ACS Publications. Most Trusted. Most Cited. Most Read
Optimization of Vancomycin Aptamer Sequence Length Increases the Sensitivity of Electrochemical, Aptamer-Based Sensors In Vivo
My Activity
  • Open Access
Article

Optimization of Vancomycin Aptamer Sequence Length Increases the Sensitivity of Electrochemical, Aptamer-Based Sensors In Vivo
Click to copy article linkArticle link copied!

  • Alexander Shaver
    Alexander Shaver
    Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United States
  • J.D. Mahlum
    J.D. Mahlum
    Chemistry-Biology Interface Program, Zanvyl Krieger School of Arts & Sciences, Johns Hopkins University, Baltimore, Maryland 21218, United States
    More by J.D. Mahlum
  • Karen Scida
    Karen Scida
    Lieber Institute for Brain Development, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United States
    More by Karen Scida
  • Melanie L. Johnston
    Melanie L. Johnston
    Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United States
    Biochemistry, Cellular and Molecular Biology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21287, United States
  • Miguel Aller Pellitero
    Miguel Aller Pellitero
    Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United States
  • Yao Wu
    Yao Wu
    Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United States
    More by Yao Wu
  • Gregory V. Carr
    Gregory V. Carr
    Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United States
    Lieber Institute for Brain Development, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United States
  • Netzahualcóyotl Arroyo-Currás*
    Netzahualcóyotl Arroyo-Currás
    Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United States
    Chemistry-Biology Interface Program, Zanvyl Krieger School of Arts & Sciences, Johns Hopkins University, Baltimore, Maryland 21218, United States
    Biochemistry, Cellular and Molecular Biology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21287, United States
    *Email: [email protected]. Phone: 443-287-4798.
Open PDFSupporting Information (1)

ACS Sensors

Cite this: ACS Sens. 2022, 7, 12, 3895–3905
Click to copy citationCitation copied!
https://doi.org/10.1021/acssensors.2c01910
Published November 23, 2022

Copyright © 2022 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY-NC-ND 4.0 .

Abstract

Click to copy section linkSection link copied!

The measurement of serum vancomycin levels at the clinic is critical to optimizing dosing given the narrow therapeutic window of this antibiotic. Current approaches to quantitate serum vancomycin levels are based on immunoassays, which are multistep methods requiring extensive processing of patient samples. As an alternative, vancomycin-binding electrochemical, aptamer-based sensors (E-ABs) were developed to simplify the workflow of vancomycin monitoring. E-ABs enable the instantaneous measurement of serum vancomycin concentrations without the need for sample dilution or other processing steps. However, the originally reported vancomycin-binding E-ABs had a dissociation constant of 45 μM, which is approximately 1 order of magnitude higher than the recommended trough concentrations of vancomycin measured in patients. This limited sensitivity hinders the ability of E-ABs to accurately support vancomycin monitoring. To overcome this problem, here we sought to optimize the length of the vancomycin-binding aptamer sequence to enable a broader dynamic range in the E-AB platform. Our results demonstrate, via isothermal calorimetry and E-AB calibrations in undiluted serum, that superior affinity and near-equal sensor gain in vitro can be achieved using a one-base-pair-longer aptamer than the truncated sequence originally reported. We validate the impact of the improved binding affinity in vivo by monitoring vancomycin levels in the brain cortex of live mice following intravenous administration. While the original sequence fails to resolve vancomycin concentrations from baseline noise (SNR = 1.03), our newly reported sequence provides an SNR of 1.62 at the same dose.

This publication is licensed under

CC-BY-NC-ND 4.0 .
  • cc licence
  • by licence
  • nc licence
  • nd licence
Copyright © 2022 The Authors. Published by American Chemical Society
Vancomycin is a critically important antibiotic that is used for a variety of treatments including methicillin-resistant Staphylococcus aureus (MRSA). (1,2) It is one of the few antibacterials that multidrug-resistant bacteria are susceptible to and, as such, has been used as a first response to bacterial infection, even when patients do not require antibiotics. (1−3) Great care must be taken to ensure that the concentration of vancomycin remains high enough above the minimum inhibitory concentration of MRSA to keep the bacteria from gaining resistance. (2,4) To monitor serum vancomycin concentrations in clinical laboratories, a variety of methods exist, including enzyme immunoassays, enzyme-multiplied immunoassays, and fluorescence polarization immunoassays. (5) These commercial methods often present measurement discrepancies that can drastically affect therapeutic toxicity and, consequently, a patient’s health. Therefore, there is a continued need for the advancement and optimization of techniques to measure vancomycin concentrations in serum samples or directly in patients.
To ensure proper dosing of vancomycin, the standard of care is to use therapeutic drug monitoring (TDM). However, TDM presents the problem of sample management, needing phlebotomy, transport of serum to central facilities, multiple sample processing steps, and backlogged labs. (2) A potential solution to simplify TDM is to measure serum vancomycin concentrations at the patient’s location, or alternatively, to continuously monitor concentrations within a patient’s body. Recent work has shown that electrochemical, aptamer-based sensors (E-ABs) could be ideally suited for both of these applications. (6) E-ABs can reagentlessly measure the concentration of a target molecule in biological samples both in vivo and in vitro. (6−10) Generally, they consist of single-stranded DNA (ssDNA) aptamers attached to a sensing electrode. A redox reporter (e.g., methylene blue) attached to the distal end of the ssDNA can exchange electrons with the electrode at different rates, depending on target binding-induced changes in the conformation of the aptamer (Figure 1A), which leads to changes in electrochemical currents (Figure 1B). These currents can be translated to dose–response curves and used to determine target concentration in a sample (Figure 1C). E-ABs are ideally suited for the application of point-of-need vancomycin sensing because they do not require multiple dilutions, reagent additions, or any chemical processing of patient samples. (6) Thus, E-ABs uniquely achieve accurate vancomycin measurements in unprocessed biological fluids.

Figure 1

Figure 1. Vancomycin binding to E-ABs results in an electrochemical signal change. (A) E-ABs consist of several parts: (i) the electrode, (ii) the self-assembled monolayer, (iii) the SAM-bound aptamer (which binds to the target, vancomycin), and (iv) the redox reporter. (B) The redox reporter exchanges electrons at rates dependent on the reporter–electrode distance, keT1 and keT2. Because target binding induces conformational changes in the aptamer that affect the reporter–electrode distance, such changes result in observable changes in voltammetric current. These voltammograms were measured via square wave voltammetry at a frequency of 250 Hz, amplitude of 25 mV, and step size of 1 mV in phosphate-buffered saline (PBS) with 2 mM MgCl2. (C) Change in voltammetric peak current (red markers) or area under the curve (blue markers) can then be plotted to build dose–response curves from which thermodynamic parameters can be extracted. Here, for example, we performed nonlinear regression of the curve to a Hill model (red trace), obtaining a dissociation constant of 11.7 ± 0.7 μM and a Hill coefficient of 0.66 ± 0.02. The clinically relevant range for serum vancomycin is 6–30 μM as indicated in Table 1. (2,6)

The first vancomycin-sensing E-ABs were developed by Dauphin-Ducharme and colleagues, following a truncation-based approach that produced a conformation-switching DNA aptamer. (6) Although demonstrated to be functional both in vitro in whole blood and in vivo in the veins of rats, the dissociation constant of these E-ABs for vancomycin is, at 45 μM, 1 order of magnitude higher than the lowest recommended trough concentrations measured in patients (see Table 1 for reference). Unfortunately, given the natural and known subject-to-subject variability in vancomycin pharmacokinetics, (11) this sensitivity is not sufficient to accurately measure the full range of concentrations expected during therapeutic monitoring (see Table 1). The authors originally reported the use of a modified sequence truncated by four base pairs (Table S1 and Figure 2). Here, seeking to further increase the dynamic range of these E-ABs, we investigated the effect of a series of aptamer truncations starting from the aptamer termini, up to a maximum of four base pairs (all sequences included in Table S1). Deploying such aptamer sequences, we performed isothermal titration calorimetry (ITC), circular dichroism (CD) spectroscopy, and square wave voltammetry to determine the effect of the various truncations on vancomycin affinity, aptamer conformation dynamics, and E-AB signaling output. Through these measurements, we demonstrate that higher sensitivity and a broader dynamic range are achieved in serum with a sequence one-base-pair longer than the truncated aptamer originally reported by Dauphin-Ducharme et al., (6) without compromising maximum signal gain. In addition, the increased affinity of the longer sequence allows the real-time monitoring of vancomycin levels in the brain of live rodents following intravenous dosing, which the originally reported sequence fails to resolve from baseline noise. Thus, the corrected aptamer sequence length allows vancomycin E-AB sensing that fully covers the range required for accurate therapeutic drug monitoring in ex vivo specimens as well as pharmacokinetic measurements in vivo.

Figure 2

Figure 2. ITC measurements reveal worsening of vancomycin affinity with increasing aptamer truncations. (A) The modeled secondary structure of the aptamers with symmetric truncations considered in this work. From left to right: no truncations, one (1trunc), two (2trunc), three (3trunc), and four (4trunc) base-pair truncations. This modeling was done using Nupack (14) assuming 140 mM Na+ and 2 mM Mg2+ at a temperature of 25 °C. (B) An example of two raw ITC graphs. Negative is exothermic. (C) We calculated the areas for each titration point and plotted them against the vancomycin to aptamer mole ratio in the ITC cell. We then fit the curves to a binding polynomial (red, black, and blue traces). KD for the full aptamer is 0.14 ± 0.02 μM, KD for 3trunc is 3.92 ± 0.90 μM, and KD for 4trunc is 52 ± 20 μM. Error bars represent the standard deviation of the heat and mole ratio. Errors are calculated from three replicates. Shading represents the propagation of error of the 95% confidence interval for each regression based on three replicates. (D) Red markers represent the KD, and blue markers represent the n of each symmetrical truncation. Dashed lines are point-to-point connections. Error bars are based on three replicates. All measurements were done in PBS solution with 2 mM MgCl2 added.

Table 1. Clinical Range and Dissociation Constants for the Previously Optimized 4trunc Aptamer and the 3trunc Sequence Introduced in This Work
clinically relevant range (μM)a4trunc aptamer’s KD in flowing blood (μM)b4trunc aptamer’s KD in serum (μM)c3trunc aptamer’s KD in serum (μM)c
6–3045110 ± 1111.7 ± 0.7
a

Refs (2,12).

b

Ref (6).

c

This work.

Results and Discussion

Click to copy section linkSection link copied!

Affinity of Truncated Vancomycin Aptamers

We first investigated the effect that six different truncations (Table S1) of the full aptamer originally reported by Dauphin-Ducharme and colleagues had on vancomycin binding affinity using ITC (Figures 2 and S1). We performed these measurements in PBS supplemented with 2 mM MgCl2. The full aptamer sequence is predicted to fold into two stem regions acting as spacers between two loop regions (Figure 2A). The goal of truncating the sequence was to destabilize the shorter-stem region of the aptamer to favor an unfolded state. This strategy promotes vancomycin-induced folding of the aptamer upon binding, generating a large electrochemical signal in E-ABs. However, because the binding pocket of the aptamer is currently unknown, identifying the ideal number of truncations to achieve maximum conformation-switching behavior and, thereby, E-AB signaling involves a process of trial and error. Here, under the above-described conditions, the full sequence presented a perfect sigmoidal curve by ITC (Figure 2C) with an estimated dissociation constant (KD) of 0.14 ± 0.03 μM and stoichiometric coefficient (n) of 0.721 ± 0.007 (Figure 2D). Given that the aptamer is dual HPLC-purified and its concentration and that of vancomycin were calibrated via spectrophotometric measurements, the n < 1 may indicate that the simple independent site binding model is inappropriate to fit our data. In other words, vancomycin binding to the aptamer may not necessarily bind in a 1:1 ratio. In addition, while we observed a strong binding response for the full aptamer, it lacked the dynamic range necessary for vancomycin sensing. Thus, we aimed to truncate the vancomycin aptamer to balance a large signal change with a greater dynamic range.
Systematically truncating the parent aptamer sequence one base-pair at a time affected the apparent affinity and net ITC heat rates. For example, single base-pair truncations incrementally decreased ITC heat rate magnitudes (Figure 2B) and resulted in a nearly imperceptible binding isotherm when we performed a four base-pair truncation (Figure 2C, black trace), here termed 4trunc. In addition, performing progressive sequence truncations also affected the vancomycin binding affinity and corresponding stoichiometric coefficients, resulting in worse affinity with increasing number of truncations (Figure 2D). Overall, these results indicate a greater stabilization of an unfolded state for the 4trunc sequence, which is predicted to lack the first stem of the full aptamer (Figure 2A). Additionally, the change in stoichiometric coefficients from the full aptamer (n = 0.721 ± 0.007) to the three base-pair truncation (n = 0.57 ± 0.05), 3trunc, indicates the possibility of a 1:2 interaction (2 aptamers per vancomycin molecule) under the conditions tested (Figure 2D). We observe this change in conjunction with a drop in binding affinity relative to the full aptamer sequence (KD of 3.92 ± 0.90 μM). We note that the ratio of the measured aptamer concentration over the dissociation constant (or c-value) (13) for each truncation (c = 125, 3.4, and 0.4 for the full, 3trunc, and 4trunc variants, respectively) reflects the decrease in aptamer affinity with increasing number of truncations. Re-titrating the 4trunc variant with c = 2 by increasing the aptamer concentration resulted in a broad titration curve that still did not reflect higher affinity (Figure S2).
A single base-pair truncation (1trunc) produced an ITC isotherm identical to the full sequence (Figure S1A), and a two base-pair truncation (2trunc) produced intermediate behavior between the full and 3trunc sequences (Figure S1B). Asymmetric truncations resulted in ITC isotherms like those obtained from base-pair truncations (Figure S1C). Based on these results, we continued our structure-switching and electrochemical analysis using the full, 3trunc, and 4trunc sequences.
To evaluate the effect on the binding affinity of the aptamer modifications needed for E-AB fabrication, we also performed ITC affinity measurements after functionalizing the 3trunc sequence with hexylthiol and methylene blue as redox reporter (Figure S1D). All surface constructs employed in this work, including 3trunc, included a covalent bond to methylene blue at the 3′ terminus, and one to hexylthiol at the 5′ terminus (see the Methods section for structure). This configuration favors the strongest signaling from E-AB sensors as reported by Chamorro-García and colleagues. (15) Regression analysis of the resulting isotherms revealed an expected destabilization of the folded state of the 3trunc aptamer via a 3-fold worsening of the dissociation constant (KD = 9.13 ± 1.9 vs 3.92 ± 0.90 μM without functionalization, Figure S1D vs Figure 2C).

Conclusions from the Affinity Measurements

Symmetrically truncating the stem of the parent vancomycin-binding aptamer sequence by only three base pairs (3trunc) instead of four (4trunc, previously published) results in a 13× stronger binding affinity. However, the 3trunc aptamer-vancomycin interaction may not follow a 1:1 stoichiometry, a factor that must be considered when designing E-AB sensors using this sequence.

Circular Dichroism Spectroscopy of Truncated Vancomycin Aptamers

To confirm that base-pair truncations effectively induce structure-switching behavior in the vancomycin aptamer, we examined the full, 3trunc, and 4trunc sequences in the presence and absence of vancomycin via CD spectroscopy (Figure 3). We note that vancomycin exhibits strong CD absorption (Figure S3); thus, the CD spectra reported in this work were vancomycin-spectrum subtracted. For CD measurements, we employed vancomycin solutions at 100 μM because such a concentration elicited a large binding response in ITC while not unduly interfering with our CD measurements at any wavelength (200–320 nm). Collecting CD spectra in the presence of 2 mM MgCl2, each of the constructs showed a band transition upon the addition of vancomycin (Figure 3). In more detail, the full and 3trunc sequences demonstrated a shift of about Δλ ∼ 10 nm, from ∼260 to 270 nm (Figure 3A,B). The 4trunc sequence, on the other hand, portrayed a Δλ ∼ 5 nm (Figure 3C). This smaller change for the 4trunc sequence may be explained by a less stable secondary structure. The shortened stem destabilizes the folded conformation of 4trunc, causing the same concentration of vancomycin to have a smaller effect on 4trunc compared to longer-stemmed constructs.

Figure 3

Figure 3. CD spectroscopy illustrates the change in aptamer conformation upon vancomycin binding. CD spectra for each construct with no vancomycin (red trace) and 100 μM vancomycin (black trace). Red and gray shaded areas represent the standard deviation of three measurements without and with vancomycin, respectively. Measurements were done at a concentration of 5 μM aptamer in PBS solution with 2 mM Mg2+. Insets are expanded regions to highlight band shift upon vancomycin addition. For the 4trunc sequence (C), minimal vancomycin binding-induced conformational change is seen compared to full (A) or 3trunc (B). Measurements were performed at 25 °C, with an averaging time of 2 seconds, and a 1 mm path length. Data were background-subtracted ([aptamer ± vancomycin] – [buffer ± vancomycin]).

Conclusions from CD Measurements

For our tested concentration of vancomycin, addition of target causes a larger band shift for the full and 3trunc constructs compared to 4trunc. This result likely indicates that full and 3trunc constructs are fully bound to vancomycin at 100 μM, while 4trunc─with its destabilized stem─is only partially bound and does not exhibit a clear two-state model. For developing E-ABs, aptamers that undergo distinct conformational changes are ideal, suggesting 3trunc as a better candidate than 4trunc.

Performance of Truncated Aptamers in E-AB Format

The enhanced performance of the 3trunc aptamer becomes even more apparent when using electrochemical interrogation of hexylthiol- and methylene blue-modified sequences on gold electrodes (Figure 4). To illustrate this behavior, we co-deposited the modified sequences with mercaptohexanol to form self-assembled monolayers following previously published protocols. (6,16) We then challenged the resulting E-AB sensors in PBS with increasing concentrations of vancomycin to build calibration curves. Notably, we observed no signal gain (i.e., maximum signal output at saturating vancomycin concentrations) when the E-AB sensors were fabricated with either the full aptamer or 1trunc sequences, indicating that these constructs do not undergo the switching dynamics needed for E-AB sensing. While the signal gain for the 3trunc and 4trunc sensors is within error of each other (Figure 4A) the dynamic range is larger and the apparent dissociation constant, KD, is lower for the 3trunc-functionalized sensors. This latter result agrees with our ITC measurements in the same matrix (Figure 2D). Since aptamer deposition concentration is known to play a role in signaling, (17,18) we also determined that the optimal concentration for deposition under these conditions is 200 nM (Figure S4).

Figure 4

Figure 4. Performance of truncated aptamers in E-AB format. Here, we show dose–response curves for E-ABs functionalized with a 200 nM deposition concentration of (A) 3trunc (KD = 19.5 ± 1.6 μM) and 4trunc (KD = 70.7 ± 6 μM) aptamers in PBS with 2 mM Mg2+ or (B) 3trunc (KD = 18.1 ± 1.6 μM) and 4trunc (KD = 87.3 ± 9.1 μM) E-ABs in whole serum. The y-axis in these titrations represents signal gain, computed as the change in signal relative to measurements in the absence of vancomycin. Error bars are calculated based on eight sensors representing one standard deviation for each point. All measurements were performed via square wave voltammetry with an amplitude of 25 mV, step size of 1 mV, and frequency of either 80 Hz (PBS) or 250 Hz (serum).

When interrogated in undiluted human serum, both the 3trunc and 4trunc vancomycin aptamer variants achieved a 3-fold increase in signal gain (Figure 4B) relative to the maximum gain observed in titrations measured in PBS (Figure 4A). This effect may be due to the presence of polyelectrolytes in serum, which can interact with the aptamer backbone and favor an unfolded state. The addition of target then drives a larger conformational change, which is reflected in the larger sensor gain observed. While we observed a slight loss in affinity for the 4trunc variant when measuring in serum (KD = 87.3 ± 9.1 μM in serum vs KD = 70.7 ± 6 μM in PBS), the 3trunc variant remained within error (KD = 18.1 ± 1.6 μM in serum vs 19.5 ± 1.6 μM in PBS). We speculate that the less structured 4trunc sequence is more affected by nonspecific polyelectrolyte interactions or by changes in viscosity, making it harder to close the aptamer’s stem in the presence of vancomycin and, therefore, achieving worse affinity in serum. The 3trunc sequence, in contrast, benefits from its longer stem to remain more structured despite the increased viscosity of serum.

Conclusions from E-AB Testing of Truncated Aptamers

The 3trunc- and 4trunc-functionalized sensors achieve comparable signal gain in buffered solution and whole serum. However, 3trunc sensors have a broader dynamic range and lower apparent dissociation constant in both fluids, enabling measurements of clinically relevant vancomycin concentrations in clinical serum samples.

Calibration Considerations for Vancomycin Sensing in Static Solutions

One of our goals is to quantify the concentration of vancomycin in unknown samples (e.g., ex vivo clinical serum samples). In pursuit of this goal, we discovered that one must be mindful of how calibration curves are obtained. Specifically, we found that vancomycin-binding E-AB signal gain is different whether the sensors are exposed to (1) sequential increments of vancomycin concentration or (2) a single addition of vancomycin, at any concentration. To demonstrate this effect, we prepared three independent batches of eight 3trunc-functionalized sensors using the above-described conditions. We used one batch to measure a full titration in serum, consisting of 35 points with increasing vancomycin concentrations, ranging from 1 nM to 1 mM (gold data in Figure 5A). The 28th addition corresponded to a concentration of 100 μM vancomycin (top blue square in Figure 5A). We used the second batch to perform 27 electrochemical measurements without vancomycin additions, followed by one vancomycin addition on the 28th measurement to instantaneously raise the concentration to 100 μM (red data in Figure 5A). For this experiment, we did not mix the serum between measurements, other than during the vancomycin addition. The third batch was like the second one, except we pipette-mixed the serum for about thirty seconds after each measurement (black data in Figure 5A). Our goal was to determine if the discrepancy of the gain obtained at 100 μM vancomycin was due to the effects of serum mixing, time, or number of E-AB interrogations. The results show that irrespective of how the sensors were interrogated or the serum was treated, the 3trunc-functionalized sensors challenged by a single addition to reach 100 μM never reached the gain seen when performing full calibration curves (blue boxes in Figure 5A). In fact, repeating these measurements with additional sensor batches but this time targeting lower vancomycin concentrations produced similar results (Figure 5B), with the single vancomycin additions always underperforming in E-AB gain relative to identical concentrations reached by sequentially raising target concentration. This phenomenon also holds true for measurements performed with 4trunc-functionalized E-ABs, even in 50% diluted serum (Figure 5C). Thus, these results suggest that the appropriate way to quantify an unknown concentration of vancomycin using a single-point sensor measurement would be to calibrate the sensor also using single-point vancomycin concentration measurements, instead of the sequential target additions standard in the field.

Figure 5

Figure 5. Single-point serum vancomycin measurements in static solution exhibit worse gain than a full titration. (A) Illustration of this effect by showing a 3trunc full titration curve (gold) side-by-side with single-point additions of vancomycin into unmixed (red) and mixed (black) serum. The full vancomycin titration shown here includes 35 measurements each performed at increasing vancomycin concentrations, converted to time scale using the time stamp on the files generated after each measurement. The two single-addition traces display the same number of E-AB measurements up until the addition of 100 μM vancomycin. The blue squares indicate the E-AB signal gain measured at 100 μM vancomycin concentration 1 min after addition, either via a progressive titration or via a single concentrated spike. (B, C) When comparing the E-AB gain between three points from a full vancomycin titration and single additions to reach equivalent vancomycin concentrations in (B) 100% serum or (C) 50% serum, the titration always achieved higher gains than measurements performed after single additions, for both 3trunc- (B) and 4trunc-functionalized sensors (C), irrespective of the number of measurements or sample mixing.

Conclusions from E-AB Calibrations in Serum

For static solutions, the signal output of E-ABs following a single-point vancomycin addition does not match the output obtained after progressively raising the concentration in a full calibration. Sample mixing improves the gain after single-point additions, likely indicating vancomycin diffusion in serum and interactions with serum proteins affect the response of E-AB sensors in this medium. We note, however, that the amount of mixing time in both green and black data from Figure 5 is the same at the point of vancomycin addition (the 28th point) and afterward, indicating that progressive addition of vancomycin always achieves higher gains.
Considering the low stoichiometric coefficient determined for 3trunc vancomycin affinity via ITC (n ∼ 0.6, Figure 2D), the unexpected improvement in gain when moving from PBS to serum (Figure 4), and the high molecular weight and size of vancomycin (1485.73 g/mol) relative to other small molecules, it is likely that the 3trunc vancomycin interaction does not follow a 1:1 stoichiometry. Our results may further indicate that treatment of the 3trunc aptamer with sequentially increasing concentrations of vancomycin enables a sensor state that cannot be produced via direct addition of a single target concentration. In the future, we aim to pursue further biophysical investigations to better understand the binding physics of this sensor interaction.

Reversibility of 3trunc-Functionalized E-ABs

In addition to single-point measurements in ex vivo clinical serum samples, a potentially important future application of vancomycin-binding E-ABs is continuous, real-time therapeutic drug monitoring, as initially proposed by Dauphin-Ducharme and colleagues. (6) To test if our 3trunc variant could support continuous vancomycin monitoring, we deposited the aptamer onto microfabricated sensors embedded in a flow cell (Figure 6). To create this setup, we three-dimensional (3D)-printed flow cells using methyl methacrylate-based resins as previously reported by our group (Figure 6A). (19) The main inlet of these devices was connected to a peristaltic pump flowing PBS from a beaker. The solutions overfilled the sensor chamber, then flowed out into two external drains located on the sides of each cell and open to an underlying waste beaker. This approach allowed us to mimic excretion kinetics as they would occur in a living rodent by the exit of drug from blood to urine via the kidneys. Two additional, secondary inlets are connected to a computer-controlled syringe pump, enabling us to perform vancomycin injections through one and rapid mixing of drug levels (<10 s) through the other (by pulling liquid out and back in). This mixing rate is analogous to the natural mixing of drugs in the bloodstream of rats as determined by E-AB measurements. (20)

Figure 6

Figure 6. Reversibility and reusability of 3trunc-functionalized E-AB sensors for vancomycin detection in vitro. (A) 3D-printed flow cell to evaluate the reversibility of 3trunc-functionalized E-ABs when exposed to changing vancomycin concentrations. Two pumps, a peristaltic pump for continuous solution flow and a syringe pump for drug additions, are suspended above a plate containing gold microelectrodes. The microelectrodes are functionalized with vancomycin-binding 3trunc aptamer to make E-AB sensors, and interrogated via a handheld potentiostat. Reprinted with permission from Anal. Chem. 2022, 94, 23, 8335–8345. Copyright 2022, Analytical Chemistry. (B) Demonstration of continuous, real-time monitoring of vancomycin in the flow system following three injections of 100 μM vancomycin. Voltammetric measurements were performed every 12 s in PBS containing 2 mM MgCl2. (C) Kinetic differential measurements (KDM) obtained after subtracting data collected at 10 Hz (signal-OFF output) from data collected at 80 Hz (signal-ON output).

To evaluate the reversibility of vancomycin binding to 3trunc-functionalized E-ABs, we serially interrogated the sensors at square wave frequencies of 80 and 10 Hz every 12 s, while flowing PBS at a rate of 0.66 mL/min. To achieve real-time data processing and visualization, we used SACMES, an open-access Python script previously reported by our group. (21) We emulated in vivo IV boluses by repeatedly injecting 100 μM vancomycin concentrations after collecting an initial 10 min long baseline (Figure 6B). We specifically adjusted the injection (0.1 mL/min) and mixing (0.5 mL/min) rates of the syringe pump to emulate a bolus followed by rapid mixing. By performing these measurements, we demonstrate that 3trunc-functionalized sensors support reversible, continuous vancomycin monitoring in flowing buffered solutions.

3trunc-Functionalized E-ABs Support Continuous Vancomycin Monitoring In Vivo

Motivated by the higher sensitivity of 3trunc-functionalized E-ABs in biological media (Figure 4B) and their reversibility in vitro (Figure 6B), we decided to evaluate the performance of the new sensors in vivo. Specifically, our laboratory is pursuing investigations of drug transport across the blood–brain barrier following intravenous (IV) therapeutic dosing. For these experiments, we surgically implanted in vivo probes (Figure 7A) in the right hemisphere of mice, targeting the cortex (Figure 7B). Placement was confirmed via histological staining of brain slices post measurements (Figures 7C and S5). Upon sensor placement, we recorded E-AB baseline signals for 1 h (only last 30 min shown in Figure 7D), prior to administering a bolus of 75 mg/kg vancomycin via the tail vein. The 3trunc-functionalized sensors immediately responded to the bolus, showing a rapid rise of sensor signal that plateaued at ∼1 h after IV dosing and was clearly distinguishable from vehicle boluses (Figure 7D). Although we could not find other published works reporting vancomycin pharmacokinetics in the brain after an IV injection (the only attempt we found could not resolve any brain levels after a 30 mg/kg dose (22)), a similar pharmacokinetic profile has been reported via microdialysis sampling for Fosfomycin. (23) In this case, fosfomycin readily penetrates the blood–brain barrier and has slow excretion kinetics from brain compartments, only decaying from plateau levels 4–6 h after dosing. Although a direct comparison cannot be made due to the molecular size difference between fosfomycin and vancomycin, this publication serves as suggestive evidence supporting the pharmacokinetic profile (i.e., long-lasting plateau) we report in this manuscript.

Figure 7

Figure 7. 3trunc-Functionalized E-ABs support continuous vancomycin monitoring in vivo. (A) Electrode fabrication protocols from the field of fast-scan cyclic voltammetry to make gold-based E-AB probes. The probes are 50 μm in diameter and ∼500 μm in length. (B) Placing the probes on the right hemisphere of mice brains, within the brain cortex (sagittal view). (C) In-brain placement is confirmed via histology post in vivo measurements (coronal view). (D) 3trunc-Based E-AB measurements performed every 2 min revealing an immediate increase of cortex vancomycin levels following IV dosing relative to vehicle boluses. (E) Identical experiments performed using 4trunc-based E-ABs cannot confidently resolve cortex vancomycin levels from background noise. Solid red markers represent the average of three measurements performed on n = 3 independent mice. Shaded areas represent the standard deviation. Solid black markers indicate control measurements (n = 1 for each sensor type) following vehicle dosing.

Alternatively, when comparing the same measurements performed with 4trunc-functionalized sensors, we observed 60% weaker sensor signals (SNR4trunc = 1.03 vs SNR3trunc = 1.62) that could not be fully resolved from the sensor response obtained from vehicle boluses (Figure 7E). The positive drift of these sensors likely arises from error in the correction of sensor baseline via kinetic differential measurements (the dual frequency method used to correct for sensor baselines in real time). In this experiment, we used the same two square wave frequencies employed for the measurements shown in panel D. However, because the sensors are not made with the exact same aptamer sequence, it is possible that sensor drift when using 4trunc is not fully corrected at the employed frequencies. Nevertheless, these results agree with our in vitro (in serum) data, where the 3trunc-functionalized sensors displayed better affinity and dynamic range relative to 4trunc sensors (Figure 4B). Additionally, when comparing the signal gain of the 3trunc sensor to a calibration performed in bovine cerebrospinal fluid (bCSF) in vitro (Figure S6), we note that the signal gain seen in vivo is above the dynamic range and therefore cannot be accurately quantified. This barrier likely arises due to the fact that bCSF does not fully reflect the environment of the cortex. We also injected a high vancomycin dose of 75 mg/kg to clearly demonstrate that, even at 3× the therapeutically effective dose, 4trunc sensors do not respond to the presence of vancomycin in the cortex. In contrast, 3trunc-functionalized sensors achieve a strong signal gain. If we instead convert the in vivo signal output to concentration based on a calibration in PBS (Figure S7), we can determine that the cortex vancomycin concentration plateaus at around 8 μM. In future work, lower doses of vancomycin will be administered to the animals to fully test the ability of the 3trunc aptamer to measure clinically relevant concentrations in vivo.

Conclusions

Click to copy section linkSection link copied!

We followed a systematic approach to evaluate aptamer truncations based on affinity measurements via ITC, conformation-switching dynamics via CD spectroscopy, and electrochemical sensing output in the E-AB format. For this purpose, we made serial base-pair truncations to a previously reported vancomycin-binding aptamer (Table S1). (6) Through our ITC measurements, we observed that truncating the parent aptamer’s stem greatly reduces affinity for vancomycin, but simultaneously broadens the dynamic range of the aptamer (Figures 2 and S1). Qualitative CD measurements suggest that the 3trunc form of the aptamer undergoes stronger binding-induced dynamics than 4trunc at low concentrations of vancomycin, a characteristic desirable for translation to the E-AB platform (Figure 3). Finally, our E-AB measurements revealed 3trunc as the optimal sequence in terms of affinity and broad dynamic range when deployed in undiluted serum (Figure 4).
Our experiments also revealed a few considerations related to this aptamer that require further biophysical investigations beyond the scope of this work. On the one hand, we observed stoichiometric coefficients in our ITC affinity measurements that do not correspond to 1:1 vancomycin–aptamer interactions. This observation suggests that intermolecular interactions between adjacent aptamers may affect binding during our sensor calibration experiments. Moreover, interaptamer interactions may explain why vancomycin-binding E-ABs do not achieve the same signal output in static solutions if vancomycin is added progressively through a sensor titration experiment vs punctual addition to specific target concentrations (Figure 5). In the future, we will dedicate efforts to studying the structure of the vancomycin aptamer via two-dimensional nuclear magnetic resonance measurements coupled to molecular dynamics simulations. Such studies should provide further insights into how vancomycin is binding to these aptamers. We will also continue our translational efforts by pursuing a clinical validation of single-point serum vancomycin E-AB measurements relative to benchmark immunoassays performed at our medical institution.
As final conclusions, this study demonstrated that 3trunc-functionalized E-ABs support reversible continuous sensing in flow systems in vitro, and have superior performance in vivo compared to 4trunc sensors. In addition, this work reports for the first time the continuous monitoring of vancomycin penetration into the cortex of live mice following IV dosing in real time. Unlike previous approaches that relied on microdialysis sampling of cerebral fluid followed by ex vivo molecular quantification of brain drug levels, this study demonstrates the ability of properly optimized E-AB sensors to monitor, in real time, vancomycin transport into the brain of live rodents with minutes resolution and clinically relevant sensitivity.

Methods

Click to copy section linkSection link copied!

Chemicals and Materials

6-Mercaptohexanol (MCH), gold surface cleaning solution, tris(2-carboxyethyl)phosphine hydrochloride (TCEP), and 50 μm gold wire were ordered from Sigma-Aldrich (St. Louis, MO). Gender-pooled human serum was ordered from BioIVT (Washington, D.C.). Phosphate-buffered saline (PBS, 11.9 mM HPO32–; 137 mM NaCl; 2.7 mM KCl; pH = 7.4), trace-metal-grade 18.4 M sulfuric acid (H2SO4), magnesium chloride (MgCl2) hexahydrate, 200 proof ethanol, sodium hydroxide pellets, iron (III) chloride, and hydrochloric acid were purchased from Fisher Scientific (Waltham, MA). Vancomycin hydrochloride, 10% acetic acid, and 0.5 mm platinum wire used for the counter electrodes were ordered from Alfa Aesar (Ward Hill, MA); 2 mm gold working electrodes (PN: CHI101) and Ag|AgCl reference electrodes (PN: CHI111) were purchased from CH Instruments (Austin, TX). Polishing pads (PN: MF-1043), cloth pads (PN: MF-1040), and alumina slurry (PN: CF-1050) were purchased from BASi (West Lafayette, IN). Superfrost Plus charged glass slides, granular paraformaldehyde, HEMO-DE xylene substitute, and 3% hydrogen peroxide were obtained from VWR (Radnor, PA). 30 G 1/2 Precision Glide needles were purchased from BD (Franklin Lakes, NJ). Cu wire (30 AWC) and Betadine (7.5%) were acquired from Amazon. Gold pins (AD1238-ND) were purchased from Digi-Key (Thief River Falls, MN). The anesthetic isoflurane (Fluriso) was obtained from VetOne (Boise, Idaho). Cresyl violet was purchased from Electron Microscopy Sciences (Hatfield, PA). Silver wire was acquired from World Precision Instruments (Sarasota, FL). Stainless steel screws (19010-10) were purchased from Fine Science Tools (Foster City, CA).
All aqueous solutions were prepared using deionized water from a Milli-Q Direct purification system (resistivity = 18 MΩ). All oligonucleotide sequences (Table S1) with and without a 5′-hexanethiol modification and a 3′-methylene blue modification were purchased from and HPLC-purified by Sigma-Aldrich (Houston, TX) with the following structure:
We prepared all DNA solutions by reducing 1 μL aliquots of 100 μM DNA aptamer with 2 μL of 5 mM TCEP for an hour at room temperature. We then added freshly prepared 1 mM MCH to the DNA until a concentration of 200 nM was reached as determined by an Implen Nanophotometer (Munich, Germany, model NP80).
We prepped fresh stock concentrations of vancomycin on the day of titration/experiment. First, we made a stock solution of 10 mM vancomycin in PBS with MgCl2. Next, we made serial dilutions of 1 mM, 100 μM, and 1 μM in either PBS or whole serum. For in vivo experiments, vancomycin was dissolved in 0.5 mL of 1× PBS right before the tail IV injection was performed.

CD Measurements

CD experiments were performed in a 1 mm pathlength cuvette in a total volume of 250 μL. Scans were obtained at 25 °C from 200 to 320 nm, with a 1 nm step and 2 s averaging time. Experiments were conducted in PBS solution with 2 mM Mg2+ and with 0 or 100 μM vancomycin. All experiments were performed on an Aviv (Lakewood, NJ) 420 CD spectrometer. Due to the large CD signal from vancomycin alone, the CD spectra were background-subtracted as follows: [aptamer in buffer ± vancomycin] – [buffer ± vancomycin]. Fresh vancomycin stocks were prepared in buffer no more than 24 h before use and stored at 4 °C. CD experiments were performed in triplicate and analyzed with GraphPad Prism.

ITC Measurements

We used an Affinity ITC (TA Instruments, New Castle, DE) coupled to an autosampler and associated software to run all ITC titrations. We used PBS + 2 mM MgCl2 as a supporting electrolyte to prepare vancomycin and aptamer solutions, and removed air bubbles by keeping the solutions under vacuum for 10 min prior to the ITC experiments. Before each titration, we purged the system three times with deionized water, and let it equilibrate in buffer for 300 s before the first injection. We loaded the syringe with a 300 μM vancomycin solution and the sample cell with at 20 μM solution of the different sequences tested. We recorded the heat-exchanged after the injection of 1 μL of the vancomycin solution for a total of 30 times, allowing the system to equilibrate between 150 and 180 s after each injection. All titrations were done at 25 °C and performed in triplicate.

Fabrication of Electrodes for In Vivo Measurements

In this work, we fabricated the E-ABs probes by adapting the carbon-fiber electrode fabrication protocol from the field of fast-scan cyclic voltammetry (24) to create gold-based microprobes for in-brain measurements (Figure 7A, top). The active sensor windows of the probes were approximately 500 μm long and 50 μm in diameter (Figure 7A, bottom).
The reference electrodes were fabricated by soldering one extreme of Cu-cable to a Au pin and the other extreme to a Ag wire. We added adhesive to the soldered area to provide more support. Before implantation in the brain, we immersed ∼2 mm of the Ag wire in a solution of 0.3 M FeCl3 in 0.1 M HCl for 30 s to produce the Ag/AgCl reference electrode.
For the counter electrodes, we soldered a stainless steel screw to a Cu-cable.

Sensor Preparation for In Vitro Beaker Applications

To prepare electrodes for deposition we first polished the electrodes using a polishing pad and a cloth pad with a drop of alumina slurry on it. Once polished, we placed the electrodes in 10 mL of 200 proof ethanol and sonicated for 30 s. We wiped the electrodes with antistatic wipes before stirring them for 30 s in 10 mL gold surface cleaning solution, after which we immediately placed the electrodes in 10 mL of 200 proof ethanol. We left the electrodes to incubate for 30 min. Afterward, using a CH Instruments electrochemical analyzer (CHI 1040C) multichannel potentiostat and associated software, we used cyclic voltammetry (CV) at a scan rate of 0.5 V to sweep the potential in a window of −0.3 to −1.6 V (vs Ag|AgCl) and back for 200 scans in 0.5 M NaOH. We then placed the electrodes in 0.5 M H2SO4 and used CV to sweep from 0 to 1.6 V (vs Ag|AgCl) for 200 scans at a scan rate of 0.5 V. For those electrodes used in serum measurements, we also increased the electroactive surface area of the electrodes by roughening their surface via chronoamperometry. (25) Briefly, we pulsed from 0 to 2 V (vs Ag|AgCl) with 320 steps and a pulse width of 0.02 s. We then placed prepared electrodes in 1.5 mL Eppendorf tubes containing ∼60 μL of monolayer/DNA solution. After sealing the containers with parafilm, we left the electrodes to incubate overnight at room temperature. Prepared sensors were not stored longer than 24 h before use.

Microneedle Sensor Preparation for In Vitro Flow Cell Applications

Microneedle devices were fabricated by 3D printing using a Formlab 3D printer. The microneedles were sputtered with chromium-supported layers of chromium for 1 min at 100 W (chromium) and gold for 16 min at 100 W. To clean the microneedles, we incubated the microneedle device in pure ethanol for 10 min, followed by electrochemical cleaning in 0.5 M NaOH and 0.5 M H2SO4. (19) Electrochemical cleaning steps include sweeping from −0.4 to −1.4 V (vs Ag/AgCl) for 100 cycles in 0.5 M NaOH at a scan rate of 4 V/s; sweeping from 0.2 to 1.6 V (vs Ag/AgCl) in 0.5 M H2SO4 for 100 cycles at a scan rate of 4 V/s; second time sweeping from 0.2 to −1.6 V (vs Ag/AgCl) in 0.5 M H2SO4 for 100 cycles at a scan rate of 4 V/s. We repeated the ethanol incubation and electrochemical cleaning steps three times to completely remove gold oxides. Prior to sensor fabrication, we immersed the microneedles in 600 μL of 10% commercial gold cleaning solution for 10 s, followed by 10 min ethanol incubation and another round of electrochemical cleaning steps described as above. To prepare E-AB sensors on microneedles, we first reduced 3trunc aptamers by mixing 100 mM TCEP with 100 μM aptamers for 1 h. We then incubated the microneedles with 600 μL 200 μM reduced aptamer solution for 1h, followed by a 3 h-long incubation in 30 mM mercaptohexanol solution at room temperature for sensor dilution. Both aptamer and mercaptohexanol solutions were diluted to the indicated concentration in PBS (11.9 mM HPO32–, 137 mM NaCl, 2.7 mM KCl, pH = 7.4) with added 2 mM MgCl2.

Sensor Preparation for In Vivo Applications

The sensors were fabricated in a similar manner as described before. We electrochemically cleaned the in vivo probes using a Gamry Reference 600+ potentiostat (Warminster, PA) by performing 200 cycles using cyclic voltammetry first from −0.3 to −1.6 V (vs Ag|AgCl) in 0.5 M NaOH at a scan rate of 0.5 V/s, and then from 0 to 1.55 V (vs Ag|AgCl) in 0.5 M H2SO4 also at a scan rate of 0.5 V/s. We then increased the electroactive surface area of the electrodes by roughening their surface via chronoamperometry. (25) For this, we performed 300 pulses (0.01 s pulse width and lowest sensitivity) from 0 to 2 V (vs Ag|AgCl). This was repeated 100 times using a macro (for a total of 30,000 pulses). Finally, we performed an additional 20 cycles in H2SO4 (same conditions as during electrochemical cleaning in H2SO4) to ensure the stability of the roughened surface.
While the electrodes are being cleaned and roughened, a 1 μL aliquot of 100 μM 3trunc or 4trunc aptamer solution was reduced with 2 μL of 5 mM TCEP for 1 h and then diluted to 500 nM using PBS. At this point, the electrodes were thoroughly rinsed with Millipore water before incubating them in the diluted aptamer solution for 2 h at room temperature (manually stirred the electrode in the solution to homogenize the solution/electrode interface). Finally, the electrodes were removed from the aptamer solution and immediately transferred into 1 mM MCH at room temperature overnight. During both incubation periods, we protected the solutions from evaporation by covering the electrodes/solution with a petri dish that was taped to the bench to minimize air flow.
The next day, we removed the electrodes from the MCH solution and thoroughly rinsed with water before placing them in a beaker with PBS solution. We then checked the quality of the sensors by performing square wave voltammograms from 0 to 0.5 V (vs Ag|AgCl) at 30 and 300 Hz (50 mV amplitude and 1 mV step size). At this point, we ran 100 consecutive square wave voltammograms at 300 Hz to “warm up” the sensor and remove any nonspecifically bound aptamer from the electrode surface. We kept the sensors in this PBS solution until ready to implant in the mice brain.

Surgical Procedures for Electrodes Implantation

We used male C57BL/6J mice (Strain #: 000664; The Jackson Laboratory, Bar Harbor, ME) for all in vivo experiments. All procedures were approved by the Johns Hopkins University Animal Care and Use Committee (Protocols: MO19L258 and MO22M234) and in accordance with the Guide for the Care and Use of Laboratory Animals.
Before surgery, we removed the mice from their home cage and placed them in an anesthesia box where we flow O2 at 1 L/min and 5% isoflurane until the mouse is under the surgical place of anesthesia (confirmed by lack of toe-pinch reaction). We shaved the hair off the head from eyes to ears level and, once we immobilized the head in the stereotaxic frame (Stoelting, Wood Dale, IL), we lowered the isoflurane level to 3% (always checking for lack of toe-pinch reaction). We applied betadine to the shaved area using a cotton-tip applicator before making an incision using a scalpel. We removed the connective tissue from the exposed skull using a cotton-tip applicator and added 3% hydrogen peroxide solution to aid in the visualization of the sagittal, bregma, and lambda suture lines (for alignment of the head with the stereotaxic frame). Next, we drilled three holes in the skull with the following coordinates:
1.

Reference electrodes: +4.80 AP (from bregma), −1.00 ML (from bregma), and −2.00 DV (from brain surface)

2.

E-ABs electrodes: −1.70 AP (from bregma), +0.50 ML (from bregma), and −1.00 DV (from brain surface)

3.

Counter-electrode screws: −2.90 AP (from bregma), −2.30 ML (from bregma), and 3–4 full rotations inside the hole.

To find the brain surface with the E-ABs, we connected the Gamry Reference 600+ potentiostat to all three electrodes but kept the E-ABs near the entrance of its hole. We initiated 10 consecutive cyclic voltammograms from 0 to 0.5 V at 0.01 V/s and slowly lowered the sensor tip toward the brain. When the sensor is not touching the brain, the cyclic voltammogram shows just noise because the electrical circuit is open. Once the tip of the sensor touches the brain, a spike in the cyclic voltammogram current is observed, indicating the circuit is closed. At this point, we zeroed the DV coordinate of the digital stereotaxic arm reader and lowered the electrode to its final depth inside the brain. Finally, we run another “warm up” session of 60 square wave voltammograms from 0 to 0.5 V at a frequency of 200 Hz (1 mV step, 50 mV amplitude) to ensure the sensor is homogenized with the brain tissue before starting the experiment.

Electrochemical Measurements In Vivo

We set up the Gamry Reference 600+ potentiostat software so that two consecutive square wave voltammograms at frequencies of 15 and 200 Hz (from 0 to 0.5 V vs Ag|AgCl, 1 mV potential step, 50 mV amplitude) were run every 2 min (no need for better time resolution) for the duration of the experiment (∼5 h). We recorded the baseline sensor signal for 1 h before administering a 75 mg/kg bolus via tail vein injection (30G needles). We continued monitoring the sensor signal for at least 3 more hours post IV injection. Immediately after the sensor monitoring was stopped, we passed 300 μA of current for 5 s (using chronopotentiometry) through the probe to make a lesion in the area of the brain in contact with the Au wire. This facilitated sensor placement identification during histology.

Electrochemical Measurements In Vitro

All in vitro electrochemical measurements were performed using a CH Instruments electrochemical analyzer (CHI 1040C) multichannel potentiostat and associated software. To keep the cell at a constant temperature, a special glass cell was employed made by the glassblowing shop at the University of Maryland. This allows a water bath to be linked to the cell where the temperature of the water regulates the temperature of the cell. To interrogate the electrodes, we used square wave voltammetry (SWV) with frequencies of 5, 10, 40, 80, and 250 Hz with an amplitude of 25 mV and a quiet time of 2 s while scanning from 0 to −0.5 V (vs Ag|AgCl). We used the frequency with the highest signal gain for all data; 80 Hz for PBS, 200 Hz for CSF, and 250 Hz for serum. Prior to vancomycin measurements, we preconditioned the electrodes until a flat baseline was established (∼20 scans).

Data Fitting

We fit our in vitro cerebrospinal fluid titration (Figure S6) to the following Hill equation
%signal=0.748+57.26+0.7481+(0.00000967/x)0.696
The signal output of 3trunc sensors in vivo in the cortex was ∼57%, which would calculate to ∼100 mM vancomycin in the brain. However, because this gain is higher than the maximum we observed for the in vitro calibration, the calculated values are not accurate.
We fit our in vitro PBS titration using the electrode configuration of our in vivo probes (Figure S7, left) to the following Hill equation
%signal=0.45986+123.22+0.459861+(0.000011357/x)0.4045
We converted the signal output of 3trunc sensors in vivo to concentration versus time using this equation (Figure S7, right).

Time Course Experiments

Time course experiments were kept at a constant temperature of 23 °C. We interrogated the electrodes using SWV with the conditions described above. Between each set of tested frequencies, we let the electrodes go without electrochemical stimulation for at least 90 s to simulate the time it takes to perform another titration. For time courses taken that include mechanical mixing, the cell was mixed for ∼35 s before being allowed to rest for ∼60 s. Time of the measurements was calculated based on the time stamp of the data files and plotted against total signal gain.

Flow Cell Measurements

We serially interrogated the sensors at square wave frequencies of 80 and 10 Hz every 12 s, while flowing undiluted serum at a rate of 0.66 mL/min. To achieve real-time data processing and visualization, we used SACMES, an open-access Python script previously reported by our group. (21) We emulated in vivo IV boluses by injecting vancomycin at a concentration of 100 μM, after collecting an initial 10 min long baseline (Figure 6B). We specifically adjusted the injection (0.1 mL/min) and mixing (0.5 mL/min) rates of the syringe pump to emulate a bolus followed by rapid mixing.

Brain Tissue Staining and Histology

After we performed the lesions in the brain tissue via chronopotentiometry, we decapitated the mice and extracted the brains into 4% paraformaldehyde. We stored all mice brains in this solution until ready to begin the tissue staining and histology protocols described hereafter.
Sliced brain tissue was stained using Cresyl Violet. First, we mounted the brain slices on charged glass slides. After allowing the tissue and glass to dry overnight, we performed consecutive immersions of the mounted glass slides in the following solutions: (1) 95% ethanol for 15 min, (2) 70% ethanol for 2 min, (3) 50% ethanol for 2 min, (4) deionized water for 3 min, (5) deionized water for 3 min, (6) Cresyl Violet for 5 min, (7) deionized water for 2 min, (8) 50% ethanol for 2 min, (9) short dips in 95% ethanol with 1 mL of 10% until the desired coloration was obtained, (10) 100% ethanol for 3 min, (11) 100% ethanol for 3 min, and (12) HEMO-DE xylene substitution for 5 min. After preparing the glass slides for imaging, we analyzed the stained tissue using an Aperio CS2 microscope scanner (Leica Biosystems, Nussloch, Germany) and compared the acquired images to the Franklin & Paxinos brain map (26) to confirm the final stereotaxic coordinates of the implanted sensors.

Supporting Information

Click to copy section linkSection link copied!

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acssensors.2c01910.

  • Aptamer variant sequences, measured ITC parameters, and Nupack-predicted secondary structures (Table S1)

  • and additional ITC, CD, and electrochemical measurements (Figures S1–S7) (PDF)

Terms & Conditions

Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.

Author Information

Click to copy section linkSection link copied!

  • Corresponding Author
    • Netzahualcóyotl Arroyo-Currás - Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United StatesChemistry-Biology Interface Program, Zanvyl Krieger School of Arts & Sciences, Johns Hopkins University, Baltimore, Maryland 21218, United StatesBiochemistry, Cellular and Molecular Biology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21287, United StatesOrcidhttps://orcid.org/0000-0002-2740-6276 Email: [email protected]
  • Authors
    • Alexander Shaver - Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United StatesOrcidhttps://orcid.org/0000-0002-5478-5291
    • J.D. Mahlum - Chemistry-Biology Interface Program, Zanvyl Krieger School of Arts & Sciences, Johns Hopkins University, Baltimore, Maryland 21218, United StatesOrcidhttps://orcid.org/0000-0003-0220-3058
    • Karen Scida - Lieber Institute for Brain Development, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United States
    • Melanie L. Johnston - Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United StatesBiochemistry, Cellular and Molecular Biology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21287, United States
    • Miguel Aller Pellitero - Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United StatesOrcidhttps://orcid.org/0000-0001-8739-2542
    • Yao Wu - Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United States
    • Gregory V. Carr - Department of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United StatesLieber Institute for Brain Development, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, United StatesOrcidhttps://orcid.org/0000-0002-6091-6729
  • Author Contributions

    A.S. and J.D.M. contributed equally to this work. A.S. and J.D.M. performed all of the electrochemical measurements. M.L.J. collected CD spectra. M.A.P. performed the affinity ITC measurements. Y.W. designed and built the flow system and performed all continuous E-AB measurements. K.S. and G.V.C. contributed to the experimental design and conducted all in vivo measurements. N.A.C. collaboratively designed the experiments with A.S., J.D.M., and K.S. All authors contributed to the writing of this manuscript.

  • Notes
    The authors declare no competing financial interest.

Acknowledgments

Click to copy section linkSection link copied!

The authors thank Zio Health for providing research-sponsored funds in support of this work. They also thank Dr. Katie Tripp and the JHU Center for Molecular Biophysics for providing access to CD spectrometers and technical assistance. The in vivo work reported here was supported by the National Institute of General Medical Sciences of the National Institutes of Health under Award Number R01GM140143. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health

References

Click to copy section linkSection link copied!

This article references 26 other publications.

  1. 1
    Organization, W. H. World Health Organization Model List of Essential Medicines: 21st List , 2019, Vol. 21.
  2. 2
    Rybak, M. J.; Le, J.; Lodise, T. P.; Levine, D. P.; Bradley, J. S.; Liu, C.; Mueller, B. A.; Pai, M. P.; Wong-Beringer, A.; Rotschafer, J. C. Therapeutic Monitoring of Vancomycin for Serious Methicillin-Resistant Staphylococcus Aureus Infections: A Revised Consensus Guideline and Review by the American Society of Health-System Pharmacists, the Infectious Diseases Society of America, the Pediatric Infectious Diseases Society, and the Society of Infectious Diseases Pharmacists. Am. J. Heal. Pharm. 2020, 77, 835863,  DOI: 10.1093/ajhp/zxaa036
  3. 3
    Dilworth, T. J.; Schulz, L. T.; Rose, W. E. Vancomycin Advanced Therapeutic Drug Monitoring: Exercise in Futility or Virtuous Endeavor to Improve Drug Efficacy and Safety?. Clin. Infect. Dis. 2021, 72, E675E681,  DOI: 10.1093/cid/ciaa1354
  4. 4
    Elbarbry, F. Vancomycin Dosing and Monitoring: Critical Evaluation of the Current Practice. Eur. J. Drug Metab. Pharmacokinet. 2018, 43, 259268,  DOI: 10.1007/s13318-017-0456-4
  5. 5
    Chen, C. Y.; Li, M. Y.; Ma, L. Y.; Zhai, X. Y.; Luo, D. H.; Zhou, Y.; Liu, Z. M.; Cui, Y. M. Precision and Accuracy of Commercial Assays for Vancomycin Therapeutic Drug Monitoring: Evaluation Based on External Quality Assessment Scheme. J. Antimicrob. Chemother. 2020, 75, 21102119,  DOI: 10.1093/jac/dkaa150
  6. 6
    Dauphin-Ducharme, P.; Zhang, Y.; Gerson, J.; Kurnik, M.; Kippin, T. E.; Stojanovic, M. N. Electrochemical Aptamer-Based Sensors for Improved Therapeutic Drug Monitoring and High-Precision, Feedback-Controlled Drug Delivery ́. ACS Sens. 2019, 4, 28322837,  DOI: 10.1021/acssensors.9b01616
  7. 7
    Li, H.; Dauphin-ducharme, P.; Ortega, G.; Plaxco, K. W. Calibration-Free Electrochemical Biosensors Supporting Accurate Molecular Measurements Directly in Undiluted Whole Blood. J. Am. Chem. Soc. 2017, 139, 1120711213,  DOI: 10.1021/jacs.7b05412
  8. 8
    Arroyo-Currás, N.; Somerson, J.; Vieira, P. A.; Ploense, K. L.; Kippin, T. E.; Plaxco, K. W. Real-Time Measurement of Small Molecules Directly in Awake, Ambulatory Animals. Proc. Natl. Acad. Sci. U.S.A. 2017, 114, 645650,  DOI: 10.1073/pnas.1613458114
  9. 9
    Parolo, C.; Idili, A.; Ortega, G.; Csordas, A.; Hsu, A.; Arroyo-Currás, N.; Yang, Q.; Ferguson, B. S.; Wang, J.; Plaxco, K. W. Real-Time Monitoring of a Protein Biomarker. ACS Sens. 2020, 5, 18771881,  DOI: 10.1021/acssensors.0c01085
  10. 10
    Idili, A.; Gerson, J.; Parolo, C.; Kippin, T.; Plaxco, K. W. An Electrochemical Aptamer-Based Sensor for the Rapid and Convenient Measurement of l-Tryptophan. Anal. Bioanal. Chem. 2019, 411, 46294635,  DOI: 10.1007/s00216-019-01645-0
  11. 11
    Zaric, R. Z.; Milovanovic, J.; Rosic, N.; Milovanovic, D.; Zecevic, D. R.; Folic, M.; Jankovic, S. Pharmacokinetics of Vancomycin in Patients with Different Renal Function Levels. Open Med. 2018, 13, 512519,  DOI: 10.1515/med-2018-0068
  12. 12
    Suzuki, Y.; Kawasaki, K.; Sato, Y.; Tokimatsu, I.; Itoh, H.; Hiramatsu, K.; Takeyama, M.; Kadota, J. I. Is Peak Concentration Needed in Therapeutic Drug Monitoring of Vancomycin? A Pharmacokinetic-Pharmacodynamic Analysis in Patients with Methicillin- Resistant Staphylococcus Aureus Pneumonia. Chemotherapy 2012, 58, 308312,  DOI: 10.1159/000343162
  13. 13
    Turnbull, W. B.; Daranas, A. H. On the Value of c: Can Low Affiniity Systems Be Studied by Isothermal Titration Calorimetry?. J. Am. Chem. Soc. 2003, 125, 1485914866,  DOI: 10.1002/asi.4630310417
  14. 14
    Zadeh, J. N.; Steenberg, C. D.; Bois, J. S.; Wolfe, B. R.; Pierce, M. B.; Khan, A. R.; Dirks, R. M.; Pierce, N. A. NUPACK: Analysis and Design of Nucleic Acid Systems. J. Comput. Chem. 2011, 32, 170173,  DOI: 10.1002/jcc.21596
  15. 15
    Chamorro-Garcia, A.; Ortega, G.; Mariottini, D.; Green, J.; Ricci, F.; Plaxco, K. W. Switching the Aptamer Attachment Geometry Can Dramatically Alter the Signalling and Performance of Electrochemical Aptamer-Based Sensors. Chem. Commun. 2021, 57, 1169311696,  DOI: 10.1039/d1cc04557a
  16. 16
    Shaver, A.; Curtis, S. D.; Arroyo-Currás, N. Alkanethiol Monolayer End Groups Affect the Long-Term Operational Stability and Signaling of Electrochemical, Aptamer-Based Sensors in Biological Fluids. ACS Appl. Mater. Interfaces 2020, 12, 1121411223,  DOI: 10.1021/acsami.9b22385
  17. 17
    White, R. J.; Phares, N.; Lubin, A. A.; Xiao, Y.; Plaxco, K. W. Optimization of Electrochemical Aptamer-Based Sensors via Optimization of Probe Packing Density and Surface Chemistry. Langmuir 2008, 24, 1051310518,  DOI: 10.1021/la800801v
  18. 18
    Liu, Y.; Canoura, J.; Alkhamis, O.; Xiao, Y. Immobilization Strategies for Enhancing Sensitivity of Electrochemical Aptamer-Based Sensors. ACS Appl. Mater. Interfaces 2021, 13, 94919499,  DOI: 10.1021/acsami.0c20707
  19. 19
    Wu, Y.; Tehrani, F.; Teymourian, H.; Mack, J.; Shaver, A.; Reynoso, M.; Kavner, J.; Huang, N.; Furmidge, A.; Duvvuri, A. Microneedle Aptamer-Based Sensors for Continuous, Real-Time Therapeutic Drug Monitoring. Anal. Chem. 2022, 94, 83358345,  DOI: 10.1021/acs.analchem.2c00829
  20. 20
    Arroyo-Currás, N.; Dauphin-Ducharme, P.; Ortega, G.; Ploense, K. L.; Kippin, T. E.; Plaxco, K. W. Subsecond-Resolved Molecular Measurements in the Living Body Using Chronoamperometrically Interrogated Aptamer-Based Sensors. ACS Sens. 2018, 3, 360366,  DOI: 10.1021/acssensors.7b00787
  21. 21
    Curtis, S. D.; Ploense, K. L.; Kurnik, M.; Ortega, G.; Parolo, C.; Kippin, T. E.; Plaxco, K. W.; Arroyo-Currás, N. Open Source Software for the Real-Time Control, Processing, and Visualization of High-Volume Electrochemical Data. Anal. Chem. 2019, 91, 1232112328,  DOI: 10.1021/acs.analchem.9b02553
  22. 22
    Miura, Y.; Fuchigami, Y.; Nomura, S.; Nishimura, K.; Hagimori, M.; Kawakami, S. Brain Microdialysis Study of Vancomycin in the Cerebrospinal Fluid After Intracerebroventricular Administration in Mice. AAPS PharmSciTech 2019, 20, 5  DOI: 10.1208/s12249-018-1232-8
  23. 23
    Nau, R.; Sörgel, F.; Eiffert, H. Penetration of Drugs through the Blood-Cerebrospinal Fluid/Blood-Brain Barrier for Treatment of Central Nervous System Infections. Clin. Microbiol. Rev. 2010, 23, 858883,  DOI: 10.1128/CMR.00007-10
  24. 24
    Meunier, C. J.; Sombers, L. A. Fast-Scan Voltammetry for In Vivo Measurements of Neurochemical Dynamics. In The Brain Reward System; Springer, 2021.
  25. 25
    Arroyo-Currás, N.; Scida, K.; Ploense, K. L.; Kippin, T. E.; Plaxco, K. W. High Surface Area Electrodes Generated via Electrochemical Roughening Improve the Signaling of Electrochemical Aptamer-Based Biosensors. Anal. Chem. 2017, 89, 1218512191,  DOI: 10.1021/acs.analchem.7b02830
  26. 26
    Franklin, K.; Paxinos, G. The Mouse Brain in Stereotaxic Coordinates, 3rd ed.; Elsevier Academic Press: San Diego, 2007.

Cited By

Click to copy section linkSection link copied!
Citation Statements
Explore this article's citation statements on scite.ai

This article is cited by 32 publications.

  1. Karen Scida, Elysse Ornelas-Gatdula, Michael DePasquale, Gregory V. Carr, Netzahualcóyotl Arroyo-Currás. Therapeutic Drug Distribution across the Mouse Brain Is Heterogeneous as Revealed by In Vivo, Spatially Resolved Aptamer-Based Sensing. ACS Pharmacology & Translational Science 2025, Article ASAP.
  2. Lisa C. Fetter, Matthew H. McDonough, Tod E. Kippin, Kevin W. Plaxco. Effects of Physiological-Scale Variation in Cations, pH, and Temperature on the Calibration of Electrochemical Aptamer-Based Sensors. ACS Sensors 2024, 9 (12) , 6675-6684. https://doi.org/10.1021/acssensors.4c02274
  3. Lisa Deflandre, Philippe Dauphin-Ducharme. “Binding” or “Binding and Switching”? A Perspective on Resolving Conformational Changes of Surface-Attached Biomolecular Receptors. ACS Sensors 2024, 9 (11) , 5615-5625. https://doi.org/10.1021/acssensors.4c01677
  4. Elsi Verrinder, Julian Gerson, Kaylyn Leung, Tod E. Kippin, Kevin W. Plaxco. Dual-Frequency, Ratiometric Approaches to EAB Sensor Interrogation Support the Calibration-Free Measurement of Specific Molecules In Vivo. ACS Sensors 2024, 9 (6) , 3205-3211. https://doi.org/10.1021/acssensors.4c00516
  5. Lars Alexander Olivan, Kaitlyn Hand, Ryan J. White. Utilization of Spontaneous Alkyne–Gold Self-Assembly Chemistry as an Alternative Method for Fabricating Electrochemical Aptamer-Based Sensors. Langmuir 2024, 40 (23) , 12117-12123. https://doi.org/10.1021/acs.langmuir.4c00972
  6. Yu Liu, John O. Mack, Maryam Shojaee, Alexander Shaver, Ankitha George, William Clarke, Neel Patel, Netzahualcóyotl Arroyo-Currás. Analytical Validation of Aptamer-Based Serum Vancomycin Monitoring Relative to Automated Immunoassays. ACS Sensors 2024, 9 (1) , 228-235. https://doi.org/10.1021/acssensors.3c01868
  7. Alexander Shaver, Kyle Mallires, Jonathan Harris, Jonathan Kavner, Bo Wang, Rebecca Gottlieb, Juan Lión-Villar, María Ángeles Herranz, Nazario Martín, Netzahualcóyotl Arroyo-Currás. Survey of Conductive Polymers for the Fabrication of Conformation Switching Nucleic Acid-Based Electrochemical Biosensors. ACS Applied Polymer Materials 2024, 6 (1) , 541-551. https://doi.org/10.1021/acsapm.3c02206
  8. Ya-Chen Tsai, Wei-Yang Weng, Yu-Tung Yeh, Jun-Chau Chien. Dual-Aptamer Drift Canceling Techniques to Improve Long-Term Stability of Real-Time Structure-Switching Aptasensors. ACS Sensors 2023, 8 (9) , 3380-3388. https://doi.org/10.1021/acssensors.3c00509
  9. Qiuyue Yang, Julia Pedreira-Rincón, Leire Balerdi-Sarasola, Luis Baptista-Pires, Jose Muñoz, Daniel Camprubí-Ferrer, Andrea Idili, Claudio Parolo. An aptamer-based electrochemical sensor for the quantification of the malaria biomarker lactate dehydrogenase. Biosensors and Bioelectronics 2025, 274 , 117152. https://doi.org/10.1016/j.bios.2025.117152
  10. Jian Yu, Bozhao Xiao, Jianlong Qiu, Yuanbin Tang, Yanjie Guo, Lei Yang, Naipeng Li, Kai Jiang. Artificial intelligence-enabled smart monitoring of vancomycin concentration using a droplet-driven triboelectric nanogenerator. Nano Energy 2025, 136 , 110691. https://doi.org/10.1016/j.nanoen.2025.110691
  11. Malek Bibani, Magdolna Casian, Bogdan Feier, Diana Bogdan, Oana Hosu-Stancioiu, Nadia Ktari, Rafik Kalfat, Cecilia Cristea. Electrochemical aptasensor for the selective detection of vancomycin based on nanostructured “in-lab” printed electrodes. Microchimica Acta 2025, 192 (2) https://doi.org/10.1007/s00604-025-06952-1
  12. Zhou‐Jie Liu, Yu‐Qi Liang, Jia‐Yi Li, Bing Wu, Chen Huang, Yi‐Wei Liu, Chen‐Zhi Zhang, Ye Yang, Nai‐Qing Cai, Jin‐Yuan Chen, Xin‐Hua Lin. Engineered Aptamer‐Derived Fluorescent Aptasensor: the Label‐Free, Single‐Step, Rapid Detection of Vancomycin in Clinical Samples. Small 2025, 21 (5) https://doi.org/10.1002/smll.202407799
  13. Jing Li, Vincent Clark, Chen‐Hsu Yu, Karen Scida, Miguel Aller Pellitero, Rolando L. Albarracín Rivera, Wenrui Zhong, Erin Demek, Jeffrey Fountain, J.D. Mahlum, Richard E. Haaland, Gregory V. Carr, Jonathan Sczepanski, Netzahualcóyotl Arroyo‐Currás. Monitoring HIV Antiretroviral Therapy via Aptamer‐Based Measurements in Preclinical Animal Models, in Human Plasma. Advanced Sensor Research 2025, https://doi.org/10.1002/adsr.202400191
  14. Alejandro Chamorro-Garcia, Gabriel Ortega-Quintanilla, Andrea Idili, Claudio Parolo. Structure Switching Bioreceptors as Novel Tools for Point-of-Care Diagnostics. 2025, 490-510. https://doi.org/10.1016/B978-0-323-99967-0.00266-0
  15. Cátia Santa, Soohyun Park, Artur Gejt, Heather A. Clark, Bastian Hengerer, Khulan Sergelen. Real-time monitoring of vancomycin using a split-aptamer surface plasmon resonance biosensor. The Analyst 2024, 150 (1) , 131-141. https://doi.org/10.1039/D4AN01226G
  16. Laura Ferrer Pascual, Eero Gustafsson, Juha Siitonen, Vasuki Durairaj, Tomi Laurila. Challenges in aptamer-based sensor development using carbon nanotube networks. Sensors & Diagnostics 2024, 3 (12) , 1935-1946. https://doi.org/10.1039/D4SD00250D
  17. Minh-Dat Nguyen, Meghan T. Osborne, Guy Terence Prevot, Zachary R. Churcher, Philip E. Johnson, Lena Simine, Philippe Dauphin-Ducharme. Truncations and in silico docking to enhance the analytical response of aptamer-based biosensors. Biosensors and Bioelectronics 2024, 265 , 116680. https://doi.org/10.1016/j.bios.2024.116680
  18. John Mack, Raygan Murray, Kenedi Lynch, Netzahualcóyotl Arroyo-Currás. 3D-printed electrochemical cells for multi-point aptamer-based drug measurements. Sensors & Diagnostics 2024, 3 (9) , 1533-1541. https://doi.org/10.1039/D4SD00192C
  19. Tran N. H. Nguyen, Lisa F. Horowitz, Timothy Krilov, Ethan Lockhart, Heidi L. Kenerson, Taranjit S. Gujral, Raymond S. Yeung, Netzahualcóyotl Arroyo-Currás, Albert Folch. Label-free, real-time monitoring of cytochrome C drug responses in microdissected tumor biopsies with a multi-well aptasensor platform. Science Advances 2024, 10 (36) https://doi.org/10.1126/sciadv.adn5875
  20. Isabel M. Jensen, Vincent Clark, Harper L. Kirby, Netzahualcóyotl Arroyo-Currás, David M. Jenkins. Tuning N-heterocyclic carbene wingtips to form electrochemically stable adlayers on metals. Materials Advances 2024, 5 (17) , 7052-7060. https://doi.org/10.1039/D4MA00648H
  21. Seyed Vahid Hamidi, Arash Khorrami Jahromi, Imman I. Hosseini, Roozbeh Siavash Moakhar, Cesar Collazos, Qinghua Pan, Chen Liang, Sara Mahshid. Surface‐Based Multimeric Aptamer Generation and Bio‐Functionalization for Electrochemical Biosensing Applications. Angewandte Chemie 2024, 136 (32) https://doi.org/10.1002/ange.202402808
  22. Seyed Vahid Hamidi, Arash Khorrami Jahromi, Imman I. Hosseini, Roozbeh Siavash Moakhar, Cesar Collazos, Qinghua Pan, Chen Liang, Sara Mahshid. Surface‐Based Multimeric Aptamer Generation and Bio‐Functionalization for Electrochemical Biosensing Applications. Angewandte Chemie International Edition 2024, 63 (32) https://doi.org/10.1002/anie.202402808
  23. Shihao Pei, Samuel Babity, Ana Sara Cordeiro, Davide Brambilla. Integrating microneedles and sensing strategies for diagnostic and monitoring applications: The state of the art. Advanced Drug Delivery Reviews 2024, 210 , 115341. https://doi.org/10.1016/j.addr.2024.115341
  24. Élodie V. d’Astous, Philippe Dauphin-Ducharme. Whole blood multiplex measurements using electrochemical aptamer-based biosensors. Chemical Communications 2024, 60 (50) , 6419-6422. https://doi.org/10.1039/D4CC01452A
  25. Eun Ryung Kim, Thi Thanh-Thao Dang, Su Jin Lee, Thi Thanh-Qui Nguyen, Jee-Woong Park, Man Bock Gu. A highly sensitive sandwich-type electrochemical aptasensor using a pair of novel truncated aptamers for the detection of vaspin. Chemical Engineering Journal 2023, 477 , 147002. https://doi.org/10.1016/j.cej.2023.147002
  26. Attilio Marino, Matteo Battaglini, Marie Celine Lefevre, Maria Cristina Ceccarelli, Kamil Ziaja, Gianni Ciofani. Sensorization of microfluidic brain-on-a-chip devices: Towards a new generation of integrated drug screening systems. TrAC Trends in Analytical Chemistry 2023, 168 , 117319. https://doi.org/10.1016/j.trac.2023.117319
  27. Yechang Guo, Shaofeng Wang, Peiyue Li, Pan Zhang, Wei Wang. Rapid Colloidal Gold Immunoassay for Pharmacokinetic Evaluation of Vancomycin in the Cerebrospinal Fluid and Plasma of Beagle Dogs. Sensors 2023, 23 (21) , 8978. https://doi.org/10.3390/s23218978
  28. Yanwei Fu, Chunjing Yang, Congmin Liu, Xiqiao Xu, Zhengyuan Shi, Dechun Jiang, Dan Yan. A simple and rapid strategy for monitoring vancomycin concentration in serum using a multicolor immunosensor based on the ratio of gold nanobipyramids and a product of cetyltrimethylammonium bromide-blue oxide of 3,3′,5,5′-tetramethylbenzidine interaction. Analytical Methods 2023, 15 (30) , 3650-3660. https://doi.org/10.1039/D3AY00873H
  29. Yuqiao Liu, Junmin Li, Shenghao Xiao, Yanhui Liu, Mingxia Bai, Lixiu Gong, Jiaqian Zhao, Dajing Chen. Revolutionizing Precision Medicine: Exploring Wearable Sensors for Therapeutic Drug Monitoring and Personalized Therapy. Biosensors 2023, 13 (7) , 726. https://doi.org/10.3390/bios13070726
  30. Yuchan Yuan, Netzahualcóyotl Arroyo-Currás. Continuous molecular monitoring in the body via nucleic acid–based electrochemical sensors: The need for statistically-powered validation. Current Opinion in Electrochemistry 2023, 39 , 101305. https://doi.org/10.1016/j.coelec.2023.101305
  31. Charlotte Flatebo, William R. Conkright, Meaghan E. Beckner, Robert H. Batchelor, Tod E. Kippin, Jason Heikenfeld, Kevin W. Plaxco. Efforts toward the continuous monitoring of molecular markers of performance. Journal of Science and Medicine in Sport 2023, 26 , S46-S53. https://doi.org/10.1016/j.jsams.2023.01.010
  32. Amani A. Hariri, Alyssa P. Cartwright, Constantin Dory, Yasser Gidi, Steven Yee, Kaiyu Fu, Kiyoul Yang, Diana Wu, Ian A. P. Thompson, Nicolò Maganzini, Trevor Feagin, Brian E. Young, Behrad Habib Afshar, Michael Eisenstein, Michel Digonnet, Jelena Vuckovic, H. Tom Soh. Continuous optical detection of small-molecule analytes in complex biomatrices. 2023https://doi.org/10.1101/2023.03.03.531030
  33. Alexander Shaver, Netzahualcóyotl Arroyo-Currás. Expanding the Monolayer Scope for Nucleic Acid-Based Electrochemical Sensors Beyond Thiols on Gold: Alkylphosphonic Acids on ITO. ECS Sensors Plus 2023, 2 (1) , 010601. https://doi.org/10.1149/2754-2726/acc4d9

ACS Sensors

Cite this: ACS Sens. 2022, 7, 12, 3895–3905
Click to copy citationCitation copied!
https://doi.org/10.1021/acssensors.2c01910
Published November 23, 2022

Copyright © 2022 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY-NC-ND 4.0 .

Article Views

3861

Altmetric

-

Citations

Learn about these metrics

Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.

Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.

The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.

  • Abstract

    Figure 1

    Figure 1. Vancomycin binding to E-ABs results in an electrochemical signal change. (A) E-ABs consist of several parts: (i) the electrode, (ii) the self-assembled monolayer, (iii) the SAM-bound aptamer (which binds to the target, vancomycin), and (iv) the redox reporter. (B) The redox reporter exchanges electrons at rates dependent on the reporter–electrode distance, keT1 and keT2. Because target binding induces conformational changes in the aptamer that affect the reporter–electrode distance, such changes result in observable changes in voltammetric current. These voltammograms were measured via square wave voltammetry at a frequency of 250 Hz, amplitude of 25 mV, and step size of 1 mV in phosphate-buffered saline (PBS) with 2 mM MgCl2. (C) Change in voltammetric peak current (red markers) or area under the curve (blue markers) can then be plotted to build dose–response curves from which thermodynamic parameters can be extracted. Here, for example, we performed nonlinear regression of the curve to a Hill model (red trace), obtaining a dissociation constant of 11.7 ± 0.7 μM and a Hill coefficient of 0.66 ± 0.02. The clinically relevant range for serum vancomycin is 6–30 μM as indicated in Table 1. (2,6)

    Figure 2

    Figure 2. ITC measurements reveal worsening of vancomycin affinity with increasing aptamer truncations. (A) The modeled secondary structure of the aptamers with symmetric truncations considered in this work. From left to right: no truncations, one (1trunc), two (2trunc), three (3trunc), and four (4trunc) base-pair truncations. This modeling was done using Nupack (14) assuming 140 mM Na+ and 2 mM Mg2+ at a temperature of 25 °C. (B) An example of two raw ITC graphs. Negative is exothermic. (C) We calculated the areas for each titration point and plotted them against the vancomycin to aptamer mole ratio in the ITC cell. We then fit the curves to a binding polynomial (red, black, and blue traces). KD for the full aptamer is 0.14 ± 0.02 μM, KD for 3trunc is 3.92 ± 0.90 μM, and KD for 4trunc is 52 ± 20 μM. Error bars represent the standard deviation of the heat and mole ratio. Errors are calculated from three replicates. Shading represents the propagation of error of the 95% confidence interval for each regression based on three replicates. (D) Red markers represent the KD, and blue markers represent the n of each symmetrical truncation. Dashed lines are point-to-point connections. Error bars are based on three replicates. All measurements were done in PBS solution with 2 mM MgCl2 added.

    Figure 3

    Figure 3. CD spectroscopy illustrates the change in aptamer conformation upon vancomycin binding. CD spectra for each construct with no vancomycin (red trace) and 100 μM vancomycin (black trace). Red and gray shaded areas represent the standard deviation of three measurements without and with vancomycin, respectively. Measurements were done at a concentration of 5 μM aptamer in PBS solution with 2 mM Mg2+. Insets are expanded regions to highlight band shift upon vancomycin addition. For the 4trunc sequence (C), minimal vancomycin binding-induced conformational change is seen compared to full (A) or 3trunc (B). Measurements were performed at 25 °C, with an averaging time of 2 seconds, and a 1 mm path length. Data were background-subtracted ([aptamer ± vancomycin] – [buffer ± vancomycin]).

    Figure 4

    Figure 4. Performance of truncated aptamers in E-AB format. Here, we show dose–response curves for E-ABs functionalized with a 200 nM deposition concentration of (A) 3trunc (KD = 19.5 ± 1.6 μM) and 4trunc (KD = 70.7 ± 6 μM) aptamers in PBS with 2 mM Mg2+ or (B) 3trunc (KD = 18.1 ± 1.6 μM) and 4trunc (KD = 87.3 ± 9.1 μM) E-ABs in whole serum. The y-axis in these titrations represents signal gain, computed as the change in signal relative to measurements in the absence of vancomycin. Error bars are calculated based on eight sensors representing one standard deviation for each point. All measurements were performed via square wave voltammetry with an amplitude of 25 mV, step size of 1 mV, and frequency of either 80 Hz (PBS) or 250 Hz (serum).

    Figure 5

    Figure 5. Single-point serum vancomycin measurements in static solution exhibit worse gain than a full titration. (A) Illustration of this effect by showing a 3trunc full titration curve (gold) side-by-side with single-point additions of vancomycin into unmixed (red) and mixed (black) serum. The full vancomycin titration shown here includes 35 measurements each performed at increasing vancomycin concentrations, converted to time scale using the time stamp on the files generated after each measurement. The two single-addition traces display the same number of E-AB measurements up until the addition of 100 μM vancomycin. The blue squares indicate the E-AB signal gain measured at 100 μM vancomycin concentration 1 min after addition, either via a progressive titration or via a single concentrated spike. (B, C) When comparing the E-AB gain between three points from a full vancomycin titration and single additions to reach equivalent vancomycin concentrations in (B) 100% serum or (C) 50% serum, the titration always achieved higher gains than measurements performed after single additions, for both 3trunc- (B) and 4trunc-functionalized sensors (C), irrespective of the number of measurements or sample mixing.

    Figure 6

    Figure 6. Reversibility and reusability of 3trunc-functionalized E-AB sensors for vancomycin detection in vitro. (A) 3D-printed flow cell to evaluate the reversibility of 3trunc-functionalized E-ABs when exposed to changing vancomycin concentrations. Two pumps, a peristaltic pump for continuous solution flow and a syringe pump for drug additions, are suspended above a plate containing gold microelectrodes. The microelectrodes are functionalized with vancomycin-binding 3trunc aptamer to make E-AB sensors, and interrogated via a handheld potentiostat. Reprinted with permission from Anal. Chem. 2022, 94, 23, 8335–8345. Copyright 2022, Analytical Chemistry. (B) Demonstration of continuous, real-time monitoring of vancomycin in the flow system following three injections of 100 μM vancomycin. Voltammetric measurements were performed every 12 s in PBS containing 2 mM MgCl2. (C) Kinetic differential measurements (KDM) obtained after subtracting data collected at 10 Hz (signal-OFF output) from data collected at 80 Hz (signal-ON output).

    Figure 7

    Figure 7. 3trunc-Functionalized E-ABs support continuous vancomycin monitoring in vivo. (A) Electrode fabrication protocols from the field of fast-scan cyclic voltammetry to make gold-based E-AB probes. The probes are 50 μm in diameter and ∼500 μm in length. (B) Placing the probes on the right hemisphere of mice brains, within the brain cortex (sagittal view). (C) In-brain placement is confirmed via histology post in vivo measurements (coronal view). (D) 3trunc-Based E-AB measurements performed every 2 min revealing an immediate increase of cortex vancomycin levels following IV dosing relative to vehicle boluses. (E) Identical experiments performed using 4trunc-based E-ABs cannot confidently resolve cortex vancomycin levels from background noise. Solid red markers represent the average of three measurements performed on n = 3 independent mice. Shaded areas represent the standard deviation. Solid black markers indicate control measurements (n = 1 for each sensor type) following vehicle dosing.

  • References


    This article references 26 other publications.

    1. 1
      Organization, W. H. World Health Organization Model List of Essential Medicines: 21st List , 2019, Vol. 21.
    2. 2
      Rybak, M. J.; Le, J.; Lodise, T. P.; Levine, D. P.; Bradley, J. S.; Liu, C.; Mueller, B. A.; Pai, M. P.; Wong-Beringer, A.; Rotschafer, J. C. Therapeutic Monitoring of Vancomycin for Serious Methicillin-Resistant Staphylococcus Aureus Infections: A Revised Consensus Guideline and Review by the American Society of Health-System Pharmacists, the Infectious Diseases Society of America, the Pediatric Infectious Diseases Society, and the Society of Infectious Diseases Pharmacists. Am. J. Heal. Pharm. 2020, 77, 835863,  DOI: 10.1093/ajhp/zxaa036
    3. 3
      Dilworth, T. J.; Schulz, L. T.; Rose, W. E. Vancomycin Advanced Therapeutic Drug Monitoring: Exercise in Futility or Virtuous Endeavor to Improve Drug Efficacy and Safety?. Clin. Infect. Dis. 2021, 72, E675E681,  DOI: 10.1093/cid/ciaa1354
    4. 4
      Elbarbry, F. Vancomycin Dosing and Monitoring: Critical Evaluation of the Current Practice. Eur. J. Drug Metab. Pharmacokinet. 2018, 43, 259268,  DOI: 10.1007/s13318-017-0456-4
    5. 5
      Chen, C. Y.; Li, M. Y.; Ma, L. Y.; Zhai, X. Y.; Luo, D. H.; Zhou, Y.; Liu, Z. M.; Cui, Y. M. Precision and Accuracy of Commercial Assays for Vancomycin Therapeutic Drug Monitoring: Evaluation Based on External Quality Assessment Scheme. J. Antimicrob. Chemother. 2020, 75, 21102119,  DOI: 10.1093/jac/dkaa150
    6. 6
      Dauphin-Ducharme, P.; Zhang, Y.; Gerson, J.; Kurnik, M.; Kippin, T. E.; Stojanovic, M. N. Electrochemical Aptamer-Based Sensors for Improved Therapeutic Drug Monitoring and High-Precision, Feedback-Controlled Drug Delivery ́. ACS Sens. 2019, 4, 28322837,  DOI: 10.1021/acssensors.9b01616
    7. 7
      Li, H.; Dauphin-ducharme, P.; Ortega, G.; Plaxco, K. W. Calibration-Free Electrochemical Biosensors Supporting Accurate Molecular Measurements Directly in Undiluted Whole Blood. J. Am. Chem. Soc. 2017, 139, 1120711213,  DOI: 10.1021/jacs.7b05412
    8. 8
      Arroyo-Currás, N.; Somerson, J.; Vieira, P. A.; Ploense, K. L.; Kippin, T. E.; Plaxco, K. W. Real-Time Measurement of Small Molecules Directly in Awake, Ambulatory Animals. Proc. Natl. Acad. Sci. U.S.A. 2017, 114, 645650,  DOI: 10.1073/pnas.1613458114
    9. 9
      Parolo, C.; Idili, A.; Ortega, G.; Csordas, A.; Hsu, A.; Arroyo-Currás, N.; Yang, Q.; Ferguson, B. S.; Wang, J.; Plaxco, K. W. Real-Time Monitoring of a Protein Biomarker. ACS Sens. 2020, 5, 18771881,  DOI: 10.1021/acssensors.0c01085
    10. 10
      Idili, A.; Gerson, J.; Parolo, C.; Kippin, T.; Plaxco, K. W. An Electrochemical Aptamer-Based Sensor for the Rapid and Convenient Measurement of l-Tryptophan. Anal. Bioanal. Chem. 2019, 411, 46294635,  DOI: 10.1007/s00216-019-01645-0
    11. 11
      Zaric, R. Z.; Milovanovic, J.; Rosic, N.; Milovanovic, D.; Zecevic, D. R.; Folic, M.; Jankovic, S. Pharmacokinetics of Vancomycin in Patients with Different Renal Function Levels. Open Med. 2018, 13, 512519,  DOI: 10.1515/med-2018-0068
    12. 12
      Suzuki, Y.; Kawasaki, K.; Sato, Y.; Tokimatsu, I.; Itoh, H.; Hiramatsu, K.; Takeyama, M.; Kadota, J. I. Is Peak Concentration Needed in Therapeutic Drug Monitoring of Vancomycin? A Pharmacokinetic-Pharmacodynamic Analysis in Patients with Methicillin- Resistant Staphylococcus Aureus Pneumonia. Chemotherapy 2012, 58, 308312,  DOI: 10.1159/000343162
    13. 13
      Turnbull, W. B.; Daranas, A. H. On the Value of c: Can Low Affiniity Systems Be Studied by Isothermal Titration Calorimetry?. J. Am. Chem. Soc. 2003, 125, 1485914866,  DOI: 10.1002/asi.4630310417
    14. 14
      Zadeh, J. N.; Steenberg, C. D.; Bois, J. S.; Wolfe, B. R.; Pierce, M. B.; Khan, A. R.; Dirks, R. M.; Pierce, N. A. NUPACK: Analysis and Design of Nucleic Acid Systems. J. Comput. Chem. 2011, 32, 170173,  DOI: 10.1002/jcc.21596
    15. 15
      Chamorro-Garcia, A.; Ortega, G.; Mariottini, D.; Green, J.; Ricci, F.; Plaxco, K. W. Switching the Aptamer Attachment Geometry Can Dramatically Alter the Signalling and Performance of Electrochemical Aptamer-Based Sensors. Chem. Commun. 2021, 57, 1169311696,  DOI: 10.1039/d1cc04557a
    16. 16
      Shaver, A.; Curtis, S. D.; Arroyo-Currás, N. Alkanethiol Monolayer End Groups Affect the Long-Term Operational Stability and Signaling of Electrochemical, Aptamer-Based Sensors in Biological Fluids. ACS Appl. Mater. Interfaces 2020, 12, 1121411223,  DOI: 10.1021/acsami.9b22385
    17. 17
      White, R. J.; Phares, N.; Lubin, A. A.; Xiao, Y.; Plaxco, K. W. Optimization of Electrochemical Aptamer-Based Sensors via Optimization of Probe Packing Density and Surface Chemistry. Langmuir 2008, 24, 1051310518,  DOI: 10.1021/la800801v
    18. 18
      Liu, Y.; Canoura, J.; Alkhamis, O.; Xiao, Y. Immobilization Strategies for Enhancing Sensitivity of Electrochemical Aptamer-Based Sensors. ACS Appl. Mater. Interfaces 2021, 13, 94919499,  DOI: 10.1021/acsami.0c20707
    19. 19
      Wu, Y.; Tehrani, F.; Teymourian, H.; Mack, J.; Shaver, A.; Reynoso, M.; Kavner, J.; Huang, N.; Furmidge, A.; Duvvuri, A. Microneedle Aptamer-Based Sensors for Continuous, Real-Time Therapeutic Drug Monitoring. Anal. Chem. 2022, 94, 83358345,  DOI: 10.1021/acs.analchem.2c00829
    20. 20
      Arroyo-Currás, N.; Dauphin-Ducharme, P.; Ortega, G.; Ploense, K. L.; Kippin, T. E.; Plaxco, K. W. Subsecond-Resolved Molecular Measurements in the Living Body Using Chronoamperometrically Interrogated Aptamer-Based Sensors. ACS Sens. 2018, 3, 360366,  DOI: 10.1021/acssensors.7b00787
    21. 21
      Curtis, S. D.; Ploense, K. L.; Kurnik, M.; Ortega, G.; Parolo, C.; Kippin, T. E.; Plaxco, K. W.; Arroyo-Currás, N. Open Source Software for the Real-Time Control, Processing, and Visualization of High-Volume Electrochemical Data. Anal. Chem. 2019, 91, 1232112328,  DOI: 10.1021/acs.analchem.9b02553
    22. 22
      Miura, Y.; Fuchigami, Y.; Nomura, S.; Nishimura, K.; Hagimori, M.; Kawakami, S. Brain Microdialysis Study of Vancomycin in the Cerebrospinal Fluid After Intracerebroventricular Administration in Mice. AAPS PharmSciTech 2019, 20, 5  DOI: 10.1208/s12249-018-1232-8
    23. 23
      Nau, R.; Sörgel, F.; Eiffert, H. Penetration of Drugs through the Blood-Cerebrospinal Fluid/Blood-Brain Barrier for Treatment of Central Nervous System Infections. Clin. Microbiol. Rev. 2010, 23, 858883,  DOI: 10.1128/CMR.00007-10
    24. 24
      Meunier, C. J.; Sombers, L. A. Fast-Scan Voltammetry for In Vivo Measurements of Neurochemical Dynamics. In The Brain Reward System; Springer, 2021.
    25. 25
      Arroyo-Currás, N.; Scida, K.; Ploense, K. L.; Kippin, T. E.; Plaxco, K. W. High Surface Area Electrodes Generated via Electrochemical Roughening Improve the Signaling of Electrochemical Aptamer-Based Biosensors. Anal. Chem. 2017, 89, 1218512191,  DOI: 10.1021/acs.analchem.7b02830
    26. 26
      Franklin, K.; Paxinos, G. The Mouse Brain in Stereotaxic Coordinates, 3rd ed.; Elsevier Academic Press: San Diego, 2007.
  • Supporting Information

    Supporting Information


    The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acssensors.2c01910.

    • Aptamer variant sequences, measured ITC parameters, and Nupack-predicted secondary structures (Table S1)

    • and additional ITC, CD, and electrochemical measurements (Figures S1–S7) (PDF)


    Terms & Conditions

    Most electronic Supporting Information files are available without a subscription to ACS Web Editions. Such files may be downloaded by article for research use (if there is a public use license linked to the relevant article, that license may permit other uses). Permission may be obtained from ACS for other uses through requests via the RightsLink permission system: http://pubs.acs.org/page/copyright/permissions.html.