key: cord-1019733-u3rzoiab
authors: Abrego-Martinez, Juan Carlos; Jafari, Maziar; Chergui, Siham; Pavel, Catalin; Che, Diping; Siaj, Mohamed
title: Aptamer-based electrochemical biosensor for rapid detection of SARS-CoV-2: Nanoscale electrode-aptamer-SARS-CoV-2 imaging by photo-induced force microscopy
date: 2021-08-30
journal: Biosens Bioelectron
DOI: 10.1016/j.bios.2021.113595
sha: 045da9a38b964323a7d1937a149397da66781d56
doc_id: 1019733
cord_uid: u3rzoiab

Rapid, mass diagnosis of the coronavirus disease 2019 (COVID-19) is critical to stop the ongoing infection spread. The two standard screening methods to confirm the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are polymerase chain reaction (PCR), through the RNA of the virus, and serology by detecting antibodies produced as a response to the viral infection. However, given the detection complexity, cost and relatively long analysis times of these techniques, novel technologies are urgently needed. Here, we report an aptamer-based biosensor developed on a screen-printed carbon electrode platform for rapid, sensitive, and user-friendly detection of SARS-CoV-2. The aptasensor relies on an aptamer targeting the receptor-binding domain (RBD) in the spike protein (S-protein) of the SARS-CoV-2. The aptamer immobilization on gold nanoparticles, and the presence of S-protein in the aptamer-target complex, investigated for the first time by photo-induced force microscopy mapping between 770 and 1910 cm(-1) of the electromagnetic spectrum, revealed abundant S-protein homogeneously distributed on the sensing probe. The detection of SARS-CoV-2 S-protein was achieved by electrochemical impedance spectroscopy after 40 min incubation with several analyte concentrations, yielding a limit of detection of 1.30 pM (66 pg/mL). Moreover, the aptasensor was successfully applied for the detection of a SARS-CoV-2 pseudovirus, thus suggesting it is a promising tool for the diagnosis of COVID-19.

During the early stages of the COVID-19 outbreak, infected patients were correctly diagnosed through deoxyribonucleic acid (DNA) sequencing assays (Ji et al. 2020) , nevertheless, the widespread use of this technology is not feasible, since it is expensive, time-consuming, and it must be performed under rigorous laboratory conditions by highly trained technicians. Reverse transcription polymerase chain reaction (RT-PCR) and serological tests (detection of antibodies)

are the standard COVID-19 diagnostic techniques (Ji et al. 2020; Udugama et al. 2020) ; however, these assays are not without limitations. RT-PCR yields a relatively high rate of false-negative results, especially in the early stages of infection due to the low levels of SARS-CoV-2 RNA and the low detection sensitivity of the method (Ji et al. 2020; Tahamtan and Ardebili 2020) . In addition, several hours are needed to obtain a result (Long et al. 2020 ) and, although the cost of PCR assays is lower than DNA sequencing technology, it is relatively expensive for mass testing in most countries. On the other hand, serological tests are not meaningful for early diagnosis of COVID-19 due to the lengthy delay between infection and seroconversion (De Assis et al. 2021) .

Therefore, antibody tests are more reliable to identify past infection. Meanwhile, countless efforts pursued optimizing detection and diagnosis methods at point-of-care (POC) facilities and as selfdetection kits, such as Lateral Flow Immunochromatographic Assay Strips (LFICS) for COVID-19 (Alpdagtas et al. 2020; Yüce et al. 2021) . Unlike gene-targeting tests, the sensitivity of rapid immunochromatographic test (ICT) kits is dependent on the time past infection (Fujigaki et al. 2020 ). Mandatorily, 10-15 days past infection are essential to provide enough time for the viral immune-response to produce a measurable amount of IgM and IgG SARS-CoV-2 recognizing antibodies (Fujigaki et al. 2020) . Thus, ICT assays are biased by delay time and can only provide a reliable diagnosis after a controlled environment quarantine period, although they are showing an increasing role as screening method due to their ease of use (Pegoraro et al. 2021) .

Alternatively, biosensors could potentially be employed as diagnostic tools for COVID-19, since in general, they offer several advantages over conventional detection methods, such as rapid, selective, and low-cost detection over a wide range of analytes, including viruses (Bahadır and Sezgintürk 2016; Mehrotra 2016) . Several groups have recently reported the development of biosensors for COVID-19, including devices targeting the RNA of SARS-CoV-2 (Zhao et al. 2021) , and COVID-19 antibodies (Ali et al. 2021; Elledge et al. 2021; Yakoh et al. 2021; Zeng et al. 2020) , as well as immunosensors targeting the spike and nucleocapsid proteins (Cerutti et al. 2020; Eissa and Zourob 2021; Mavrikou et al. 2020) . Nonetheless, short strands of oligonucleotides, known as aptamers, could also be employed as biorecognition elements for COVID-19 diagnosis, as they are able to recognize targets with high selectivity (Darmostuk et al. 2015) and possess advantages compared with immunosensors in terms of detection capabilities.

J o u r n a l P r e -p r o o f For instance, aptamers are subjected to less steric hindrance on the surface of CoVs (65-125 nm in diameter) (Shereen et al. 2020) due to their smaller size, typically about 2-3 nm in diameter (30-60 nucleotides) in comparison with antibodies (12-15 nm in diameter) . This, in theory, allows the binding of more recognition elements on the surface of the CoV, which leads to enhanced sensing performances. In addition, aptamers are stable in a wider range of temperature and pH and have lower production costs than antibodies (Darmostuk et al. 2015; Tombelli et al. 2005) . For these reasons, we propose the use of an aptamer-based electrochemical biosensor for detection of SARS-CoV-2 S-protein. Song et al. recently identified two single-stranded (ss) DNA aptamer sequences with the ability to bind to the RBD of the SARS-CoV-2 . Therefore, in this work, we employed one of those reported sequences to develop an aptamer-based sensing platform for impedimetric detection of SARS-CoV-2 Sprotein. The immobilization of the ssDNA aptamer on gold nanoparticles (AuNPs) and the binding of S-protein with the aptamer were investigated through photo-induced force microscopy (PiFM) and electrochemical techniques. The resulting device exhibited excellent sensitivity and selectivity towards the SARS-CoV-2 S-protein and towards a SARS-CoV-2 pseudovirus, thus being a promising tool for COVID-19 detection.

The ssDNA aptamer sequence used in this work is a 51-nt 3-hairpin-structured aptamer targeting the SARS-CoV-2 RBD (Kd=5.8 nM), recently reported by Song et al Sigma-Aldrich. All reagents were analytical grade and were used without further processing.

Prior to the fabrication of the aptasensor, the duration of the potential pulse for AuNPs deposition, aptamer incubation time and target incubation time (Fig. S1 , S2 and S3, respectively) were investigated using a glassy carbon electrode (GCE, BASi, 0.07 cm 2 area) in a typical 3-electrode cell with an Ag/AgCl (3 M NaCl) reference electrode (RE) and a Pt wire counter electrode (CE).

The final sensing platform was therefore constructed with the optimum parameters for Au (Heiskanen et al. 2008) which is well known to yield a highly clean Au surface (Fischer et al. 2009; Ho et al. 2019) . Before proceeding to aptamer immobilization, 1.5 µM of disulfide-labeled ssDNA solution in binding buffer (BB, 50 mM Tris-HCl + 150 mM NaCl + 2 mM MgCl2, pH= 7.5) was subjected to heating at 90 °C for 5 min, followed by cooling at -4 °C for 10 min for DNA renaturation. Finally, the solution was allowed to stand at room temperature for 5 min. Afterwards, 5 µL of aptamer solution was drop-casted onto the surface of the WE to immobilize the aptamer on the AuNPs via a self-assembled monolayer (SAM) (Oberhaus et al. 2020) , which was spontaneously formed during incubation at 4 °C for 8 h. The aptamer-modified electrode was then gently rinsed with BB to remove non-bound aptamer. The Aptamer/AuNPs/SPCE was subsequently incubated with 1 mM of 6-mercapto-1-hexanol (MCH) + 10 mM PBS solution for 30 min at room temperature to displace the non-specific adsorption of the aptamers and to block the non-modified Au area to avoid pinhole formation. A final rinsing step with BB was performed to remove non-bound J o u r n a l P r e -p r o o f molecules, thus obtaining the aptamer-based sensing probe. The development process and working principle of the aptasensor are illustrated in Scheme 1. A photograph of the fabricated device is shown in Fig. S4 .

Stepwise fabrication of aptasensor for SARS-CoV-2 S-protein detection.

Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) measurements were performed throughout every step of the sensor fabrication to follow the surface modification.

Cyclic voltammograms were obtained in the potential range of [-0.4 ─ 0.7 V] at 50 mV s -1 scan rate, while EIS was recorded at a direct current (DC) potential of 0.11 V in the frequency range from 100 kHz to 100 mHz with a sinusoidal voltage perturbation of 10 mV amplitude, both acquired in 30 µL of PBS (pH = 7.4) solution containing 5 mM [Fe(CN)6] 3-/4as redox probe. All electrochemical measurements were performed on a SP-300 potentiostat/galvanostat (Bio-Logic Science Instruments) controlled by EC-Lab® software, which was also employed for the fitting of the EIS data.

The morphology of AuNPs was studied by SEM with a JEOL JCM-6000 tabletop microscope operated at 15 kV.

Atomic force microscopy (AFM) and photo-induced force microscopy (PiFM) images of the modified electrode were captured with a VistaScope instrument (MolecularVista, San-Jose, USA).

diameter of ~10 nm. The microscope was coupled to a mid-infrared radiation source comprised of two quantum cascade lasers (QCLs), together spanning the 770 -1910 cm -1 spectral range.

The focused laser polarization was at 45°, partly P-and partly S-polarized. The P-polarized component dominated the PiFM signal. The elliptical irradiation spot size was roughly λ by 1.5 λ on the short and long axes of the ellipse respectively. The laser intensity was 0.1 -1 W and the depth of the laser into the sample was 20 nm. 10-point PiF-IR line-spectra were recorded on 2 × 2 µm 2 topography scans of the AuNPs-modified SPCE, aptamer/AuNPs/SPCE and Sprotein/Aptamer/AuNPs/SPCE at a rate of 30 seconds/spectrum, with ≤ 10 nm lateral surface resolution and with 0.5 cm -1 spectral resolution. The spectra were collected at a baseline intensity The selectivity of the aptasensor was investigated using Spike proteins from SARS-CoV, MERS-CoV and SARS-CoV-2. The probe was separately incubated during 40 min with 10 nM of these analytes and the impedimetric response was then recorded in redox probe/PBS.

To assess the potential application of the aptasensor as a COVID-19 detection method, the device was incubated with 5 µL of a SARS-CoV-2 pseudovirus, kindly provided by Prof. Benoit Barbeau, Université du Québec à Montréal. The SARS-CoV-2 pseudovirus was produced by transfection of HEK293T cells with an Env-deficient HIVNL4-3 plasmid and a plasmid expressing SARS-CoV-2

Spike. Experiments with the HIVNL4-3 plasmid (no spike) were also performed as control. The concentration of the HIVNL4-3Env-luc plasmid and the SARS-CoV-2 pseudovirus was 12294 cpm/µL, as measured by scintillation counter. increase of cathodic current starting at ≈0.75 V with a peak at 0.47 V, which is ascribed to the reduction of Au(III) to Au(0), according to (1):

In the backward scan (anodic direction), a current crossover is observed at 0.65 V, which is a typical feature of nucleation on the electrode surface (Grujicic and Pesic 2002; Inamdar et al. 2007 ). However, during the second cathodic scan (solid line), the Au(III) reduction peak shifted towards a less cathodic potential (0.7 V), indicating that the reduction of Au(III) occurs on the Au previously deposited. Therefore, under these conditions, electrodeposition of Au on the NPs deposited during the first scan is thermodynamically more favorable than nucleation of new AuNPs on the carbon electrode. Based on this CV, a potential of -0.3 V was applied to the working electrode through chronoamperometry for electrodeposition of AuNPs, since at this value the overpotential is large enough to favour the nucleation of new sites on the C electrode rather than the growth of previously established nuclei (Hezard et al. 2012 ). The potential pulse was applied during 840 s, according to the optimization reported in Fig. S1 . The current transient presented in Fig. 1b exhibits a typical high current at very short times due to the charging of the double layer.

Formation of initial nuclei also occurred at this early stage. Then, a current decay is observed until steady state is achieved, which corresponds to overlapping of the diffusion zones defined by each nucleus, as described by the Cottrell's equation (Grujicic and Pesic 2002) . (2): 

The development of the sensing device was followed up by CV and EIS. Fig. 2a and 2b show the cyclic voltammograms and Nyquist plots, respectively, obtained after every step of the aptasensor fabrication process. The Randles equivalent circuit used to fit the impedance data is displayed in the inset of Fig. 2c and 2e ). This profile is consistent with the height of the spike protein of the SARS-CoV and SARS-CoV-2 viruses (Neuman et al. 2006; Zhu et al. 2020) . Given that the samples exhibited similar features in terms of NPs size and shape suggests that no major topography changes occurred as the electrode underwent modification. Figure 2d shows the PiF-IR spectra (green and blue) of Aptamer/AuNPs-and S-protein/Aptamer/AuNPs-modified electrodes, respectively. In the 1500-1800 cm -1 region, known for the in-plane vibrations and double bond (C=C, C=N and C=O) stretching of the nucleic acid moieties, it was found that the initially observed ssDNA aptamer signal at ~1680 cm -1 experienced a hypsochromic shift to ~1730 cm -1 upon complexation with the viral spike protein (Wood 2016) . Stabilization of the ss nucleic acids when binding to the S-protein accounts for higher energy light absorption needed to promote the C=X bonds to energetically increasing vibrational states. On top of the sum of simultaneous J o u r n a l P r e -p r o o f specific non-covalent ligand-receptor interactions, being electron-rich molecules, nucleic acids are influenced by neighbouring π-π interactions that occur between aromatic residues of the Sprotein and the ssDNA aptamer. This observation is fortified by the noticeable localized region of high intensity in the north-east corner of the PiFM map when scanning the Sprotein/Aptamer/AuNPs modified SPCE at ~1730 cm -1 , highlighting an area of greater binding between ligand and receptor, in comparison to the rest of the scanned frame. Moreover, the uneven signal observed in the latter scan contrasts with the more uniformly distributed signal on the Aptamer/AuNPs-modified SPCE scanned at ~1680 cm -1 (Fig. 2e) , evidencing a change in the chemistry of the ssDNA aptamer once the S-protein was bound. At this same wavenumber, no signal was observed on the bare AuNPs modified SPCE, in accordance with its acquired PiF-IR representative spectra (Fig. S6) . As for the peak located in the ~1050 cm -1 , originating from the phosphate-backbone symmetric stretch of the ssDNA aptamer (Polito et al. 2021) , no peak shift occurred. Figure AuNPs electrode, the PiFM signal at ~1400 cm -1 detected from the Aptamer/AuNPs sample arises from the DNA sugar-backbone/base bending modes, sensitive to glycosidic torsion angles (Parker and Quinn 2013) . Indeed, following incubation with S-protein, PiFM mapping showed that the signal intensified, densified, and appeared on larger zones of the sampled area (Fig. 2c) . Peptide bonds, comprised of amide groups, resonate in the ca.1400 cm -1 region of the spectrum through the amide III vibration modes (Mallamace et al. 2015) . Moreover, it was found that the PiFM mapping signal corresponded accurately to the morphology of the AuNPs-modified electrode, thus revealing abundant S-protein present in the aptamer-target complex homogeneously distributed on the sensing probe. Hence, the PiFM mapping images at ~1050 cm -1 , arising from the PO2 ssDNA aptamer backbone vibrations, are consistent with this explanation (Fig. 2e) . A lesser change is perceived between the Aptamer/AuNPs-and S-protein/Aptamer/AuNPsmodified SPCEs signal intensity distributions at this wavenumber. The subtle difference in the PiFM mapping images is perceived exclusively at the AuNP contours. A slight intensification in signal distribution on the AuNPs perimeters post-incubation with the S-protein, without any PiF-IR spectral peak shifts, confers a non-specific binding role to the PO2 aptamer backbone toward J o u r n a l P r e -p r o o f groups by electrostatic forces, strengthening the ligand-receptor complex stability. In summary, as observed in Fig. 2c , the PiFM signal intensity distributions are null, uniform and locally elevated at ~1400 cm -1 due to negligeable resonance on the bare AuNPs, to medium resonance of the glycosidic backbone in the Aptamer/AuNPs-and to perceivable amide III resonance on the Sprotein/Aptamer/AuNPs-modified SPCEs, respectively. The PiFM scans at ~1390 cm -1 in Fig. 2e concur the latter observations. The ssDNA aptamer's nucleic acid resonance peak at ~1680 cm -1 undergoes a blue shift to ~1730 cm -1 after complexation with the S-protein, explained by ligandreceptor complex stabilization and electronic interactions. Finally, the detected aptamer PO2backbone symmetric stretching peak at ~1050 cm -1 remains invariable after the incubation with the S-protein, implying it contributes to non-specific electrostatic stabilization of the aptamer─Sprotein complex.

J o u r n a l P r e -p r o o f According to the PiF-IR spectra (Fig.2d) and targeting selected peak shifts for best chemical contrast, the common wavenumbers at (~1050, ~1390 cm -1 and 1680-1730 cm −1 )

were selected for PiFM mapping. d) Representative PiF-IR spectra acquired on the Aptamer/AuNPs-and S-protein/Aptamer/AuNPs-modified SPCEs, respectively in blue and green.

Highlighted regions are mutual resonance peaks observed in both samples e) AFM and PiFM 500

x 500 nm 2 micrographs of the AuNPs-, Aptamer/AuNPs-and S-protein/Aptamer/AuNPs-modified SPCEs at the highlighted region wavenumbers in d), ~1050, ~1390 and ~1680-1730 cm -1 .

The analytical performance of the aptasensor was investigated through EIS by measuring the 

Given the similarity between SARS-CoV, MERS-CoV and SARS-CoV-2 Spike proteins, the aptasensor was separately incubated in 10 nM of these analytes for 40 min and the impedimetric response was recorded to study the selectivity towards SARS-CoV-2 S-protein. The percentage change in charge transfer resistance relative to BB is displayed in Fig. 4a . While the response of the aptasensor towards MERS S-protein was only 7%, the device exhibited relatively high activity towards the SARS S-protein (%ΔRct ═ 26%). This value represents more than half the response obtained with the SARS-CoV-2 S-protein (43%). This is not surprising, since the S-proteins of SARS-CoV and SARS-CoV-2 share a sequence identity of 77% (Hassanzadeh et al. 2020 ).

However, despite the high sequence and structural similarity, the SARS-CoV-2 S-protein is slightly more positively charged than the SARS-CoV S-protein, which is one of the reasons why the former exhibits greater affinity to the ACE2 receptor that contributes to its transmission efficiency (Hassanzadeh et al. 2020) . This difference in charge is likely to be the main differentiator in the behavior of the aptasensor towards SARS-CoV-2 and SARS-CoV S-proteins, which causes the former to be detected with more sensitivity. Since the SARS is considered eradicated in humans (Smith 2019), a false positive result caused by SARS-CoV is unlikely.

To assess the stability of the aptamer probe, the impedimetric response of the aptasensor towards 50 nM SARS-CoV-2 S-protein was recorded after up to 3 weeks storage in BB at 4 °C and contrasted with the response obtained with the freshly fabricated device. The EIS J o u r n a l P r e -p r o o f measurements shown in Fig. 4b revealed that the Rct increased from 1855 to 1872 Ω after 21 days, which represents a sensing activity loss of only 1% with respect to the fresh device. This result demonstrates that the aptamer probe is remarkably stable under the mentioned storage conditions, and it suggests that the aptasensor is capable of reliable detection up to 3 weeks after device fabrication. 

To investigate the feasibility of practical application for COVID-19 detection, the aptasensor was tested with a pseudovirus consisting of HIVNL4-3Env-luc + SARS-CoV-2 S-protein. Experiments with a spike-deficient NL4-3 plasmid were also performed as control. The Nyquist plots and the relative response (%) are shown in Fig. 5 . The fitting of the EIS data for the control measurement revealed a slightly higher Rct in comparison with the response towards BB (no analyte). Since several different biomolecules are present in the HIVNL4-3Env-luc sample, the observed response could be attributed to nonspecific binding of some of these species with the aptamer probe. On the other hand, after incubation with the SARS-CoV-2 pseudovirus, a lower Rct was obtained with respect to the response towards BB. As shown in Fig. 3a , the presence of analyte in the aptamer probe induces a decrease in Rct, therefore, the behavior of the aptasensor with the pseudovirus is consistent with the results obtained with purified S-protein, which indicates that the aptamer effectively captured the SARS-CoV-2 pseudovirus.

J o u r n a l P r e -p r o o f the ICT tests being consistent only during the second week after infection (Pegoraro et al. 2021) are substantial barriers that setback a successful deconfinement and delay the end of the pandemic. On the other hand, the electrochemical aptasensor developed in this work was fabricated on a miniaturized, low-cost platform using widely available techniques and equipment, and it offers the possibility of early COVID-19 diagnosis by detecting the S-protein present in the SARS-CoV-2. This approach makes it possible to reproduce it in practically any bioelectrochemistry laboratory and even provides the feasibility of mass production in adequate facilities, although admittedly, the next phase (clinical testing) with groups of patients potentially infected with SARS-CoV-2 is necessary. In addition, the aptasensor can be used directly in a miniaturized, low-cost potentiostat coupled to a smartphone (Ainla et al. 2018) , which adds to the user-friendly characteristics of this biosensor. The use of smartphones is an encouraging horizon J o u r n a l P r e -p r o o f for the miniaturized detection of SARS-CoV-2. Coupling the present biosensor with a smartphone would enable real-time fast detection of SARS-CoV-2 S-protein directly on the smartphone's screen, thus being a promising tool to complement or even replace existing SARS-CoV-19 detection methods.

In the present work, an aptamer-based impedimetric biosensor for detection of SARS-CoV-2 Sprotein was successfully developed using a AuNPs-modified SPCE platform and an aptamer targeting the RBD of the SARS-CoV-2. The physicochemical characterization confirmed that the disulfide-modified aptamer was homogeneously immobilized on the surface of the AuNPs, which allowed the probe to capture S-protein atop practically all the AuNPs-modified surface, as revealed by the PiFM mapping imaging in the 1400 cm -1 region. The aptasensor demonstrated excellent sensing performance, as it required an analyte incubation time of 40 min to obtain a reliable reading, which is faster than the standard diagnostic tests, and it exhibited a LOD of 1.30 pM (66 pg/mL) for SARS-CoV-2 S-protein, lower than the LOD of PCR for the same analyte. The selectivity studies showed that the aptasensor is active to both SARS-CoV and SARS-CoV-2 Sproteins, however, the latter is more easily detected due to its more positively charged nature.

Moreover, the aptasensor demonstrated consistent sensing activity when tested with a SARS-CoV-2 pseudovirus, which reaffirms the feasibility of practical application for COVID-19 detection.

As a final remark, our biosensor was developed with practicality in mind, given the urgency for portable and fast detection methods. The use of screen-printed electrodes with a proven aptamer allows a reliable and straightforward application, bringing up the possibility of using a hand-held potentiostat connected to a smartphone. By implementing a straightforward fabrication method, our device has a greater potential to be reproduced in most facilities, which is favorable in case of an eventual mass production. 

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Infrared Spectroscopy of DNA

We thank the Natural Sciences and Engineering Research Council of Canada (NSERC), the Canada Research Chairs program (CRC), Canada Foundation for Innovation (CFI) and NSERC-Alliance Covid-19 program. NanoQAM center at UQAM is gratefully acknowledged for all the characterization experiments. We thank Quebec Centre for Advanced Materials (QCAM).J o u r n a l P r e -p r o o f