the unique genomic, proteomic, metabolomic, and epigenomic composition of individuals. Within this context, a central diagnostic goal is to develop high-throughput, point-ofcare assays to report accurately on the identity and progress of any disease states. Traditionally, this has meant analyzing for specific molecular biomarkers, often Current Opinion in Electrochemistry 2019, 14:104–112
The added complexities of multitarget and/or functional assays, especially within the constraints of an inexpensive point-of-care application, present a singular challenge in clinical diagnostics design. Owing to their ease of use, high sensitivity, and low cost, electrochemical-based platforms have attracted considerable interest toward this goal [9,10]. In particular, DNA-modified electrodes have been widely examined, as the inherent specificity of DNA for biomoleculebased recognition allows it to function as an effective sensing element [11]. There are two general strategies to use DNA as a recognition element (Figure 1). The most common approach relies on nucleic acid hybridization, either direct capture of a complementary nucleic acid target or the use of redox-labeled DNA hairpin constructs or www.sciencedirect.com
DNA-CT-based electrochemical sensors Furst et al.
105
Figure 1
Current Opinion in Electrochemistry
DNA-based electrochemical detection of biomolecule targets. Common strategies involving DNA hybridization (left column) and DNA CT (right column) are illustrated. Hybridization: electrodes are covalently modified with a single-stranded probe sequence, and changes in hybridization upon target binding produces a change in the electrochemical signal. Variations of this strategy include the following: (a) Single-stranded DNA capture; upon hybridization of target DNA, electroactive molecules (blue spheres) are recruited to the duplex and produce an electrochemical signal; (b) Labeled aptamer DNA; upon binding by target proteins or small molecules (purple crescent), the electroactive molecule is brought closer to the electrode surface and produces a signal; (c) Labeled hairpin DNA; hybridization of complementary target DNA causes the hairpin to unfold, separating the electroactive molecule from the surface and turning the signal off. DNA CT: electrodes are modified with single-stranded or duplex DNA, and redox-active reporter molecules that are electronically coupled to the DNA p-stack allow for the detection of intervening biomolecules which either complete (signal on) or perturb (signal off) this stack. Common strategies include the following: (d) Capture of complementary nucleic acids which complete the DNA duplex and produce an electrochemical signal; (e) abasic sites and other DNA lesions; or (f) Certain DNA binding proteins, both of which can disrupt the DNA p-stack and attenuate the electrochemical signal.
aptamers. In these latter constructs, target binding causes a conformational change that alters the electrochemical signal. An alternative signaling strategy relies on the unique property of double-stranded DNA to mediate long-range charge transport (DNA-CT) [12]. By modifying electrodes with DNA duplexes that feature electronically coupled, redox-active reporter molecules, a DNA-mediated electrochemical signal that depends on the integrity of the intervening DNA bases can be measured [13]. Because DNA CT is interrupted by DNA lesions, mismatches, and bound proteins that perturb the continuous p-stack, this is an extremely sensitive strategy for detecting biomarkers that bind to a specific DNA sequence [9]. Although DNA-based electrochemical approaches have shown promise in laboratory proof-of-principle studies, they have not translated into successful clinical applications. A major obstacle has been developing assays compatible with unpurified, complex biological samples: excessive sample processing can result in unintentional elimination of biomarker targets and precludes assessing www.sciencedirect.com
protein (or other biomolecule) activity in a native biological matrix. This task is particularly difficult using DNA-modified electrodes because of the propensity of biomolecules to adhere nonspecifically to electrode surfaces. Practically, overcoming this challenge involves assembling electrode surfaces with enough space between the probe DNA sequences to provide access for target binding in the constrained surface environment. Moreover, with the decreased number of DNA recognition elements and the use of dilute samples, there is an additional need to amplify the correspondingly lower electrochemical signals. Here, we focus on DNA CTe based detection schemes, describing recent work that we and others have performed to address these issues to move DNA-modified electrode detection closer to clinical utility.
Interhelical spacing in DNA films DNA-modified electrodes are typically formed via selfassembly of thiolated duplexes onto gold, followed by passivation of any uncovered surface with a small organic thiol such as mercaptohexanol (Figure 2a) [14]. As Current Opinion in Electrochemistry 2019, 14:104–112
106 Bioelectrochemistry
interhelical interactions of the DNA dominate the selfassembly process, even ‘low-density’ films tend to appear as islands of close-packed helices within a sea of passivating reagent. This inhomogeneity decreases both the sensitivity and reproducibility of detection. In recent years, three main strategies for controlling probe spacing and surface homogeneity have emerged that are as follows: manipulating the underlying electrode morphology [15e17]; developing next-generation passivation methods for DNA self-assembly [18]; and chemically controlling DNA attachment to a preformed monolayer [19,20,21]. Owing to their simplicity, consistency, and high level of control, we focus here on chemical (as opposed to physical) strategies to control DNA self-assembly. Improving self-assembled mixed monolayers
The Ferapontova group has significantly advanced the detection capabilities of self-assembled DNA monolayers by manipulating the nature of the adjuvant underlying passivating film [18]. Using a polyethylene glycol (PEGylated-) thiol (Figure 2b), they were able to suppress nonspecific adsorption of serum proteins onto dilute DNA monolayers, while still maintaining extremely efficient passivation against the DNAbinding electrochemical reporter molecules (methylene blue [MB] and ferricyanide), thus preventing false signals by direct electrochemical reduction. Using this PEGylated passivating agent and a DNA probe that featured an aptamer sequence specific for the cancer biomarker Her-2/neu, they successfully detected picomolar concentrations of the protein in serumd conditions amenable for clinical applications. Using click chemistry to assemble DNA
Enhancing target capture from complex mixtures also has been achieved by controlling the spacing and density of DNA duplexes on electrode surfaces via orthogonal coupling based on click chemistry (Figure 2c). Here, a mixed alkanethiol monolayer doped with a small fraction of molecules that possess a reactive head group is first self-assembled onto a gold electrode. Then, an appropriately labeled DNA duplex is coupled to the surface. Changing the concentration of the reactive head group in the underlying monolayer allows for homogeneous surfaces with controlled DNA probe duplex loadings. One strategy for this type of coupling involves surface attachment based on catalyst-free [3 þ 2] cycloaddition between an azide head group and a DNA duplex modified with a strained cyclooctyne [19]. The resulting density of DNA on the electrode surface is directly proportional to the mole fraction of azide in the underlying monolayer. Importantly, atomic force microscopy reveals homogenous films with minimal DNA aggregation, and these monolayers display key characteristics Current Opinion in Electrochemistry 2019, 14:104–112
of DNA-mediated electrochemistry, including signal decrease upon incorporation of a mismatch. Direct electrochemistry of reporter molecules intercalated into the DNA helices is extremely sensitive to proteinbinding events on this surface, with the ability to detect transcription factors well below the solution KD of the protein. Activatable catalysis for click chemistry
Cu(I)-catalyzed Huisgen 1,3-dipolar cycloaddition has been widely used to prepare functionalized monolayers [22]. In these systems, the Cu(I) is typically generated in situ using either a chemical reductant or by electrochemical reduction of a Cu(II) precursor. Unfortunately, electrochemical reduction of common water-soluble Cu(II) precursors often results in coppercontaining deposits on the electrode surface, impeding subsequent DNA CT to DNA-bound reporter molecules. One solution to this problem is to use a twoelectrode DNA-grafting platform in which active Cu(I) is generated at an auxiliary microelectrode [20,21]. In this system (Figure 2d), the microelectrode is positioned above a mixed monolayer doped with terminal azide groups: in the presence of alkyne-terminated DNA, electrochemical reduction of Cu(II) triggers DNA coupling to the underlying film in a spatially controlled manner. Any electrochemically induced copper deposits occur at the auxiliary electrode, shielding the DNA monolayer from damage. Direct electrochemistry of intercalated reporter molecules at a patterned array featuring different DNA sequences allowed simultaneous detection at physiological concentrations of multiple transcription factors from a complex mixture. Indeed, with appropriate signal amplification [23], these surfaces are sufficiently selective to allow detection of methyltransferase activity directly from human tumor tissue via DNA CT [24,25]. Reagentless oxidative coupling
By combining features of chemical-based cyclooctyne surface coupling with electrochemical catalyst activation, a reagentless method was recently developed to control DNA coupling to gold surfaces from a single pot [26]. This technique (Figure 2e) relies on a suite of oxidative coupling reactions developed in the Francis group for site-specific bioconjugations [27]. The key coupling reaction between substituted aniline and catechol is typically initiated by a mild chemical oxidant (either ferricyanide or periodate). However, in many cases, chemical oxidants are incompatible with sensitive biomolecules. Therefore, electrode-surface coupling of DNA was accomplished by direct electrochemical oxidation of catechol. Using this technology, aniline, alkyl thiols, and amine-modified DNA were all successfully coupled to catechol-modified electrodes. Interestingly, DNA-modified electrodes prepared in this www.sciencedirect.com
DNA-CT-based electrochemical sensors Furst et al.
107
Figure 2
Current Opinion in Electrochemistry
Strategies for controlling the spacing and homogeneity of DNA-modified electrodes. (a) In conventional monolayer formation, DNA sequences modified with a terminal alkanethiol self-assemble onto gold electrodes, followed by treatment with small thiol-containing molecules, such as mercaptohexanol, to cover and passivate any remaining electrode surface. (b) After the self-assembly of alkanethiol-modified DNA, a small molecule PEG-thiol can also be used as a more effective surface passivation agent for biological samples. (c) The spacing of individual DNA duplexes can be controlled by first forming a mixed-alkylthiol monolayer doped with chemically active head groups that can be later coupled to DNA. (d) When a terminal azide is used as the active www.sciencedirect.com
Current Opinion in Electrochemistry 2019, 14:104–112
108 Bioelectrochemistry
Figure 3
Current Opinion in Electrochemistry
Immobilization of electroactive cells onto an electrode allows for DNA CT–dependent current generation. Cells modified with single-stranded DNA are directed to adhere specifically to an electrode modified with the complementary sequence through DNA hybridization. When fully complementary DNA sequences are used, the cells generate more current (WM DNA) than either free cells or cells immobilized with mismatched (MM) DNA sequences, indicating a DNA CT–mediated pathway.
fashion were used recently to immobilize electron transfereproficient cells onto electrode surfaces (Figure 3). Thus, in addition to providing a route to whole-cell displays, these constructs enable the expansion of DNA CT to green energy technology [28].
Generation and amplification of electrochemical signals With straightforward (and scalable) strategies available for preparing DNA-modified electrodes to capture biomarkers from complex mixtures, there is still the challenge of generating sufficiently large signaling currents for clinical use. Recent work has focused on three main strategies: direct electrochemical readout; electrocatalysis; and collectoregenerator amplification using a bipotentiostat. Direct readout
In the simplest case, electrochemical readout can be accomplished by comparing the electrochemical signal
of a redox-active reporter molecule (e.g., MB, Nile blue, or daunomycin [9]) associated into DNA before versus after a signaling event (e.g., capture of a small-molecule target, binding of a protein, etc.). For helix-dependent DNA-CT assays, it is important that the reporter molecules intercalate into the p-stack [13]. This type of readout was recently used by Kahanda et al [29] in a functional assay for the activity of a chemotherapeutic agent in raw cell lysate. b-Lapachone is an orthoquinone anticancer drug that acts as a substrate for NAD(P)H: quinone oxidoreductase 1 (NQO1)dan NAD(P)H-driven enzyme overexpressed in a number of aggressive solid tumors. Upon reduction by NQO1, the resulting hydroquinone is rapidly reoxidized to b-lapachone, with concomitant generation of H2O2 and depletion of NAD(P)H. This redox cycling results in the buildup of reactive oxygen species, leading to the formation of 8-oxoguanine. Repair enzymes subsequently excise the damaged bases, leading to strand scissions and abasic sites within the DNA.
head group in this mixed monolayer, DNA can be patterned onto specific regions of an electrode by using a secondary electrode to activate an inert precatalyst and attach oligonucleotides onto particular locations via click chemistry. (e) A one-pot, electrochemical coupling method between a catecholterminated monolayer and aniline-modified DNA enables precise control over the final surface density of DNA. PEG, polyethylene glycol. Current Opinion in Electrochemistry 2019, 14:104–112
www.sciencedirect.com
DNA-CT-based electrochemical sensors Furst et al.
109
Figure 4
Current Opinion in Electrochemistry
Amplification strategies for DNA-CT detection. (a) Conventional electrocatalysis (left) versus collector-generator amplification (right). Electrocatalysis (left) involves amplification of the DNA-mediated redox signal of a DNA-intercalating probe (methylene blue, MB) with a small signaling molecule (ferricyanide) in solution. MB is reduced to leucomethylene blue (LB) by DNA CT, but is rapidly reoxidized back to MB by ferricyanide. If the individual electron-transfer rates are sufficiently large, the resulting current is limited only by ferricyanide diffusion, providing a substantially larger signal compared to direct electrochemistry. Incorporation of a secondary electrode (right) in collector-generator mode allows for the regeneration of ferricyanide at rates that exceed diffusion. Given the efficiency of long-range DNA-CT, this detection strategy generates significantly more current than conventional electrocatalysis www.sciencedirect.com
Current Opinion in Electrochemistry 2019, 14:104–112
110 Bioelectrochemistry
In an elegant assay, DNA duplexes featuring a covalent redox reporter molecule were self-assembled onto an electrode array to serve as a stand-in for cellular DNA. After treating these surfaces with cell lysate from tumor samples, the electrochemical signal arising from DNACT to the redox probe was then monitored upon addition of b-lapachone. The more damage to the individual helices within the film caused by futile b-lapachone redox cycling, the more attenuation in the electrochemical signal. Using a pulsed voltammetric readout technique (Osteryoung square-wave voltammetry) helped isolate the relatively small Faradaic currents from charging current background. This assay nicely exploits DNA-CT, as the end products of b-lapachone activity are charge transportedisrupting lesions of the p-stack. Notably, because formation of DNA lesions depends also on the activity of downstream repair proteins and other factors specific to the unique cell biology of the cancer cell, drug-induced DNA damage can deviate markedly from NQO1 expression levels. Thus, this DNA-CTe based functional assay provides critical clinical data that would be inaccessible using conventional biomarker analyses that simply quantify the expression level of a single component of a multifactorial pathway. Electrocatalysis
The finite surface concentrations of DNA-bound redox reporter molecules (determined by the electrode area and DNA surface density) ultimately limit the inherent sensitivity of direct electrochemical readout, highlighting the need for selective Faradaic signal amplification. This is commonly accomplished by an electrocatalytic cycle in which a DNA-bound reporter (e.g., MB) mediates the reduction of a soluble redox substrate (e.g., ferricyanide). As long as the substrate is blocked from interacting with the electrode surface, its reduction occurs only upon DNA-CT to the reporter. Because the ultimate electrochemical substrate is in the solution phase, the current is no longer limited by the amount of redox-probe molecule that can be adsorbed onto the electrode surface. In addition to electrocatalytic readout of low concentrations of circulating blood markers [18], non-DNAemediated electrocatalytic cycles (e.g., electrostatically bound Ru(NH3)3þ 6 /ferricyanide pairs) have been widely used for an impressive host of biomarker detection assays [10,11,30]. Collector/generator amplification
Although electrocatalysis has enabled detection of a broader range of analytes, often at lower concentrations,
the currents remain limited by the diffusion of the soluble redox signaling substrate. Moreover, under practical operating conditions, there is typically background current owing to some direct substrate reduction because of incomplete electrode passivation. Thus, while detection can be identified in a controlled laboratory setting by monitoring the ‘percent difference’ between electrocatalytic versus background signals with and without an analyte present, the absolute change in current may be quite small compared with the overall baseline currents. One approach to enhance the signal-to-noise ratio is to use a bipotentiostat to drive a two-electrode system in collector-generator mode, Figure 4 [31], Combined with DNA-CT, this alignment virtually eliminates backgrounds, as the signaling reporter is generated only via DNA-mediated processes; moreover, orienting the collector and generator electrodes at sufficiently small spacings allows recycling of the signaling reporter at rates that exceed the diffusion limit (cf. positive feedback in scanning electrochemical microscopy) [32]. We have applied this collector-generator concept to macroscopic platforms [20,21]. This amplification platform has proven useful in the development of an activity assay for the human methyltransferase DNMT1 from crude tumor cell lysate (Figure 4b). DNMT1 plays a critical role in the epigenetic regulation of gene expression, and aberrant DNA methylation by DNMT1 has been associated with tumorigenesis in a wide variety of cancers [33]. This electrochemical assay [34] is based on a two-step procedure in which electrode-bound DNA substrates are first treated with DNMT1, followed by incubation with a restriction enzyme that is incompetent toward fully methylated duplexes. DNMT1 activity, therefore, provides protection from excision by the restriction enzyme, and methyltransferase activity is reflected in the magnitude of the electrochemical signal that remains after restriction enzyme treatment. By carefully controlling interhelical DNA spacing during the selfassembly process and switching to a two-electrode detection platform, this assay was adapted to provide robust and high-throughput measurement of human DNMT1 activity in the crude lysate of both cultured cells [24] and patient tumor samples [25]. Most notably, the measured DNMT1 activity of tumor tissue did not correlate with DNMT1 expression or abundance, as measured by reverse-transcription polymerase chain reaction (RT-PCR) and Western blot, yet that activity was the most consistent indicator of tumor tissue
(center). (b) Electrochemical assay to detect methyltransferase activity on a two-electrode, electrocatalytic platform. Electrodes are prepared with DNA that contains overlapping recognition sites for the human methyltransferase DNMT1 and a restriction enzyme. Treatment of the electrodes with a crude cell lysate allows for methylation of the electrode-bound DNA by active DNMT1 in the lysate. Following this, the electrode is treated with a restriction enzyme and the electrocatalytic signal is measured again. As methylation protects the DNA from being cut from the surface by the restriction enzyme, the level of methyltransferase activity in the cell lysate sample is reflected by the magnitude of the electrochemical signal that remains after restriction enzyme treatment. The presence of active DNMT1 in the sample results in maintenance of the ‘on’ electrochemical signal (top) while the absence of active DNMT1 in the sample causes the electrochemical signal to be turned off (bottom). Current Opinion in Electrochemistry 2019, 14:104–112
www.sciencedirect.com
DNA-CT-based electrochemical sensors Furst et al.
[20]. This finding again highlights the importance of functional assays, conducted within a native matrix, to provide a handle for monitoring disease states that is more reliable than biomarkers which are present further upstream in a pathological process.
Conclusions It is becoming increasingly clear that routine diagnostic evaluation of unprocessed biological samples represents a great opportunity for improving the clinical management of complicated disease processes. Yet the complexity of these samples has hampered the development of high-throughput devices suitable for pointof-care analyses. In working toward next-generation DNA-based electrochemical biosensors, recent advances in controlling the surface morphology of modified electrodes, coupled with enhanced amplification methods, provide new design elements for creating diagnostic strategies tailored for specific biomarker panels and/or functional assays of multicomponent biological pathways. Indeed, DNA-CT activity assays on crude lysates have begun to yield useful information relevant to therapeutic-drug efficacy as well as disease etiology. Leveraging progress in platform design with transformative advances in chemical biology opens the door for an increasingly important role for DNA-based electrochemical devices in translating basic research findings into practical clinical tools.
Conflict of interest A.L.F. was supported by the A. O. Beckman Postdoctoral Fellowship. NBM has no conflict of interest. M.G.H. thanks the John Stauffer Charitable Trust and Fletcher Jones Foundation for support.
References Papers of particular interest, published within the period of review, have been highlighted as: of special interest of outstanding interest 1.
da Silva ETSG, Souto DEP, Barragan JTC, Giarola J de F, de Moraes ACM, Kubota LT: Electrochemical biosensors in pointof-care devices: recent advances and future trends. ChemElectroChem 2017, 4:778–794, https://doi.org/10.1002/ celc.201600758.
2.
Rifai N, Gillette MA, Carr SA: Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nat Biotechnol 2006, 24:971 – 983, https://doi.org/ 10.1038/nbt1235.
3.
Liu J, Gao J, Du Y, Li Z, Ren Y, Gu J, Wang X, Gong Y, Wang W, Kong X: Combination of plasma microRNAs with serum CA19-9 for early detection of pancreatic cancer. Int. J. Cancer 2012, 131:683–691, https://doi.org/10.1002/ijc.26422.
4.
Frisoni GB, Boccardi M, Barkhof F, Blennow K, Cappa S, Chiotis K, Démonet JF, Garibotto V, Giannakopoulos P, Gietl A, Hansson O, Herholz K, Jack CR, Nobili F, Nordberg A, Snyder HM, Ten Kate M, Varrone A, Albanese E, Becker S, Bossuyt P, Carrillo MC, Cerami C, Dubois B, Gallo V, Giacobini E, Gold G, Hurst S, Lönneborg A, Lovblad KO, Mattsson N, Molinuevo JL, Monsch AU, Mosimann U, Padovani A, Picco A, Porteri C, Ratib O, Saint-Aubert L, Scerri C, Scheltens P,
www.sciencedirect.com
111
Schott JM, Sonni I, Teipel S, Vineis P, Visser PJ, Yasui Y, Winblad B: Strategic roadmap for an early diagnosis of Alzheimer’s disease based on biomarkers. Lancet Neurol 2017, 16:661–676, https://doi.org/10.1016/S1474-4422(17) 30159-X. 5.
Pierrakos C, Vincent JL: Sepsis biomarkers: a review. Crit Care 2010, 14:R15, https://doi.org/10.1186/cc8872.
6.
Dolin HH, Papadimos TJ, Stepkowski S, Chen X, Pan ZK: A novel combination of biomarkers to herald the onset of sepsis prior to the manifestation of symptoms. Shock 2018, 49:364–370, https://doi.org/10.1097/SHK.0000000000001010.
7.
Senft D, Leiserson MDM, Ruppin E, Ronai ZA: Precision oncology: the road ahead. Trends Mol Med 2017, 23:874–898, https://doi.org/10.1016/j.molmed.2017.08.003.
[8].
Furst AL, Hoepker AC, Francis MB: Quantifying hormone disruptors with an engineered bacterial biosensor. ACS Cent Sci 2017, 3:110–116, https://doi.org/10.1021/ acscentsci.6b00322. This manuscript describes a novel electrochemical strategy for detecting a broad spectrum of endocrine-disrupting toxins based upon binding to a native protein receptor displayed on the surface of engineered E. coli. By reporting total estrogenic activity, this approach enables detection of endocrine-disrupting chemicals without requiring knowledge their structure. It represents a fundamental shift from standard assays, in which a specific capture element is constructed for a single biomarker. 9.
Drummond TG, Hill MG, Barton JK: Electrochemical DNA sensors. Nat Biotechnol 2003, 21:1192–1199, https://doi.org/ 10.1038/nbt873.
10. Boon EM, Ceres DM, Drummond TG, Hill MG, Barton JK: Mutation detection by electrocatalysis at DNA-modified electrodes. Nat Biotechnol 2000, 18:1096–1100, https://doi.org/ 10.1038/80301. [11]. Labib M, Sargent EH, Kelley SO: Electrochemical methods for the analysis of clinically relevant biomolecules. Chem Rev 2016, 116:9001–9090, https://doi.org/10.1021/ acs.chemrev.6b00220. This comprehensive review catalogs the diverse targets and electrochemical strategies under development for biomolecule detection. [12]. Beratan DN, Naaman R, Waldeck DH: Charge and spin transport through nucleic acids. Curr Opin Electrochem 2017, 4:175–181, https://doi.org/10.1016/J.COELEC.2017.08.017. This paper summarizes some of the recent developments in theoretical efforts to understand the phenomena of long-range charge transport through the double helix. [13]. Kékedy-Nagy L, Ferapontova EE: Sequence-specific electron transfer mediated by DNA duplexes attached to gold through the alkanethiol linker. J Phys Chem B 2018, 122: 10077–10085, https://doi.org/10.1021/acs.jpcb.8b07505. This study finds that the heterogeneous electron-transfer rate of MB reduction is significantly faster at DNA-modified surfaces featuring ATversus GC-duplexes. The rates correlate with the different binding modes of MB at these two surfaces (groove binding at AT versus intercalation at GC), emphasizing the importance of sequence context when designing DNA-CT assays that depend on intercalated probe molecules. 14. Levicky R, Herne TM, Tarlov MJ, Satija SK: Using self-assembly to control the structure of DNA monolayers on gold: a neutron reflectivity study. J Am Chem Soc 1998, 120: 9787–9792, https://doi.org/10.1021/ja981897r. 15. Elliott Joanne M, Birkin Peter R, Bartlett Philip N, Attard GS: Platinum microelectrodes with unique high surface areas. 1999, https://doi.org/10.1021/LA9908945. 16. Mahshid SS, Vallée-Bélisle A, Kelley SO: Biomolecular steric hindrance effects are enhanced on nanostructured microelectrodes. Anal Chem 2017, 89:9751–9757, https://doi.org/ 10.1021/acs.analchem.7b01595. 17. Arroyo-Currás N, Scida K, Ploense KL, Kippin TE, Plaxco KW: High surface area electrodes generated via electrochemical roughening improve the signaling of electrochemical aptamer-based biosensors. Anal Chem 2017, 89: 12185–12191, https://doi.org/10.1021/acs.analchem.7b02830.
Current Opinion in Electrochemistry 2019, 14:104–112
112 Bioelectrochemistry
[18]. Salimian R, Kékedy-Nagy L, Ferapontova EE: Specific pico molar detection of a breast cancer biomarker HER-2/neu protein in serum: electrocatalytically amplified electroanalysis by the aptamer/PEG-modified electrode. ChemElectroChem 2017, 4:872–879, https://doi.org/10.1002/ celc.201700025. Incorporating an aptamer sequence within an electrode-bound DNA films, a simple chemical blocking stratey for non-specific protein adsorption, and an electrocatalytic amplification strategy, this platform allows for screening of an important cancer biomarker from circulating blood, potentially obviating unnecessary and invasive biopsies. 19. Furst AL, Hill MG, Barton JK: DNA-modified electrodes fabricated using copper-free click chemistry for enhanced protein detection. Langmuir 2013, 29:16141–16149, https://doi.org/ 10.1021/la403262v. 20. Furst A, Landefeld S, Hill MG, Barton JK: Electrochemical patterning and detection of DNA arrays on a two-electrode platform. J Am Chem Soc 2013, 135:19099–19102, https:// doi.org/10.1021/ja410902j. 21. Furst AL, Hill MG, Barton JK: A multiplexed, two-electrode platform for biosensing based on DNA-mediated charge transport. Langmuir 2015, 31:6554–6562, https://doi.org/ 10.1021/acs.langmuir.5b00829. 22. Orain C, Le Poul N, Gomila A, Kerbaol J-M, Cosquer N, Reinaud O, Conan F, Le Mest Y: A generic platform for the addressable functionalisation of electrode surfaces through self-induced “electroclick. Chem - A Eur J 2012, 18:594–602, https://doi.org/10.1002/chem.201102620. 23. Furst AL, Hill MG, Barton JK: Electrocatalysis in DNA sensors. Polyhedron 2014, 84:150–159, https://doi.org/10.1016/ j.poly.2014.07.005. 24. Furst AL, Muren NB, Hill MG, Barton JK: Label-free electrochemical detection of human methyltransferase from tumors. Proc Natl Acad Sci Unit States Am 2014, 111:14985–14989, https://doi.org/10.1073/pnas.1417351111. [25]. Furst AL, Barton JK: DNA electrochemistry shows DNMT1 methyltransferase hyperactivity in colorectal tumors. Chem Biol 2015, 22:938–945, https://doi.org/10.1016/ j.chembiol.2015.05.019. In this study, the DNMT1 detection platform discribed in reference 24 was appied to human colorectal-tumor specimens. Significantly, while the electrochemical results in crude cell lysate correlate closely with the much more time-consuming (and expensive) standard tritium-labeling activity assay, they vary substantially from DNMT1 expression levels as measured by RT-PCR and western blotting. This finding underscores the importance of a functional approach for clinically relevant diagnostics involving complex biological pathways. 26. Furst AL, Smith MJ, Francis MB: Direct electrochemical bioconjugation on metal surfaces. J Am Chem Soc 2017, 139: 12610–12616, https://doi.org/10.1021/jacs.7b06385.
Current Opinion in Electrochemistry 2019, 14:104–112
27. Elsohly AM, Francis MB: Development of oxidative coupling strategies for site-selective protein modification. Acc Chem Res 2015, 48:1971–1978, https://doi.org/10.1021/ acs.accounts.5b00139. [28]. Furst AL, Smith MJ, Lee MC, Francis MB: DNA hybridization to interface current-producing cells with electrode surfaces. ACS Cent Sci 2018, 4:880–884, https://doi.org/10.1021/ acscentsci.8b00255. Shewanella oneidensis is often used as a component of bioelectrochemical cells, but gives inconsistent currents as electron injection normally depends upon biofilm formation. As highlighted in this study, wiring cells through DNA duplexes using controlled monolayer assembly highlights the ability of DNA CT to be used as both a conduit and a method to read out current production. [29]. Kahanda D, Singh N, Boothman DA, Slinker JD: Following anticancer drug activity in cell lysates with DNA devices. Biosens Bioelectron 2018, 119:1–9, https://doi.org/10.1016/ J.BIOS.2018.07.059. As described in the text, this work describes a functional electrochemical assay for a small-molecule chemotherapeutic. Operating via drug-induced enzymatic NAD(P)H reduction of dioxygen to reactive oxygen species (peroxide, superoxide, etc.), the drug is recycled many times thus providing inherent signal amplification. Because the downstream effect of this redox cycling is DNA damage that perturbs p stacking, DNA-CT is particularly well suited as a detection strategy. It is noteworthing that the degree of DNA damage does not necessarily correlate with target-enzyme expresion levels, stressing the importance of a simple activity assay performed under cell-specific conditions. 30. Fang Z, Kelley SO: Direct electrocatalytic mRNA detection using PNA-nanowire sensors. Anal Chem 2009, 81:612–617. 31. McClintock SA, Purdy WC: A bipotentiostat for four-electrode electrochemical detectors. Anal Lett 1981, 14:791–798, https:// doi.org/10.1080/00032718108055483. 32. Gorodetsky AA, Hammond WJ, Hill MG, Slowinski K, Barton JK: Scanning electrochemical microscopy of DNA monolayers modified with nile blue. Langmuir 2008, 24:14282–14288, https://doi.org/10.1021/la8029243. 33. Zhang W, Xu J: DNA methyltransferases and their roles in tumorigenesis. Biomark Res 2017, 5:1, https://doi.org/10.1186/ s40364-017-0081-z. [34]. Muren NB, Barton JK: Electrochemical assay for the signal on detection of human DNA methyltransferase activity. J Am Chem Soc 2013, 135:16632–16640, https://doi.org/10.1021/ ja4085918. This assay allowed for the first measurements of human DNMT1 using an electrochemical platform. Notably, when tested with crude lysates, cellular debris from the lysates adhered nonspecifically to the electrode, preventing access of methyltransferases to the electrode-bound DNA.
www.sciencedirect.com