Skip to main content
NCBI home page
NIST Author Manuscripts logoLink to NIST Author Manuscripts
. Author manuscript; available in PMC: 2020 Jun 23.
Published in final edited form as: ACS Sens. 2017 Oct 9;2(11):1644–1652. doi: 10.1021/acssensors.7b00486

Nanowire Aptasensors for Electrochemical Detection of Cell-Secreted Cytokines

Ying Liu 1,2, Ali Rahimian 1, Sergiy Krylyuk 3,4, Tam Vu 1, Bruno Crulhas 1,5, Gulnaz Stybayeva 1,6, Dong-Sik Shin 1,7, Albert V Davydov 4, Alexander Revzin 1,6
PMCID: PMC7311073  NIHMSID: NIHMS1588424  PMID: 28991491

Abstract

Cytokines are small proteins secreted by immune cells in response to pathogens/infections; therefore these proteins can be used in diagnosing infectious diseases. For example, release of a cytokine interferon (IFN)-γ from T-cells is used for blood-based diagnosis of tuberculosis (TB). Our lab has previously developed an aptamer-based electrochemical biosensor for rapid and sensitive detection of IFN-γ. In this study, we explored the use of silicon nanowires (Si NWs) as a way to create nanostructured electrodes with enhanced sensitivity for IFN-γ. Si NWs were covered with gold and were further functionalized with thiolated aptamers specific for IFN-γ. Aptamer molecules were designed to form a hairpin and in addition to terminal thiol groups contained redox reporter molecules methylene blue. Binding of analyte to aptamer-modified NWs (termed here nanowire aptasensors) inhibited electron transfer from redox reporters to the electrode and caused electrochemical redox signal to decrease. In a series of experiments we demonstrate that NW aptasensors responded 3x faster and were 2x more sensitive to IFN-γ compared to standard flat electrodes. Most significantly, NW aptasensors allowed detection of IFN-γ from as few as 150 T-cells/mL while ELISA did not pick up signal from the same number of cells. One of the challenges faced by ELISA-based TB diagnostics is poor performance in patients whose T-cell numbers are low, typically HIV patients. Therefore, NW aptasensors developed here may be used in the future for more sensitive monitoring of IFN-γ responses in patients co-infected with HIV/TB.

1. Introduction

Nanostructures such as nanowires (NWs) offer interesting advantages for sensing related to geometry, electrical conductivity and high surface area-to-volume ratio. Recent studies have shown the wide application of NWs as biosensor devices [1, 2], cell culture substrates [3, 4], and drug delivery vehicles [5]. NW-based devices have also been used for stimulating, rescording from, and delivering objects to single cells and tissues [6].

Vertically grown NWs have become increasing popular recently for cell culture and analysis. Recent studies demonstrated vertical NWs for penetrating cell membrane and delivering a variety of biomolecules including siRNA, peptides, DNA, and proteins into primary neurons and fibroblasts [7, 8]. “Nanostraw” structures have also been used to build fluidic connection for direct cytosolic access and molecule delivery [9].

In addition to intracellular delivery, vertical NWs have been used for intracellular and extracellular biosensing. Nanostructured surfaces are particularly attractive in electrochemistry where such surfaces enhance diffusion, decrease response times and improve electrode sensitivity [10]. Lieber group developed arrays of nano field effect transistors (nanoFET) based on vertical NWs and demonstrated intracellular recording of transmembrane potential from a single cardiomyocytes [11, 12]. Kelley group has developed nanostructured electrodes that allowed to detecting bacterial nucleic acids at attomolar concentration [13, 14]. The same group integrated nanostructured microelectrodes with microfluidic lysis of cancer cells for detection of intracellular RNA[15].

Our lab has been interested in developing biosensors for monitoring cell function [16–18], with particular emphasis on methods that allow continuous monitoring of cellular secretions [19–21]. We have previously reported on the development of electrochemical aptasensors for detection of IFN-γ and other cytokines [22, 23]. These aptasensors function as proposed by Plaxco group and others [24–26] and described in Scheme 1. Binding of analyte to aptamer-functionalized electrode causes a change in aptamer structure, decreases (or increases) rate of electron transfer and shifts the electrochemical redox signal. We hypothesized that advantages of nanostructured surfaces – improved transport and enhanced surface area per footprint – would result in more sensitive aptasensors for analysis of cell-secreted signals. Therefore, we employed Au-covered vertical Si NWs as working electrodes for aptamer-based detection of IFN-γ, an important inflammatory cytokine. Nanostructured electrodes had 3.18 fold faster response time and 3.90 fold better sensitivity to IFN-γ compared to standard flat electrodes. Most strikingly, nanostructured aptasensors enabled detection of IFN-γ release from as few as 150 T-cells per mL of solution. Standard IFN-γ ELISA was not sensitive to detect a signal from such a low number of cells.

Scheme 1.

Scheme 1.

Open in a new tab

a) Schematic of aptamer-based electrochemical sensor for IFN-γ. The aptamer was originally in hairpin structure and redox label was in close proximity to electrode surface. Upon IFN-γ binding, the hairpin changed confirmation and the redox label moved further away from the electrode, lowering the electron-transfer efficiency. The differences in faradaic current before and after IFN-γ binding were quantified using square wave voltammetry (SWV). b) Different behaviors of cell deposition and cytokine IFN-γ secretion for floating cells and surface deposited cells on planar and NW electrodes.

2. Experimental section

2.1. Materials

HEPES, sodium chloride (NaCl), potassium chloride (KCl), magnesium chloride (MgCl2), sodium bicarbonate (NaHCO3), dimethylformamide (DMF), 6-mercapto-1-hexanol (MCH), tris-(2-carboxyethyl) phosphine hydrochloride (TCEP), (3-Aminopropyl)trimethoxysilane (APTMS), N,N-Diisopropylethylamine (DIPEA), Bovine serum albumin (BSA), Paraformaldehyde (PFA), T cell activation reagents: phorbol 12-myristate-13-acetate (PMA) and ionomycin were purchased from Sigma-Aldrich (St. Louis, MO).1 Silicon wafer was purchased from WRS materials (San Jose, CA) and coated with gold in LGA Thin films (Santa Clara, CA). Human recombinant Interferon gamma (IFN-γ), Tumor necrosis factor alpha (TNF-α), Interleukin-12 (IL-12), Interleukin-6 (IL-6), Interleukin-10 (IL-10), and ELISA kit were purchased from R&D systems (Minneapolis, MN). FITC Mouse Anti-Human CD3 antibody was obtained from from BD Pharmingen. Cell culture medium RPMI 1640 with L-glutamine, Alexa Fluor 488 carboxylic acid succinimidyl ester, Alexa Fluor 594 carboxylic acid succinimidyl ester, Fluorescence-based live/dead cell viability kit were purchased from Life Technologies (Carlsbad, CA), and CellTracker Green CMFDA dye. One step NBT/BCIP western blotting test kit was purchased from Thermo Scientific (Waltham, MA). Methylene blue-carboxylic acid succinimidyl ester (MB-NHS) was purchased from Biosearch Technologies Inc. (Novato, CA). Jurkat cell line was obtained from ATCC.

The IFN-γ-binding aptamer was synthesized by IDT Technologies (San Diego, CA). The 34-mer IFN-γ-binding aptamer sequence was as follows: 5’-NH2-C6-GGG GTT GGT TGT GTT GGG TGT TGT GTC CAA CCC C-C3-SH-3’. IFN-γ aptamer were modified at the 3-terminus with a C3-disulfide [HO(CH2)3-S-S-(CH2)3-] linker and at the 5-terminus with an amine group for redox probe (MB) conjugation. The aptamer was dissolved in HEPES buffer (10mM HEPES, 150mM NaCl, 5mM KCl, 5mM MgCl2) (pH 7.4). Thiolated aptamer was reacted with 10 mM TCEP for 1h at room temperature to reduce the disulfide bond of the probe.

2.2. Conjugation of redox reporter to aptamers

The redox reporter MB-NHS was conjugated to the 5’-terminus of amino modified IFN-γ aptamer through succinimide ester coupling following previously reported protocols [22]. Briefly, 0.5M NaHCO3 was added to 100 μM aptamer solution to adjust solution pH, followed by the addition of MB-NHS for conjugation. The mixture solution was stirred and allowed to react for 4h at 4°C in the dark. After conjugation, MB-IFN-γ was filtered using centrifugal filter (Millipore, Amicon Ultra 3K 0.5mL) and stored at −20°C before use.

2.3. Growth of silicon nanowires

Silicon nanowires (Si NWs) were grown in a hot-walled chemical vapor deposition (CVD) system as reported elsewhere [27]. Briefly, 10 standard cubic centimeters per min (sccm) of SiCl4 vapor and 200 sccm of H2 were diluted with N2 carrier gas to a total flow rate of 1000 sccm through CVD system at 900 °C and 80 kPa. The growth of Si NWs was catalyzed by 30 nm, 100 nm and 200 nm Au nanoparticles (NPs) dispersed on poly-L-lysine functionalized n-Si (111) substrates, which produced NWs with approximately the same diameters. Growth time was adjusted to produce NWs with a required length, ranging from 6 μm to 145 μm. The resulting Si NWs were sputtering coated with a 10 nm Cr / 30 nm Au for electrochemical characterization and electrochemical biosensor application. Back side of a Si substrate was coated with 10 nm Cr / 100 nm Au using e-beam deposition.

2.4. Assembly of electrochemical cell and preparation of aptasensors

The IFN-γ aptasensosr were prepared in a home-made Teflon-based sample holder/electrochemical cell. Gold coated Si NW substrates were placed into the sample holder, creating a cylindrical electrochemical (e)-cell with inner diameter of 10.8 mm and volume of 0.5 mL. Ag/AgCl (3M KCl) reference electrode and a Pt wire counter electrode were immersed into e-cell for making electrochemical measurements. HEPES buffer or RPMI 1640 cell culture media were used as electrolyte.

The NWs or planar Si substrates Au-coated similarly to SiNW samples were cleaned with oxygen plasma, then thoroughly rinsed with DI water, and dried under nitrogen. Prior to modification of the electrodes, aptamer stock solution (50 μM) was reduced in 10 mM TCEP for 1 h to cleave disulfide bonds. This solution was then diluted in HEPES buffer to 1 μM and pipetted into the e-cell preloaded with NW or planar substrates. Aptamer solution was incubated with electrode surfaces overnight in the dark at 4 °C. Subsequently, electrodes were rinsed with copious amounts of DI water and then immersed in an aqueous solution of 3 mM MCH for 1 h to passivate the electrode surface. As a final step, the electrodes were rinsed with DI water, dried with nitrogen, and stored at 4 °C prior to use.

2.5. Electrochemical characterization of NW electrodes and calibration of IFN-γ aptasensor

Electrochemical measurements were carried out with a CHI 842B potentiostat (CH Instruments, Austin, TX) operating with a three electrode system as described above. Electrochemical characterization of NW electrodes was performed using cyclic voltammetry (CV) in 5 mM Fe(SCN)6 3−/4− within 0.1 M KCl. Electrochemical detection of IFN-γ from NW electrodes was performed using square wave voltammetry (SWV) with a voltage range of −0.5 V to 0 V and 40 mV amplitude, 60 Hz frequency.

After assembling MB-IFN-γ aptamer, 1 mL of HEPES buffer or RPMI 1640 medium was infused into an e-cell and an electrode was allowed to equilibrate for 30 min, as determined by stable faradaic current. Small amounts of concentrated human recombinant IFN-γ solutions were pipetted into the reaction well to adjust concentrations from 0.2 ng/mL to 280 ng/mL. Before conducting SWV measurements, the sensor was allowed to react with analyte for 30 min to ensure cytokine-aptamer binding. Binding of IFN-γ caused the redox peak to shift down. The change in redox activity was reported as signal suppression - a ratio of (initial peak redox current – final peak redox current) / initial peak redox current. To demonstrate specificity, an atpamer- modified electrodes were challenged with 100 ng/mL IFN-γ, as well as nonspecific cytokines including IL-12, IL-6, IL-10 and BSA.

2.6. Purification of CD4 T-cells from human blood

The whole blood was collected from healthy adult donors through venipuncture under sterile conditions with informed consent and approval of the Institutional Review Board of the University of California at Davis (Protocol No. 222894–6). Peripheral blood mononuclear cells (PBMCs) were purified using a density gradient media, Lymphoprep (Axis-shield, Oslo, Norway). CD4+ T-cells were isolated from the purified PBMC population through negative selection by magnetic beads using the EasySep human CD4+ T cell enrichment kit (Stemcell Technologies, Vancouver, BC, Canada) according to the manufacturer instructions and cultured in RPMI 1640 with 10 % fetal bovine serum and 1 % penicillin/streptomycin (Gibco, Chino, CA, United States). The cell count and viability test was performed using 0.4 % trypan blue solution on hemocytometer (Thermo Fisher scientific Inc, San Diego, CA, United States). Phorbol myristate acetate (50 ng/mL) and Ionomycin (1 μg/mL) were the stimuli used to induce secretion of IFN-γ and TNF-α (Sigma-Aldrich, St. Louis, MO, United States).

For experiments utilizing chimeric antibodies, T-cell population was isolated from the purified PBMC through negative selection by magnetic beads using the EsaySep human T cell enrichment kit (Stemcell Technologies, Vancouver, BC, Canada) according to manufacturer instructions. Isolated T cells were labeled with the IFN-γ capture chimeric antibody (Miltenyi Biotec, Bergisch Gladbach, Germany) according to the instructions provided by the manufacture and stimulated with PMA (50 ng/mL) and Ionomycin (1 μg/mL). The supernatant then collected and analyzed by IFN-γ/TNF-α DuoSet ELISA kit (R&D systems, Minneapolis, MN, United States).

2.7. Scanning electron microscopy

300,000 T-cells were suspended in 1 mL RPMI and cultured on 1 cm x 1cm NW electrode substrate for 4 h in an incubator at 37 °C to allow cells to settle down. The surface was then rinsed with water three times to get rid of the unattached cells and incubated in 1 % aqueous solution of PFA for 15 min to fix the cells attaching to the substrate. The samples were then dehydrated in gradually increasing concentrations of ethanol (30 % – 100 %) in water, dried in a critical point dryer, and sputter-coated with a few nanometers of gold to increase conductivity for SEM characterization. As a control experiment, the same concentration of T-cells was incubated and fixed on a flat substrate.

2.8. Analysis of IFN-γ release from T-cells

For cell secretion experiment, the Teflon device with integrated electrodes was placed into a custom-made heating incubator to maintain 37 °C and 5 % CO2. T-cell suspension in RPMI 1640 was added into an E-cell and incubated for 30 min allowing cells to settle down. Subsequently, mitogens (PMA (50 ng/mL) and ionomycin (2 mM)) dissolved in RPMI 1640 were added into e-cell to stimulate T-cells to produce cytokines. SWV measurements commenced immediately upon mitogenic stimulation and were made every 10 min for the first 50 min, every 20 min after for the course of 100 min. Different cell number from 150 to 50,000 were suspended in 1 mL RPMI 1640 media and allowed to sediment onto planar and NW electrodes. Different geometries of NW electrodes were evaluated to determine the configuration yielding the most sensitive detection.

2.9. Cell viability analysis

This experiment was carried out to determine whether cell viability was affected by placement into E-cell and subsequent electrochemical measurements. 300,000 T-cells were placed into an E-cell containing 0.5 mL of RPMI1640. The cells were mitogenically stimulated and IFN-γ release measurements were made over the course of 2 h. Subsequently, media was removed and cells were assessed for viability by LIVE/DEAD staining and fluorescence microscopy (Axiovert 200M, Zeiss, Germany).

3. Results and discussion

3.1. Characterization of NW electrodes

Building nanostructured electrodes significantly increases electrode area per foot print, enhances analyte capture efficiency and improves analyte transport to the electrode. Figure 1a shows a typical SEM image of 100 nm diameter NWs used in this study. As seen from this image, the majority of NWs remained upright – the orientation most optimal for function of the aptasensor. While NWs of different geometries were tested during this study, nanostructures most commonly used were 6 μm long and 100 nm diameter with surface grafting density of about (24 ± 4) × 106 NWs/cm2. The NWs were randomly distributed on the surface with distances between individual nanostructures ranging from 2 μm to 10 μm, which provided enough space for cells to land.

Figure 1. Characterization of NW electrodes.

Figure 1.

Open in a new tab

a) SEM of vertical nanowires at 30° tilt view. Scale bar – 10 μm. b) Cyclic voltammetry of planar Au (red dash line) and NW electrode (black solid line) in 5mM [Fe(CN)6]3− containing 0.1 M KCl at scanning rate of 100mVs−1. c) Square wave voltammetry (SWV) for planar (red dashed line) and NW electrodes (black solid line) functionalized with IFN-γ MB aptamer.

Upon sputtering Au and creating NW electrodes, we examined electrochemical properties of these electrodes using [Fe(CN)6]3−/4− cyclic voltammetry. The electrodes used in these experiments had NWs of 6 pm height and 100 μm diameter. Figure 1b shows typical [Fe(CN)6]3−/4− cyclic voltammetry curves generated from flat and nanostructured electrodes. As seen from these data the redox peaks for nanostructured surfaces were ≈2.5 times higher than standard flat electrodes. The enhancement in redox signaling is in good agreement with 2.3 fold enhancement in surface area of NW electrodes. This experiment highlights the fact that nanostructured electrode surface was easily accessible to molecules diffusing from solution. Next, we functionalized planar and NW electrodes with MB-labeled aptamers and carried out SWV analysis. As seen from Figure 1c the reduction peak current for NW electrodes was ≈2.3 times higher than for planar electrodes. This is logical given the increase in surface area for NW electrodes. SWV curves were also used to determine aptamer density on flat and nanostructured surfaces (see SI for detailed discussion of surface density determination). The MB-IFN-γ aptamer surface immobilization densities were estimated to be 5.97 × 1013 molecules/cm2 for NWs electrode and 1.27 × 1013 molecules/cm2 for planar electrode. The results were in general agreement with other reports employing MB-tagged hairpin aptamers [28–30]. One should note that values above are reported for footprint area. After accounting for higher surface area of NW electrodes, the density of aptamer molecules on the surface remains ≈2 fold higher compared to a planar electrode. The higher aptamer packing density on NW electrodes may be explained by curvature effects when nanostructure curvature approaches the size of a DNA molecule [31].

3.2. Electrochemical detection of IFN-γ using aptamer-modified NW electrodes

We have previously designed IFN-γ beacon that contains a hairpin with stem region partially obstructing aptamer nucleotides [30]. As shown in Scheme 1, hairpin probes carry redox reporters (MB) so that binding of analyte molecules causes conformation of the hairpin molecules to change and redox current to decrease. The loss in signal is presented in terms of signal suppression defined as (initial redox peak current – final redox peak current)/(original redox current).

Figure 2(a,b) show typical SWV voltammograms in RPMI cell culture media with MB reduction peak appearing at −0.25 V (vs Ag/AgCl). Challenging sensing electrode with varying concentrations of recombinant IFN-γ from 0.2 ng/mL to 280 ng/mL, resulted in decrease of MB redox current, the larger the concentration the larger the decrease in redox current. The SWV results were converted into calibration curve by plotting normalized change in redox peak current (signal suppression) vs. concentration of cytokines. MB redox peak linearly decreased in response to the addition of IFN-γ protein for both AuNW and planar electrodes (Figure 2a, b). However, nanowire aptasensors exhibited greater changes in signal in response to IFN-γ compared to planar electrodes. The dynamic range of nanowire aptasensor reached 100 ng/mL and 75 % signal suppression whereas values for planar aptasensor were 280 ng/mL and 41 %. The limit of detection, calculated as 3x signal/noise, was 0.14 ng/mL and 0.8 ng/mL for nanowire and planar aptasensor, respectively. Control experiments in Figure 3c refer to spiking aliquots of HEPES without IFN-γ and monitoring changes in the biosensor response. These control experiments point to stability of the aptasensor background signal.

Figure 2. Performance of NW aptasensors.

Figure 2.

Open in a new tab

a) SWV responses of NW aptasensors challenged with IFN-γ ranging from 0.2 ng/mL to 230 ng/mL; b) SWV analysis of planar aptasensors exposed to IFN-γ ranging from 1 ng/mL to 280 ng/mL; c) Calibration curves constructed by plotting sensors responses to various concentrations of cytokines for NW aptasensor (IFN-γ spiked: solid black square; control: open black square) and planar aptasensor (IFN-γ spiked: solid red circle; control: open red circle). The data points and error bars represent average and standard deviations of measurements from three different aptasensors. d) Specificity of NW aptasensor. Electrodes were challenged with IFN-γ, TNF-α, IL-6, IL-12, IL-10 and BSA at 100 ng/mL. Error bars are standard deviation from the average of three different aptasensors.

Figure 3. Effects of NW density and geometry on detection of IFN-γ from cells.

Figure 3.

Open in a new tab

a-b) SEM images of leukocytes on electrodes with 100 nm diameter wires at 30° tilt view. Scale bar for a) is 50 μm, for b) 2 μm. c) Detection of IFN-γ from NWs of the same length (6 μm) but different diameters. d) Detection of IFN-γ using electrodes with nanowires of the same diameter (100 nm) but different lengths.

In addition to sensitivity and dynamic range, specificity is a key requirement for a biosensor. To demonstrate specificity, NW aptasensors were challenged with inflammatory cytokines that may be produced by immune cells, TNF-α, IL-6, IL-10, IL-12, as well as with BSA – most abundant protein in blood serum. Figure 2d demonstrated MB redox peak in response to different cytokines, signal suppression less than 5 % was measured after exposure to 100 ng/mL of nonspecific proteins. Challenging the NWs aptasensor with 100 ng/mL IFN-γ was used as the positive control and resulted in 75 % signal suppression. This experiment highlights that nanowire aptasensors were indeed specific to the cytokine of interest - IFN-γ.

Several observations can be made by comparing response of NW and planar aptasensors. First, signal suppression values are almost 2x higher for nanostructured surfaces compared to planar electrodes. Signal suppression is a normalized change in redox signal and describes aptamer-analyte engagement events that lead to a change in redox current. This normalization process eliminates electrode surface area of NW electrodes as a factor contributing to signal enhancement. It would appear that NW aptasensors are either more efficient at binding cytokine molecules compared to planar aptasensors or are better at relaying change in electrochemical signal due to cytokine binding events. The first explanation is supported by an observation that aptamer surface density on NW electrodes for >2 fold higher compared to planar electrodes.

3.3. Interactions of cells with NW substrates

In the next set of experiments we wanted to characterize the nature of T-cell interactions with nanostructured surfaces. T-cells suspended in RPMI at 500,000 cell/mL were incubated with planar and NW surfaces for 2 h. This is an approximate time frame of a cytokine detection experiment. Subsequently, surfaces were imaged by SEM. Figure 3 (a–b) demonstrates T-cells on substrates containing 6 μm long, 100 nm diameter NWs spaced out by 2 μm to 10 μm. As seen from these images, cells were captured between individual nanowires but not impaled on the nanowires. Based on these observations it was possible to presume that NW electrodes will be available for detection of extracellular, secreted molecules such as IFN-γ. In the case of high cell concentration, deposition of cells onto a planar electrode surface may result in obstruction of recognition sites and loss of the sensor function. As highlighted by images in Figure 3(a–b), such obstruction may not affect aptamer-functionalized NWs which remain free of cells and available of detection of cell-secreted molecules. Importantly, interactions with NW surfaces did not compromise viability of cells. Live/dead staining (not shown here) revealed that 88% of cells were viable when incubated with planar and NW electrodes under physiological conditions for 4h.

3.4. Effect of geometry on NW aptasensor performance

In the next set of experiments we sought to characterize how aptasensors with NWs of varying geometries responded to IFN-γ. 50,000 T-cells were placed into 0.5 mL solution of an E-cell. The cells were mitogenically stimulated to secrete cytokines and were maintained at physiological conditions (37 °C, 5 % CO2) for 2 h. SWV analysis was undertaken to characterize aptasensor responses to T-cell-secreted IFN-γ. The geometries tested included 6 μm long NWs catalyzed by 30 nm, 100 nm and 200 nm Au NPs. Due to different density of Au NPs in the respective colloidal solutions, density of obtained Si NWs varied from (11 ± 2) ×108 NWs/cm2 for 30 nm Au NPs to (24 ± 4) ×106 NWs/cm2 for 100 nm NPs and (4.94 ± ???) ×106 NWs/cm2 for 200 nm NPs. As result, spacing between the NWs varied from 300 nm to 10 μm for larger diameter NWs. Figure 3c highlights geometry dependent responses of aptasensors with 100 nm diameter wires providing highest change in the redox current (80% signal suppression) while 30 nm wires recorded only a 20% change in redox signal. Therefore, our results demonstrate that the geometry and density of 100 nm wires is most suitable for cell detection experiments. It is likely that spatial arrangement of this nanostructure substrate allowed for cells to land between the wires without impeding electron transfer.

We also characterized effects of NW length on sensor responses. In this case, NW diameter was fixed at 100 nm and length was varied from 6 pm to 60 μm to 160 μm. The differences NW aptasensor responses to secreted IFN-γ were not as dramatic as those seen for varying NW diameters (Figure 3b). These differences may be attributed to decrease in physical robustness of NWs as a function of NW length. We found a significant fraction of longer NWs was lying flat on the surface; hence, they were likely not fully capable of capturing secreted cytokine molecules.

3.5. Detection of IFN-γ release from immune cells

In the next set of experiments, we proceeded to characterize responses of nanowire aptasensors to cell-secreted IFN-γ. These experiments were carried out with NWs of 100 nm in diameter and 6 μm in length. CD4 T cells were first isolated from human blood, then re-suspended in RPMI at the desired concentration and dispensed into e-cells containing either planar or nanowire aptasensors. Once cells were inside an e-cell, mitogen (PMA/ionomycin) was added to trigger cytokine release from cells. Figure 4(a,b) summarizes cytokine secretion experiment with 50,000 cells residing in an e-cell (volume of 0.5 mL). SWV measurements were made every 10 min for the first 50 min and every 20 min after that. As seen from Figure 4b, NW aptasensors responded rapidly reaching 69 % signal suppression in the first 50 min of stimulation and saturating at 83 % after 90 min. The response of NW aptasensors was considerably faster than that of planar aptasensors. In the first 30 min of interaction with activated immune cells, response of nanowire aptasensor reached 70 % of the maximum whereas planar aptasensor was at ≈55 %. Because the change of redox signal due to analyte binding was a lot greater for nanowire aptasensors, it was possible to achieve 9 to 10 fold increase in signal after 30 min of interaction with cytokine producing cells. In comparison, signal suppression for planar aptasensors ranged from 5 % in the absence of IFN-γ to 35 % at saturation – thus only a 6 fold increase over the background was possible to this aptasensor. The fold difference over background at 30 min was ≈3.

Figure 4. Detecting IFN-γ release from leukocytes with NW aptasensors.

Figure 4.

Open in a new tab

a) Continuous monitoring of IFN-γ release from cells using NWs aptasensor. Cytokine are produced from activated PBMC cells, measurements were made for 90 min and the redox peak decreased in response to cytokine secretion. b) Changes in redox peak were converted into signal suppression and plotted against measurement time for 50K stimulated PBMC cell (black solid square) and 50K unstimulated PBMC cell control (black open square) on AuNW electrode, and 50K stimulated PBMC cell (red solid circle) and 50K unstimulated PBMC cell control (red open circle). c) Comparison of AuNWs aptasensor (dense packing) and flat aptasensor (blank) responses to different number of cells secretion. The cells were activated for 100 min for all different cell numbers and the signal suppressions were calculated by measuring redox peak decrease. The data indicates means ± SD (n=3). *p<0.05.

An important question addressed in this study was related to the lowest number of cells that can be detected with planar vs. nanowire aptasensors. This question is pertinent to clinical situations where CD4 T-cell counts are low and signal due to secreted IFN-γ is weak. One prominent clinical situation where such scenario is encountered is in patients co-infected with HIV and TB. Such co-infections are common in countries where both of these diseases are endemic (e.g. Sub-Saharan Africa). HIV/AIDS patients can have CD4 T-cell counts of 200 cell/μL or less [32, 33]. Assuming that TB-specific memory T-cells in such patients constitute between 0.1 % and 1 % of the total CD4 T-cell population, there will only be between 200 and 2000 IFN-γ producing CD4 T-cells per mL of patient blood. Therefore, we tested responses of planar and NW aptasensors to T-cells varying in concentration from 300 to 50000 cells per mL. Figure 4c shows that both planar and NW electrodes were sensitive in detecting IFN-γ release from very few T-cells. However, NW aptasensors showed responses to as few as 300 cells/mL, whereas planar aptasensors were not sensitive enough to detect secretions from such a low number of cells.

In light of improved sensitivity of NW aptasensors, we carried out additional control experiments to prove that electrochemical signals were indeed due to secreted IFN-γ. We have tested specificity of IFN-γ aptasensors extensively by challenges with nonsense proteins commonly present in serum (see Figure 3d and ref [30]), however, there remained a possibility that these experiments do not account for some unknown cell-secreted nonspecific signal capable of eliciting response of the sensor. As an additional strategy for validating specificity, we utilized chimeric anti-CD45/anti-IFN-γ Ab molecules. These commercially available chimeric Abs attach to pan leukocyte CD45 antigen and then capture secreted IFN-γ molecules. In this control experiment, T-cells collected from the same venipuncture were split into two samples: one sample was pretreated with chimeric antibodies whereas the other sample was not. Both samples were stimulated with mitogens. As shown in Figure 3 c, pretreatment of T-cells (bars describing signals from 50,000 cells) with chimeric antibodies resulted in aptasensor responses comparable to those for unactivated T-cells.

In another control experiment, we used ELISA measurements to verify that chimeric Abs interfered with IFN-γ but not another secreted cytokine, for example TNF-α. Here, T-cells were pre-incubated with anti-IFN-γ chimeric Abs then mitogenically stimulated as described above. Subsequently, media was collected and analyzed for both IFN-γ and TNF-α. As shown in Figure 5a, chimeric Abs diminished the concentration of IFN-γ available for ELISA easurements while the level of TNF-α was largely unaffected. These results once again highlight that aptasensors responded to IFN-γ with high specificity.

Figure 5. ELISA analysis of cytokine release from T-cells.

Figure 5.

Open in a new tab

a) Release of IFN-γ and TNF-α from 30,000 T-cells after mitogenic stimulation. Incubation with chimeric anti-IFN-γ Abs decreased IFN-γ signal detected by ELISA while TNF-α levels remained largely unaffected. This data demonstrated that chimeric Abs bound to IFN-γ and did not affect other secreted cytokines such as TNF-α. b) Varying concentrations of CD4 T-cells were stimulated in a 12-well plate and then analyzed by ELISA for IFN-γ production. Data represent n=3 samples.

To compare aptasensor performance to a standard method of IFN-γ detection, we carried out ELISA with T-cells of varying concentrations placed into a 24 well plate and stimulated with mitogen. After stimulation, cells were removed by centrifugation and remaining media was used to quantify secreted IFN-γ. The results in Figure 5b show that after 6 h mitogenic stimulation IFN-γ concentration ranged from 3 pg/mL for 5,000 cells to 90 pg/mL for 30,000 cells. Significantly, secretion from 2,500 cells was undetectable at 6 h time point but did become detectable after 24 h stimulation.

Why is our aptasensor more sensitive than ELISA? How to reconcile the fact that limit of detection for aptasensors ranges between 0.2 ng/mL and 1 ng/mL whereas according to ELISA results in Figure 7, INF-γ is present in low pg/mL concentration? To better understand our scenario, please consider Scheme 1. In our detection approach, immune cells reside in close proximity to the sensing surfaces such that high local concentration of secreted IFN-γ is detected at the electrode surface. In contrast, in an ELISA experiment, cells secrete into a large of volume of media which is then collected and placed into an ELISA plate to generate a read-out. Along the way, concentration gradients are disturbed, homogenized and averaged out. We estimate that our method records 10 to 100 fold higher concentration than ELISA because of co-location of cells right next to the sensors.

4. Conclusion

This study demonstrates utility of NW aptasensors for detection of cell-secreted cytokines. We demonstrate that Au-coated Si NWs result in biosensors that have lower limit of detection and much better signal-to-noise ratio compared to Au-coated planar Si aptasensors. NW aptasensors possess larger surface area per footprint, have higher aptamer packing density and are minimally affected by direct deposition of leukocytes. Importantly, these nanostructured aptasensors were more sensitive than planar aptasensors, capable of detection IFN-γ release from as few as 300 T-cells in mL of solution. IFN-γ release assays are used commonly for TB diagnosis. The ability to detect IFN-γ production in a blood sample containing few CD4 T-cells is critically important for TB patients with compromised immunity, for example, those co-infected with HIV. Therefore, NW aptasensors described here may be used in the future to address an important need for more sensitive diagnosis of HIV/TB co-infections.

Supplementary Material

Supp1

Acknowledgements

S.K. acknowledges support from the U.S. Department of Commerce, National Institute of Standards and Technology under the financial assistance award 70NANB16H043.

Footnotes

1

Certain commercial equipment, instruments, or materials are identified in this paper in order to specify the experimental procedure adequately. Such identification is not intended to imply recommendation or endorsement by the National Institute of Standards and Technology, nor is it intended to imply that the materials or equipment identified are necessarily the best available for the purpose.

References

  • 1.Gao ZQ, et al. , Silicon nanowire arrays for label-free detection of DNA. Analytical Chemistry, 2007. 79(9): p. 3291–3297. [DOI] [PubMed] [Google Scholar]
  • 2.Shao MW, et al. , Silicon nanowire sensors for bioanalytical applications: Glucose and hydrogen peroxide detection. Advanced Functional Materials, 2005. 15(9): p. 1478–1482. [Google Scholar]
  • 3.Qi S, et al. , Cell Adhesion and Spreading Behavior on Vertically Aligned Silicon Nanowire Arrays. Acs Applied Materials & Interfaces, 2009. 1(1): p. 30–34. [DOI] [PubMed] [Google Scholar]
  • 4.Yoon HJ, Kozminsky M, and Nagrath S, Emerging Role of Nanomaterials in Circulating Tumor Cell Isolation and Analysis. Acs Nano, 2014. 8(3): p. 1995–2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Kwon NH, et al. , Nanowire-based delivery of Escherichia coli O157 Shiga toxin 1 A subunit into human and bovine cells. Nano Letters, 2007. 7(9): p. 2718–2723. [DOI] [PubMed] [Google Scholar]
  • 6.Tian B and Lieber CM, Synthetic Nanoelectronic Probes for Biological Cells and Tissues. Annual Review of Analytical Chemistry, Vol 6, 2013. 6: p. 31–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Shalek AK, et al. , Vertical silicon nanowires as a universal platform for delivering biomolecules into living cells. Proceedings of the National Academy of Sciences of the United States of America, 2010. 107(5): p. 1870–1875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Shalek AK, et al. , Nanowire-Mediated Delivery Enables Functional Interrogation of Primary Immune Cells: Application to the Analysis of Chronic Lymphocytic Leukemia. Nano Letters, 2012. 12(12): p. 6498–6504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.VanDersarl JJ, Xu AM, and Melosh NA, Nanostraws for Direct Fluidic Intracellular Access. Nano Letters, 2012. 12(8): p. 3881–3886. [DOI] [PubMed] [Google Scholar]
  • 10.Timko BP, et al. , Design and Implementation of Functional Nanoelectronic Interfaces With Biomolecules, Cells, and Tissue Using Nanowire Device Arrays. leee Transactions on Nanotechnology, 2010. 9(3): p. 269–280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Duan X, et al. , Intracellular recordings of action potentials by an extracellular nanoscale field-effect transistor. Nature Nanotechnology, 2012. 7(3): p. 174–179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Gao R, et al. , Outside Looking In: Nanotube Transistor Intracellular Sensors. Nano Letters, 2012. 12(6): p. 3329–3333. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Lapierre MA, et al. , Electrocatalytic detection of pathogenic DNA sequences and antibiotic resistance markers. Analytical Chemistry, 2003. 75(22): p. 6327–6333. [DOI] [PubMed] [Google Scholar]
  • 14.Das J and Kelley SO, Tuning the Bacterial Detection Sensitivity of Nanostructured Microelectrodes. Analytical Chemistry, 2013. 85(15): p. 7333–7338. [DOI] [PubMed] [Google Scholar]
  • 15.Vasilyeva E, et al. , Direct Genetic Analysis of Ten Cancer Cells: Tuning Sensor Structure and Molecular Probe Design for Efficient mRNA Capture. Angewandte Chemie-lnternational Edition, 2011. 50(18): p. 4137–4141. [DOI] [PubMed] [Google Scholar]
  • 16.Matharu Z, Enomoto J, and Revzin A, Miniature Enzyme-Based Electrodes for Detection of Hydrogen Peroxide Release from Alcohol-Injured Hepatocytes. Analytical Chemistry, 2013. 85(2): p. 932–939. [DOI] [PubMed] [Google Scholar]
  • 17.Son KJ, et al. , Micropatterned Sensing Hydrogels Integrated with Reconfigurable Microfluidics for Detecting Protease Release from Cells. Analytical Chemistry, 2013. 85(24): p. 11893–11901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Matharu Z, et al. , Detecting Transforming Growth Factor-beta Release from Liver Cells Using an Aptasensor Integrated with Microfluidics. Analytical Chemistry, 2014. 86(17): p. 8865–8872. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Liu Y, et al. , Micropatterned Aptasensors for Continuous Monitoring of Cytokine Release from Human Leukocytes. Analytical Chemistry, 2011. 83(21): p. 8286–8292. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Liu Y, Zhou Q, and Revzin A, An aptasensor for electrochemical detection of tumor necrosis factor in human blood. Analyst, 2013. 138(15): p. 4321–4326. [DOI] [PubMed] [Google Scholar]
  • 21.Kwa T, et al. , Reconfigurable microfluidics with integrated aptasensors for monitoring intercellular communication. Lab on a Chip, 2014. 14(10): p. 1695–1704. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Ying L, et al. , Detecting multiple cell-secreted cytokines from the same aptamer-functionalized electrode. Biosensors and Bioelectronics, 2015. 64: p. 43–50. [DOI] [PubMed] [Google Scholar]
  • 23.Liu Y, Kwa T, and Revzin A, Simultaneous detection of cell-secreted TNF-alpha, and IFN-gamma using micropatterned aptamer-modified electrodes. Biomaterials, 2012. 33(30): p. 7347–7355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Xiao Y, Lai RY, and Plaxco KW, Preparation of electrode-immobilized, redox-modified oligonucleotides for electrochemical DNA and aptamer-based sensing. Nature Protocols, 2007. 2(11): p. 2875–2880. [DOI] [PubMed] [Google Scholar]
  • 25.Ferguson BS, et al. , Real-Time, Aptamer-Based Tracking of Circulating Therapeutic Agents in Living Animals. Science Translational Medicine, 2013. 5(213). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Wu C, et al. , A survey of advancements in nucleic acid-based logic gates and computing for applications in biotechnology and biomedicine. Chemical Communications, 2015. 51(18): p. 3723–3734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Krylyuk S, Davydov AV, and Levin I, Tapering Control of Si Nanowires Grown from SiCl4 at Reduced Pressure. Acs Nano, 2011. 5(1): p. 656–664. [DOI] [PubMed] [Google Scholar]
  • 28.Ricci F, et al. , Effect of molecular crowding on the response of an electrochemical DNA sensor. Langmuir, 2007. 23(12): p. 6827–6834. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.White RJ, et al. , Optimization of electrochemical aptamer-based sensors via optimization of probe packing density and surface chemistry. Langmuir, 2008. 24(18): p. 10513–10518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Liu Y, et al. , Aptamer-Based Electrochemical Biosensor for Interferon Gamma Detection. Analytical Chemistry, 2010. 82(19): p. 8131–8136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Hill HD, et al. , The Role Radius of Curvature Plays in Thiolated Oligonucleotide Loading on Gold Nanoparticles. ACS Nano, 2009. 3(2): p. 418–424. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Getahun H, et al. , HIV Infection-Associated Tuberculosis: The Epidemiology and the Response. Clinical Infectious Diseases, 2010. 50: p. S201–S207. [DOI] [PubMed] [Google Scholar]
  • 33.Pawlowski A, et al. , Tuberculosis and HIV Co-Infection. Plos Pathogens, 2012. 8(2). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supp1
NIHMS1588424-supplement-Supp1.pdf (544.7KB, pdf)

RESOURCES