High Surface Area Electrodes Generated via Electrochemical

Subscriber access provided by READING UNIV

Article

High surface area electrodes generated via electrochemical roughening improve the signaling of electrochemical aptamer-based biosensors Netzahualcoyotl Arroyo-Curras, Karen Scida, Kyle L Ploense, Tod E Kippin, and Kevin W. Plaxco Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b02830 • Publication Date (Web): 27 Oct 2017 Downloaded from http://pubs.acs.org on October 29, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

High surface area electrodes generated via electrochemical roughening improve the signaling of electrochemical aptamer-based biosensors Netzahualcóyotl Arroyo-Currása,b, Karen Scidac, Kyle L. Ploensed, Tod E. Kippind,e and Kevin W. Plaxco*,a,b,1

Affiliations a

Department of Chemistry and Biochemistry, University of California Santa Barbara, Santa Barbara, CA 93106, USA. b Center for Bioengineering, University of California Santa Barbara, Santa Barbara, CA 93106, USA. c Mechanical Engineering Department, University of California Santa Barbara, Santa Barbara, CA 93106, USA d

Department of Psychological and Brain Sciences, University of California Santa Barbara, Santa Barbara, CA 93106, USA. e

The Neuroscience Research Institute and Department of Molecular Cellular and Developmental Biology, University of California Santa Barbara, Santa Barbara, CA 93106, USA. *

Corresponding author

1

Correspondence to: Kevin W. Plaxco University of California Santa Barbara Chemistry Receiving Bldg. 557 Room 1432 Santa Barbara, CA 93106 Email: [email protected] Phone: (805) 893-5558 Fax: (805) 893-4120

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 15

Abstract The electrochemical, aptamer-based (E-AB) sensor platform provides a modular approach to the continuous, real-time measurement of specific molecular targets (irrespective of their chemical reactivity) in situ in the living body. To achieve this, however, requires the fabrication of sensors small enough to insert into a vein, which, for the rat animal model we employ, entails devices less than 200 µm in diameter. The limited surface area of these small devices leads, in turn, to low faradaic currents and poor signal-to-noise ratios when deployed in the complex, fluctuating environments found in vivo. In response we have developed an electrochemical roughening approach that enhances the signaling of small electrochemical sensors by increasing the microscopic surface area of gold electrodes, allowing in this case more redox-reporter-modified aptamers to be packed onto the surface, thus producing significantly improved signal-to-noise ratios. Unlike previous approaches to achieving microscopically rough gold surfaces, our method employs chronoamperometric pulsing in a 5-minute etching process easily compatible with batch manufacturing. Using these high surface area electrodes, we demonstrate the ability of E-AB sensors to measure complete drug pharmacokinetic profiles in live rats with precision of better than 10% in the determination of drug disposition parameters. Keywords aptamer, biosensor, roughening, surface area, in vivo


2

Page 3 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Electrochemical biosensors have been in development for over 50 years,1-3 with a major goal being the continuous and real-time measurements of clinically relevant molecular targets directly in the living body.4 These sensors, which employ biomolecules (proteins, nucleic acids, carbohydrates) as their recognition elements, produce an electrochemical output in response to target binding.5 In the best cases, the selectivity and rapid reversibility achieved with biomolecular recognition, coupled to the lack of any batch processing requirements for some electrochemical detection schemes, renders these biosensors particularly well suited for continuous measurements in complex environments.4 The continuous glucose monitor,6,7 for example, is a commercially successful electrochemical biosensor that achieves real-time measurements of glucose directly in vivo for periods of days.8 It achieves this by measuring glucose levels in interstitial fluid9,10 through the enzyme-mediated oxidation of glucose to gluconolactone, which produces electrons that are transferred to an osmium-based redox mediator and detected amperometrically on a carbon electrode.11 Although of unquestionable clinical value, the continuous glucose sensor is limited in that it relies on the specific chemical reactivity of glucose (its oxidation by glucose oxidase), and thus its sensing mechanism is not generalizable to the detection of other analytes. As an alternative, our group has been pursuing electrochemical aptamer-based (E-AB) sensors, a biosensor platform that, in contrast, supports the continuous, real-time measurement of specific small molecules irrespective of their chemical reactivity.12,13 E-AB sensors consist of an aptamer “probe” attached, at one end, to an interrogating electrode via a self-assembled monolayer, and modified at the other end with a redox-active methylene blue “reporter” (Figure 1A).14 The binding of an analyte to the aptamer alters the efficiency with which this reporter approaches the electrode, thereby producing an easily measured change in current when the sensor is interrogated via square wave voltammetry (Figure 1B).15 The binding-induced conformational change exploited in E-AB sensing mimics the signal transduction mechanisms employed by naturally occurring receptors in the body, thus rendering E-AB sensors particularly insensitive to non-specific binding. Enough so that we have previously used E-AB sensors to perform continuous, multi-hour measurements of multiple drugs directly in situ in the bodies of live rats.13 Achieving a temporal resolution of just a few seconds, these measurements support the generation of complete, high-precision pharmacokinetic profiles in real time13 (Figure 1D-E), a feat not achieved by any other modular (i.e., generalizable) biosensor approach reported to date. The versatility and selectivity of E-AB sensors notwithstanding, a limitation of the platform is that, because each of the methylene blue-modified aptamers on the electrode surface produces only two electrons, the sensor’s output current per unit area is relatively low. This, in turn, can cause problems for deployment in situ in the circulatory systems of rats, as the small size of these animals (< 400 g) limits the diameter of the electrodes we can employ in even their largest veins to less than 200 µm (Figure 1C). The limited surface area of such small electrodes renders their electrochemical signal low and noisy, especially when deployed in the complex, rapidly fluctuating environments found vivo. In prior work we superficially addressed this problem by increasing the surface area of gold-plated tungsten wires via electrochemical roughening.13 This approach was limited, however, in the number of roughening steps that could be applied before etching of the gold film would expose tungsten to the solution, which in turn prevents the deposition of a homogeneous self-assembled monolayer. Here, in contrast, we describe an


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 15

improved roughening method that employs all-gold electrodes and a larger number of longer (10x longer duration) roughening steps to produce wire-fine sensors with significantly enhanced surface areas and signal-to-noise (S/N) ratios, greatly improving the precision of the measurements accessible with these sub-millimeter in-vivo devices.

Figure 1. Electrochemical, aptamer-based (E-AB) sensors for the measurement of specific molecules in situ in the living body. (A) E-AB sensors consist of a methylene-blue-modified, target biding aptamer immobilized onto the surface of a gold electrode via a self-assembled monolayer. (B) In the presence of target (here the aminoglycoside antibiotic tobramycin), the aptamer undergoes a bindinginduced conformational change that, in turn, alters the electron transfer kinetics of the methylene blue redox reporter in a manner easily measurable using square wave voltammetry. (C, D) The E-AB sensors produced in this work are, at just 75 µm in diameter, small enough to insert into the veins of a live rat. (E) When so emplaced, these microns-diameter E-AB sensors support continuous, real-time, high-precision drug measurements. As shown here, for example, we can use such a sensor to follow the pharmacokinetics of tobramycin after a 40 mg/kg intramuscular administration in vivo in a live rat with unprecedented precision. The black line is the non-linear regression of these data assuming a onecompartment open pharmacokinetic model characterized by first-order absorption kinetics.

Experimental section Reagents and materials Sodium hydroxide, sulfuric acid, tris(hydroxymethyl)aminomethane (Tris), sodium hydrogen phosphate, sodium chloride, ethylenediaminetetraacetic acid (EDTA), potassium chloride, and potassium dihydrogen phosphate were obtained from Fisher Scientific (Waltham, MA). 6mercapto-1-hexanol and tris(2-carboxyethyl)phosphine (TCEP) were obtained from Sigma Aldrich (St. Louis, MO). Tobramycin sulfate (USP grade) was obtained from Gold


4

Page 5 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


BioTechnology, Inc (St. Louis, MO). All reagents were used as received. A 1X stock solution of phosphate buffered saline (PBS) was prepared containing: 10 mM sodium hydrogen phosphate, 2.7 mM potassium chloride, 137 mM sodium chloride, and 1.76 mM potassium phosphate. A 1X stock solution of Tris-EDTA buffer was prepared containing: 10 mM Tris-HCl (pH 8.0) and 1 mM EDTA. The E-AB sensors employed here were adapted from previous work.13,16 To fabricate them we obtained the relevant methylene-blue-and-thiol-modified DNA from Biosearch Technologies, with the sequence: 5’ – HO–(CH2)6–S–S–(CH2)6–PO3O–GGGACTTGGTTTAGGTAATGAGTCCC – CH2-CCH2OH-(CH2)4 –NH– CO–(CH2)2–Methylene Blue – 3’

The 5’ end is modified with a thiol on a 6-carbon linker and the 3’ end is modified with a carboxy-modified methylene blue attached to the DNA via the formation of an amide bond to a primary amine on a 7-carbon linker. The modified DNA was purified through dual HPLC by the supplier and used as received. Upon receipt, each construct was dissolved to 200 µM in 1X TrisEDTA buffer and frozen at -20°C in individual aliquots until use. Catheters (22 gauge) and 1 mL syringes were purchased from Becton Dickinson (Franklin Lakes, NJ). Polytetrafluoroethylene (PTFE)-insulated gold, platinum and silver wires (75 µm diameter) were purchased from A-M systems (Sequim, WA). The silver wires were immersed in concentrated sodium hypochlorite (commercial bleach) overnight to form a silver chloride film for use as a reference electrode. Heat-shrink PTFE (HS Sub-Lite-Wall, 0.02, 0.005, 0.003±0.001 in, black-opaque, Lot # 17747112-3), used to electrically insulate the gold, silver and platinum wires, was purchased from ZEUS (Branchburg Township, CA). Electrode fabrication Segments of gold, platinum and silver wire 10 cm in length were insulated, first individually by applying heat to shrinkable tubing around the body of the wires, then together, to make a staggered bundle exposing approximately 3 mm (in length) of bare gold and 6 mm of both platinum and silver. The far ends of these wires (opposite to their sensing “windows”) were left without insulation for a length of 1 cm and soldered to the connectors of a mesh-protected electrophysiology cable from Plastics One, Inc. (Roanoke, VA). When appropriate, the surface of the gold wire was roughened electrochemically in 0.5 M sulfuric acid by alternating the potential of the electrode between Einitial = 0.0 V to Ehigh = 2.0 V (all potentials versus Ag/AgCl), back and forth, for 16,000 pulses. Each pulse was 20 ms long with no “quiet time.” To functionalize the E-AB sensors, an aliquot of DNA aptamer was thawed and then reduced for 1 h at room temperature with a 1,000-fold molar excess of TCEP. A freshly roughened gold electrode was then immersed in DNA at 200 nM in PBS for 1 h at room temperature. Following this the sensors were immersed overnight in 20 mM 6-mercapto-1-hexanol in PBS to coat the remaining gold surface and remove nonspecifically adsorbed DNA. Electrochemical data acquisition and instrumentation All electrochemical measurements were recorded using a potentiostat from CH Instruments, Model 1040C (Austin, TX). The E-AB sensors were interrogated using square wave


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 15

voltammetry from 0.0 V to -0.5 V using an amplitude of 50 mV, potential step sizes of 1-5 mV, and frequencies varying from 10 Hz to 500 Hz. Cyclic voltammograms for the determination of surface areas were collected between -0.35 and 1.5 V, at a scan rate of 100 mV s-1. The files corresponding to all voltammograms were recorded in serial order using macros in CH Instruments software. The post-experiment analysis of results was carried out using a script coded in Igor Pro 7. Determination of electrode surface area We determined the microscopic surface area of our gold electrodes via cyclic voltammetry experiments as described previously (17). Briefly, we scanned the potential of the electrodes from -0.35 to 1.5 V in 0.05 M sulfuric acid to induce adsorption of oxygen onto the gold in a monoatomic layer with a one-to-one correspondence with surface metal atoms. Upon scanning the potential back to -0.35 V this oxygen monolayer is reduced (sharp peak at 0.9 V in Figure 2C). Integration of the area under this reduction current gives us the charge (in Coulombs) associated with the reduction of this monolayer. Finally, we divided this charge by 400 µC cm-2, the charge density corresponding to a complete monolayer of chemisorbed oxygen on gold (17), to obtain surface area in units of cm2. Determination of packing density in E-AB sensors To determine aptamer packing density we linearly scanned the potential of E-AB sensors from 0.0 to -0.4 V in PBS at a scan rate of 100 mV s-1 to completely reduce all the methylene blue reporters present on the sensor surface. Integrating the area under this reduction current and dividing it over the scan rate gives us the total charge (in Coulombs) of reporter-modified receptors. We converted this charge to moles and, by dividing moles over the corresponding electrode surface area, obtained packing density (in picomoles cm-2). Scanning electron microscopy (SEM) image acquisition The aspect and size of the gold electrode’s microstructure were studied with a Nova Nano 650 FEI field emission gun scanning electron microscope (Hillsboro, OR) equipped with an Oxford Inca X-ray EDX detector system (Abingdon, Oxfordshire, UK). Images were obtained using a beam voltage of 10 kV, spot size of 3, dwell time of 20 µs, acquisition of 16 bit and a resolution of 1534x1024. Both field-free and immersion modes, as well as 5,000X and 60,000X magnifications were used during detection. Animal surgery Male and female Sprague Dawley rats (4-5 months old) were purchased from Charles River Laboratories (Santa Cruz, CA), weighing between 300 g and 500 g. All animals were pairhoused in a standard light cycle room (08:00 on, 20:00 off) and allowed ad libitum access to food and water. For in-vivo measurements rats were anesthesia-induced with 5% isoflurane in a Plexiglas anesthesia chamber. The rats were then maintained on 2-3% isoflurane gas for the duration of the experiment. While anesthetized, E-AB sensors were inserted into the right internal jugular vein of the rats. Briefly, the area above the jugular vein was shaved and cleaned with betadine and 70% ethanol. A small incision was made above the jugular vein, and the vein


6

Page 7 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


was isolated. A small hole was cut into the vein with spring loaded micro-scissors, a silastic catheter was inserted, and 30 units of heparin were infused into the rat. The E-AB sensor was inserted alongside the silastic catheter, and the catheter was removed. The E-AB sensor was tied into place with silk suture. In vivo measurements All in vivo measurements were performed using a three-electrode setup in which the reference electrode was a silver wire coated with a silver chloride film as described above, and the counter electrode was a platinum wire. A 20-min sensor baseline was established before the first drug infusion. Recordings were taken for up to 3 h, with sampling rates of one point every 7 seconds. The real-time plotting and analysis of voltammetric data were carried out with the help of Matlab® scripts coded in-house. Regression analysis of pharmacokinetic profiles We performed non-linear regression analysis of our concentration data using a one-compartment open model with first-order drug absorption to fit intramuscular injections. The equation employed in the regressions was the following: =

·

·

−

(1)

where CP is the measured plasma concentration, CMAX is the maximum plasma concentration, γ is the first-order time constant of drug absorption and β the drug’s elimination time constant. During the regression analysis, all variables were floating such that the best fit was determined by minimizing the squared errors. Results Our goal is to shrink the macroscopic dimensions of E-AB sensors without concomitantly reducing their signaling current, which is proportional to their microscopic surface area. To do so we have used alternating electrochemical pulses (Figure S1) to electrochemically roughen our electrodes. Specifically, a 20-ms pulse at 2.0 V (all potentials reported versus Ag/AgCl) dissolves small areas of gold from the initially smooth, pristine surface of the electrode, and a second pulse at 0.0 V randomly re-deposits the gold back onto the electrode. This increases the surface area of the electrode by promoting the formation of nanostructured architectures, as seen via scanning electron microscopy (SEM). For example, after 16,000 potential pulses micrographs of the electrode region located at the air-liquid interface of the etching bath indicate the formation of nanostructured features on the bath-exposed surface of the wires (Figure 2A). The air-exposed surface, in contrast, remains relatively smooth. When viewed at a magnification of 60,000x (Figure 2B), we observe nanostructured needles projecting from the surface along with the formation of a porous film (Figure S2) with topographical features of sizes below 250 nm.


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 15

Figure 2. Electrochemical roughening to increase the microscopic, active surface area of gold wire electrodes. (A) SEM micrographs exhibit a clear boundary at the location of the air-liquid interface of the etching bath, illustrating the effect of our electrochemical roughening approach. (B) At a magnification of 60,000X we observe the formation of a porous structure with nanometer-sized features. (C) The change in microscopic surface area produced by this effect can be measured voltammetrically by comparing the magnitude of the peak corresponding to the reduction of oxygen chemisorbed on gold obtained on roughened (red) versus smooth (black) electrodes. (D) Integrating the area under the cathodic peak in the cyclic voltammograms, we built plots of surface area versus roughening pulses to reveal a linear increase in the microscopic surface area up to 16,000 pulses, after which it plateaus. Unless otherwise specified the error bars in this figure and the figures below correspond to standard deviations calculated from a total of six electrodes.

Electrochemical roughening significantly increases the electroactive area of our electrodes. To see this we measured the reduction current of oxygen adsorbed on gold using cyclic voltammograms recorded between -0.35 and 1.5 V in 0.05 M sulfuric acid (Figure 2C). From these measurements we determined the relation between number of potential pulses applied and microscopic surface area by integrating the charge under the cathodic peaks and dividing it over the 400 µC cm-2 charge density corresponding to the formation of a complete monolayer of chemisorbed oxygen on gold.17 In doing so, we find that the estimated surface area (Figure 2D) rises monotonically with the number of pulses before reaching a plateau at approximately 16,000 pulses. At this plateau, our roughening protocol produces stable films with an average 2-fold increase in surface area relative to pristine, smooth electrodes. This 2-fold increase in surface area allows us to reduce the diameter of our wire-fine electrodes while still maintaining an excellent S/N ratio. For example, in this work we reduced the diameter of our electrodes by 25% relative to our previously published work (75 µm versus 100 µm)13 but nevertheless achieved a 50% increase in S/N in our in vivo measurements (see below).


8

Page 9 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


More aptamer molecules can be packed onto a given macroscopic surface area of a roughened electrode, thus increasing E-AB signaling current. To show this we functionalized the surfaces of both roughened and smooth electrodes with aminoglycoside-binding aptamers diluted in a selfassembled monolayer of 6-mercaptohexanol. As expected, when interrogated using square wave voltammetry (in the absence of aminoglycoside target molecules) the signaling currents observed from roughened electrodes is significantly greater than that observed for smooth electrodes (Figure 3A), with the increase in signal being a monotonic function of the number of potential pulses applied during the roughening procedure (Figure 3B). For example, treating the electrodes with 16,000 pulses, we observed a 2-fold increase in E-AB signaling current that matches the 2-fold increase in surface area (as determined via reduction of gold oxides) produced by roughening.

Figure 3. Roughened E-AB sensors produce larger signaling currents. (A) We modified smooth (untreated) and roughened gold electrodes of the same macroscopic surface area with an aminoglycosidebinding aptamer and measured the signaling currents produced by the resultant sensors in vitro in flowing whole blood in the absence of target. (B) Depending on the number of roughening pulses applied prior to aptamer-modification, roughened E-AB sensors produce square-wave peak currents 2-4 times greater than those seen for smooth sensors of the same macroscopic dimensions.

In parallel to improving absolute E-AB signaling currents (measured in microamperes), roughening also improves E-AB signal gain (the relative change in current upon the addition of target). That is, while the signaling currents produced by aminoglycoside-detecting E-AB sensors increase monotonically with increasing tobramycin for both smooth and rough electrodes, the increase is more dramatic for roughened electrodes (Figure 4A). For example, the gains of roughened and smooth electrodes at 10 mM tobramycin were 240% and 180%, respectively. Similar gain enhancement has been reported for E-AB sensors fabricated on other microscopically rough electrode geometries.18 Here the improved gain appears to be associated with the higher aptamer packing densities that we achieve (under a given set of fabrication conditions such as, used in this study, 200 nM aptamer in the deposition solution) for roughened electrodes (Figure 4B). In support of this claim, equally high gain is seen for sensors employing smooth electrodes when they are fabricated at packing densities comparable to those seen for roughened electrodes (Figure 4C), conditions that can be achieved by employing higher aptamer concentrations during fabrication.


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 15

Figure 4. Roughening-induced changes in aptamer packing density enhance E-AB signal gain. (A) When challenged with target in flowing whole blood, the signal of E-AB sensors increases monotonically with target concentration, with larger signal gain (relative signal difference between no target and target) being seen for roughened (16,000 pulses) electrodes relative to smooth electrodes (the dissociation constants are 620 µM and 250 µM for rough and smooth electrodes, respectively). (B) This enhanced gain is caused by a 4-fold increase in the density with which the aptamer probes are packed onto the electrode surface when, as was the case here, the aptamer is deposited from a 200 nM stock. (C) The same result can be reproduced on smooth electrodes by performing the deposition of the aptamer at higher concentrations; for example, increasing the deposition concentration to 1 µM increases the gain from 190% to 300%. The error bars in panel C correspond to the standard deviation from 3 independently fabricated sensors.

Improvements in E-AB signal gain can arise due to optimization of aptamer packing density19 or due to changes in the equilibrium constant of the conformational switch that drives signaling.20 The improved gain we see here may reflect a convergence of these effects. First, the higher gain may be occurring, in part, because interactions between neighboring aptamers at high packing densities reduce the current observed from the unbound state, enhancing the signal change.19 Alternatively (or additionally), it may arise due to changes in the conformational switching equilibrium constant (as a switching constant that more greatly favors the non-binding, lowsignaling state leaves more aptamers “poised” to respond to target),20 increasing the bindinginduced signal change. The coupling of binding with a more unfavorable conformational change, however, should reduce the binding affinity, which is exactly what we see at the higher packing densities achieved on roughened electrodes (caption in Figure 4A). Electrochemically-roughened E-AB sensors support multi-hour in-vivo pharmacokinetic measurements. To demonstrate this we implanted either smooth (Figure 5A, C) or roughened (Figure 5B, D) E-AB sensors in the jugular veins of four lightly anesthetized rats. As our initial test we recorded the output of these sensors over the course of two hours in the absence of target, demonstrating excellent baseline stability during multi-hour in vivo operation (Figure 5A,B). We used these measurements to estimate S/N as defined by equation 2:


10

Page 11 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. Roughened E-AB sensors achieve higher S/N when measuring target concentrations in situ in living animals. Sensors built on roughened electrodes (A) achieve the same baseline stability as sensors built on untreated, smooth electrodes (B) during multi-hour operation in the jugular veins of live rats in the absence of their target molecule, albeit with the former achieving far superior S/N. This improved S/N greatly enhances the precision with which we can measure some pharmacokinetic parameters, such as CMAX, the maximum drug plasma concentration. Here we present full pharmacokinetic profiles for tobramycin following intramuscular injections of 40 mg/kg of the drug in two rats, recorded on roughened (C) and smooth (D) electrodes. The kinetic parameters γ and β correspond to absorption and elimination time constants, respectively.

% =

∙ 100

(2)

where SD is the standard deviation of the mean. From this we see that the S/N of the roughened electrodes (7%) is twice that of the smooth electrode (3%). This enhanced S/N in turn supports the in-vivo measurement of drug concentrations with unprecedented precision. To demonstrate this we measured complete pharmacokinetic profiles of two rats following the intramuscular injection of 40 mg/kg tobramycin (Figure 5C, D). We then performed non-linear regression analysis of the two profiles assuming an open, one compartment pharmacokinetic model with first-order drug absorption kinetics (equation 1).21 From this we see that high surface area, high aptamer density E-AB sensors improve the measurement of CMAX, the maximum drug plasma concentration, with significant increase in precision: the maximum plasma concentration as defined by an untreated, smooth sensor is 30 ± 7 µM, versus 32 ± 2 µM for an electrochemicallyroughened sensor (here the confidence intervals reflect estimated standard errors). In both cases, however, the precision with which these data define the drug’s pharmacokinetic rate constants is better than 10%; for example, the half-life of the elimination phase is β = 0.60 ± 0.03 h for the animal in which we employed a rough electrode sensor versus β = 1.51 ± 0.02 h for an animal in which we used a smooth electrode sensor (the differences in β reflecting the differing metabolism of the two animals). The failure of roughening to improve the (relative) precision with which we can estimate β presumably occurs because, with temporal resolution of just 6 s,


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 15

both measurements produce sufficient data that the precision with which they can define the drug’s kinetic half-lives is relatively insensitive to changes in signal-to-noise. As we have shown previously,13 β is, in reality, not a constant as it fluctuates with the animal’s health status (e.g., dehydration when under anesthesia); it thus appears likely that physiology itself precludes the determination of β with greater precision than that achieved with our smooth electrode sensors. Conclusion Previous reports have discussed the importance of nanostructuring22-24 noble-metal-supported electrochemical biosensors to improve signaling. This past work, however, has generally employed gold electrodeposition from acidic solutions of chloroauric acid to produce microstructured films. While electrodeposition supports the formation of geometry-controlled gold nanostructures, it requires carefully controlled bath conditions (salt concentrations, temperature, ionic strength, etc.) and extensive electrode cleaning (polishing) prior to deposition. The approach we describe here, in contrast, relies on the direct roughening of gold wires via chronoamperometric pulsing in acidic media. This approach supports the formation of high surface area, microstructured architectures in a trivially easy, 5-minute-long process that does not require electrode polishing prior to roughening. These attributes render our approach particularly compatible with batch manufacturing. And thanks to their enhanced microscopic surface area and the commensurate improvement in their signaling, roughened E-AB sensors can be fabricated with diameters below 100 µm that support continuous, ultrahigh precision, multi-hour measurements of specific small molecules directly in situ in the living body. The improved signaling achieved on electrochemically roughened electrodes may also prove of benefit to other electrochemical sensing architectures, including detection schemes that measure changes in electron transfer from solution-phase redox reporters,25,26 changes in the reporter’s reorganizational energy,27,28 and/or changes in electron transfer due to increased steric hindrance. 23,29,30

Acknowledgement These studies were supported by a grant from the W. M. Keck Foundation, and by the Institute for Collaborative Biotechnologies through grant W911NF-09-0001 from the U.S. Army Research Office. The content of the information does not necessarily reflect the position or the policy of the Government, and no official endorsement should be inferred. N.A.C. is supported by the Otis Williams Postdoctoral Fellowship of the Santa Barbara Foundation. The authors thank Dr. Gabriel Ortega for providing the code used in real-time data plotting. The authors acknowledge the MRL Shared Experimental Facilities supported by the MRSEC Program of the National Science Foundation under award NSF DMR 1121053, a member of the NSF-funded Materials Research Facilities Network. Supporting Information Additional figures showing: Schematic representation of stepped-potential program employed to roughen electrodes SEM micrographs showing cross-sectional cuts of roughened gold electrodes


12

Page 13 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


References (1)

Clark, L. C.; Lyons, C., Ann. N. Y. Acad. Sci. 1962, 102, 29-45.

(2)

Mehrotra, P., J. Oral. Biol. Craniofac. Res. 2016, 6, 153-159.

(3)

Grieshaber, D.; MacKenzie, R.; Vörös, J.; Reimhult, E., Sensors (Basel, Switzerland) 2008, 8, 1400-1458.

(4)

Wilson, G. S.; Gifford, R., Biosensors Bioelectron. 2005, 20, 2388-2403.

(5)

Thévenot, D. R.; Toth, K.; Durst, R. A.; Wilson, G. S., Biosensors Bioelectron. 2001, 16, 121-131.

(6)

Bode, B.; Gross, K.; Rikalo, N.; Schwartz, S.; Wahl, T.; Page, C.; Gross, T.; Mastrototaro, J., Diabetes Technol. Ther. 2004, 6, 105-113.

(7)

Mastrototaro, J. J., Diabetes Technol. Ther. 2000, 2, 13-18.

(8)

Mastrototaro, J.; Shin, J.; Marcus, A.; Sulur, G., Diabetes Technol. Ther. 2008, 10, 385390.

(9)

Baek, Y. H.; Jin, H. Y.; Lee, K. A.; Kang, S. M.; Kim, W. J.; Kim, M. G.; Park, J. H.; Chae, S. W.; Baek, H. S.; Park, T. S., Korean Diabetes J. 2010, 34, 350-358.

(10) Cengiz, E.; Tamborlane, W. V., Diabetes Technol. Ther. 2009, 11, S-11-S-16. (11) Feldman, B.; Brazg, R.; Schwartz, S.; Weinstein, R., Diabetes Technol. Ther. 2003, 5, 769-779. (12) Li, H.; Dauphin-Ducharme, P.; Arroyo-Currás, N.; Tran, C. H.; Vieira, P. A.; Li, S.; Shin, C.; Somerson, J.; Kippin, T. E.; Plaxco, K. W., Angew. Chem. Int. Ed. 2017, 56, 74927495. (13) Arroyo-Currás, N.; Somerson, J.; Vieira, P. A.; Ploense, K. L.; Kippin, T. E.; Plaxco, K. W., Proc. Natl. Acad. Sci. U.S.A. 2017, 114, 645-650. (14) Xiao, Y.; Lubin, A. A.; Heeger, A. J.; Plaxco, K. W., Angew. Chem. Int. Ed. 2005, 44, 5456-5459. (15) White, R. J.; Plaxco, K. W., Anal. Chem. 2010, 82, 73-76. (16) Rowe, A. A.; Miller, E. A.; Plaxco, K. W. Anal. Chem. 2010; 82, 7090–7095. (17) Trasatti, S.; Petrii, O. A. Pure Appl. Chem. 1991, 63, 711-734. (18) Hauke, A.; Kumar, L. S. S.; Kim, M. Y.; Pegan, J.; Khine, M.; Li, H.; Plaxco, K. W.; Heikenfeld, J., Biosensors Bioelectron. 2017, 94, 438-442. (19) White, R. J.; Phares, N.; Lubin, A. A.; Xiao, Y.; Plaxco, K. W., Langmuir 2008, 24, 10513-10518. (20) Vallée-Bélisle. A; Ricci, F; Plaxco, K. W. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 1380213807. (21) Loftsson, T., Chapter 2 - Basic Concepts of Pharmacokinetics. In Essential Pharmacokinetics, Academic Press: Boston, 2015; pp. 9–84.


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 15

(22) Bin, X.; Sargent, E. H.; Kelley, S. O.; Anal Chem. 2010, 82, 5928-5931. (23) Mahshid, S.; Mepham, A. H.; Mahshid, S. S.; Burgess, I. B.; Saberi Safaei, T.; Sargent, E. H.; Kelley, S. O., J. Phys. Chem. C. 2016, 120, 21123-21132. (24) De Luna, P.; Mahshid, S. S.; Das, J.; Luan, B.; Sargent, E. H.; Kelley, S. O.; Zhou, R., Nano Lett. 2017, 17, 1289-1295. (25) Katayama, Y.; Ohuchi, Y.; Higashi, H.; Kudo, Y.; Maeda, Anal. Chem. 2000, 72, 46714674. (26) Le Floch, F.; Ho, H. A.; Leclerc, M., Anal. Chem. 2006, 78, 4727-4731. (27) Plumb, K.; Kraatz, H.-B., Bioconjugate Chem. 2003, 14, 601-606. (28) Feld, D. J.; Hsu, H.-T.; Eckermann, A. L.; Meade, T. J., Langmuir 2012, 28, 939-949. (29) Ge, B.; Huang, Y. C.; Sen, D.; Yu, H.-Z., Angew. Chem. Int. Ed. 2010, 49, 9965-9967. (30) Cash, K. J.; Ricci, F.; Plaxco, K. W., J. Am. Chem. Soc. 2009, 131, 6955-6957.


14

Page 15 of 15

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


for TOC only


15

High Surface Area Electrodes Generated via Electrochemical

Recommend Documents