In situ plant virus nucleic acid isothermal amplification detection on gold nanoparticle-modified electrodes

Solid-phase isothermal recombinase polymerase amplification (RPA) offers many benefits over the standard RPA in homogeneous phase in terms of sensitivity, portability, and versatility. However, RPA devices reported to date are limited by the need for heating sources to reach sensitive detection. With the aim of overcoming such limitation, we propose here a label-free highly integrated in situ RPA amplification/detection approach at room temperature that takes advantage of the high sensitivity offered by gold nanoparticle (AuNP)-modified sensing substrates and electrochemical impedance spectroscopic (EIS) detection. Plant disease ( Citrus tristeza virus (CTV)) diagnostics was selected as a relevant target for demonstration of the proof-of-concept. RPA assay for amplification of the P20 gene (387-bp) characteristic of CTV was first designed/optimized and tested by standard gel electrophoresis analysis. The optimized RPA conditions were then transferred to the AuNP-modified electrode surface, previously modified with a thiolated forward primer. The in situ-amplified CTV target was investigated by EIS in a Fe(CN6)4-/Fe(CN6)3- red-ox system, being able to quantitatively detect 1000 fg μL-1 of nucleic acid. High selectivity against nonspecific gene sequences characteristic of potential interfering species such as Citrus psorosis virus (CPsV) and Citrus caxicia viroid (CCaV) was demonstrated. Good reproducibility (RSD of 8%) and long-term stability (up to 3 weeks) of the system were also obtained. Overall, with regard to sensitivity, cost, and portability, our approach exhibits better performance than RPA in homogeneous phase, also without the need of heating sources required in other solid-phase approaches.


INTRODUCTION
Recombinase polymerase amplification (RPA) has received much attention in recent years due to its versatility and isothermal performance, which may offer effective replacement to Polymerase chain reaction (PCR) in molecular detection. [1][2][3][4] Unlike PCR, RPA technology is based on three main components (recombinase, single-stranded DNA-binding protein and polymerase) which allow amplification at constant temperature (ranging from 37°C -40°C) and as a result the need for thermal cycling is omitted. 5,6 The required temperature for RPA is lower than for other emerged isothermal amplification methods such as nucleic acid sequence-based amplification (NASBA), loop-mediated isothermal amplification (LAMP), rolling circle amplification (RCA) and helicase dependent amplification (HDA). [7][8][9][10][11] In spite of the great advantages of RPA, DNA purification and detection after amplification involving hazardous, time-consuming and expensive equipment is still required for getting qualitative information. Alternative methodologies taking advantage of the use of labeled primers for detecting RPA amplified products in lateral flow [12][13][14] , and electrochemical approaches have been proposed for such purpose. 15,16 Some efforts have been also devoted to RPA integration into microfluidic devices with fluorescence detection. 17 However, these methods lack of integration of RPA amplification and detection in the same device, which would be strongly needed for in-field diagnostic applications.
Such highly integrated devices have been recently achieved with the so-called solid-phase RPA amplification, in which one of the primers is directly immobilized on the sensing surface. Label-free optical approaches 18,19 , and enzymatic label-based electrochemical ones with high degree of integration have been reported. 20 However, heat sources are needed for getting detectable signals, which represents an important practical limitation.
In this context, we propose a label-free highly integrated in situ RPA amplification/detection approach at room temperature, taking advantage of the high sensitivity achieved combining the use of gold nanoparticle (AuNP)-modified sensing substrates and electrochemical impedance spectroscopic (EIS) detection. Plant disease (Citrus tristeza virus (CTV)) is selected as relevant target for the demonstration of the proof-of-concept. Recent studies estimating that plant diseases may cause global economic losses exceeding billions of dollars annually put in value the relevance of such diseases. 21 Early detection of pathogens in presymptomatic plants is of key importance for avoiding the development and spread of the disease. Plant pathogen determination is currently performed through antibody-based enzyme-linked immunosorbent assays (ELISA) and lateral flow immunoassays (LFIA) [22][23][24][25][26][27][28] ,and nucleic acid-based analysis [29][30][31] , as alternative to the traditional diagnostic methods including symptoms observation, regular in-field inspections, and laboratory analysis by experienced plant pathologists . 32 Over the last decade, it has been noticed that antibody and DNA based biosensing applications in the field of plant disease diagnostics have been increasing . [33][34][35][36] However, highly integrated approaches for in-field detection in presymptomatic plants, like the one we are proposing in this work, are still missing.

RPA assay of CTV related nucleic acid
The sequence of p20 gene (549nt) responsible for systemic infection was first selected in CTV genome. Such sequence is specific to CTV and does not relate to other closteridea viruses, major component of CTV and highly produced in infected trees. Three forward and two reverse primers were designed (between 25 and 35 bp) to amplify the p20 of 378-bp of CTV genomic nucleic acid, following the RPA manufacture's manual. Then the primer combinations were screened by gel electrophoresis in order to select the optimal primer pair which has great sensitivity and specificity for CTV.
For the RPA assays (50 μL reaction volume), a master mix composed of 2.4 μL of primers (10 μM), 29.5 μL of rehydration buffer and 13.8 μL of CTV-p20 gene and DNase-free water was first prepared. After dividing aliquots of the master mix into reaction tubes and mixing with TwistAmp basic Freeze-dried enzyme pellets, the RPA reactions were started immediately by adding magnesium acetate (280 mM). The reaction tubes were incubated at 37°C for 30 min. RPA reactions were performed without the target gene as no template control (NTC). Following post-amplification purification, amplicons were analysed in 2% agarose gel.

In situ isothermal RPA on gold nanoparticle modified electrodes
SPCEs modification with AuNPs and thiolated nucleic acid immobilization were performed following a previously optimized procedure (Khater et al., 2017). Briefly, the electrodes were pre-treated by applying oxidative potentials of +1.6 V for 120 s and of +1.8 V for 60 s in acetate buffer, followed by rinsing with PBS and Milli-Q water (3X) and dried using nitrogen.
Carbon working electrodes were then immersed into a gold solution (0.01% HAuCl4, / 0.1 M NaCl in the presence of 1.5 wt% HCl) and a constant negative potential of -0.4 V for 200 s was applied for achieving a homogenous formation of well distributed spherical AuNPs of 50 nm. The thiolated forward primer (SH-(AT7)-F1) was pre-reduced using TCEP as detailed at the supplementary information. AuNP-modified SPCEs were then incubated with 15 µL of 0.1µM SH-(AT7)-F1 prepared with 1 mM MCH solution at ratio (1:0.1) for 2 h at room temperature. After that, the electrodes were thoroughly rinsed using PBS and dried with nitrogen gas.
RPA solutions were then prepared for the surface amplification and detection of the target sequence of (P20 gene) on the AuNP-modified SPCEs. For master mix of RPA reaction preparation, 29.5 μL of rehydration buffer, 2.4 μL of reverse primer (10 μM) and 13.2 μL of DNase-free water were mixed and added to the freeze-dried enzyme pellet. Then, this mixture was mixed well with 2.5 μL of magnesium acetate at concentration of 280 mM. After that, the 50 μL reaction volume was divided into 10 μL aliquots and added to AuNP-modified SPCEs.
Finally, 5 μL of the P20 gene from the target plant virus genome that ranges in concentrations from 1 pg μL -1 to 1 ng μL -1 was added to the previously prepared 10 μL RPA solution. The solid phase isothermal amplification assays were performed at room temperature (25 ±3°C) for 60 min. Additionally, the AuNP-modified SPCEs with RPA solutions containing water or other unrelated DNAs as negative controls were evaluated. A scheme of the developed nucleic acid amplification/detection system is shown at Figure 1. impedance for the determination of CTV-related nucleic acid.

Electrochemical measurements
The in situ isothermal RPA was characterized by electrochemical impedance spectroscopy All EIS reported results were analyzed by calculating means and standard deviation to represent obtained data. All impedance experiments were conducted under ambient conditions.

Design of RPA for CTV detection assay
After designing of RPA primers to specifically detect CTV, gel electrophoresis was carried out to screen the selected primer sets (Table S1,

Electrochemical characterization and optimization of the in situ isothermal RPA on the electrode surface
The optimized RPA conditions were transferred to the SPCE-AuNP electrode surface. All the reagent solutions were added to the electrode, previously modified with the forward primer (SH-(AT7)-F1). CTV-p20 gene was lastly added to the RPA mixture on the electrode to start the reaction (Figure 1).
The Nyquist plots recorded for the electrodes showed a high increase in the electrode resistance upon the RPA performed with a CTV-p20 target concentration of 1 ng μL -1 ( Figure   3A), demonstrating the attachment of amplified DNA to the electrode.
However, when testing RPA solutions with negative control, unspecific EIS response was observed (data not shown). In order to prevent the interference from RPA chemical reagents and to reduce unspecific adsorption of enzymes, the three most commonly used washing Finally, the effect of the time of the in situ isothermal RPA reaction on the electrochemical signal was also studied. As shown in Figure 3C, no detectable signals were obtained for reaction times below 30 min. Rct values increased then with the reaction time in the range 30 -14 amplification time in solid-phase is longer than the one typically used for the RPA in homogeneous phase (60 min), as expected. 20 Furthermore, the lower temperature used probably also contributes to such longer time required.

In situ Citrus tristeza-related nucleic acid DNA amplification/detection
The ability of our biosensor for the quantification of CTV-related DNA was evaluated. The Other experimental conditions as described in the text.
The limit of detection achieved by the label-free in situ isothermal RPA on primer-modified SPCE-AuNP electrodes was 100 times lower than that of the modified RPA assay, showing that the integration of in situ isothermal RPA with impedance approach allowed a highly sensitive detection of nucleic acid.
Additionally, specificity studies were performed to examine the ability of our developed in situ RPA sensor to differentiate between CTV-related and non-CTV-related DNAs. These studies were performed with a non-specific gene sequences characteristic of Citrus psorosis virus (CPsV), as it is claimed to be the second important citrus virus, and also Citrus caxicia viroid (CCaV), which is likely to be present in complex infection with CTV in the infected citrus trees in nature. Under optimized experiment conditions, primer-modified SPCE-AuNP electrodes were coated with RPA solutions prepared with CTV-p20 gene target and controls.
Impedance measurements shown in Figure 5A evidence a clear discrimination between the specific and non-specific DNA, demonstrating the specificity of our sensing system to CTVrelated nucleic acid. The storage stability of SPCE-AuNP electrodes was also investigated by monitoring the impedance response of in situ RPA each week up to 1 month. The sensor has showed an extended shelf-life of working for more than three weeks. Along with that, the repeatability of responses for a 100 pg μL -1 of CTV-p20 gene target on the thiolated primer-modified SPCE-AuNP surfaces was studied through intra-and inter-day assays, obtaining a relative standard deviation (RSD) of 8 % which demonstrated the good performance of this proof-of-concept approach.

CONCLUSION
We have developed the first integrated label-free in situ isothermal RPA amplification/detection based on AuNP-modified SPCE employing impedance for the detection of CTV-related nucleic acid. Specific primers for p20 gene of CTV genome were designed and RPA amplified products were investigated by gel electrophoresis. Then, the effect of both reaction volume and amplification temperature was also evaluated by gel analysis. For in situ isothermal RPA, AuNP-modified electrodes were coated with thiolated forward primer (SH-(AT7)-F1) to form the sensing layer that recognizes CTV-p20 target and start the direct synthesis of duplex DNA onto electrode surface. A simple electrochemical detection of CTV was performed using faradic impedance to characterize the electrochemical performance before and after in situ amplification. Charge transfer resistance parameter was selected to monitor the changes on the electrode surface which results from the immobilization of sensing layer and the amplified duplex DNA. Our in situ isothermal RPA amplification/detection sensor showed a logarithm relation in the range of 1 to 1000 pg μL -1 of CTV-related nucleic acid with LOD of 1000 fg μL -1 with a total assay time of 80 min (60 min RPA amplification and 20 min readout). The sensor performance including specificity and storage life along with intra-and inter-day assays was also studied. Moreover, the results demonstrated the good reproducibility of the biosensors with RSD below 10%. The in situ RPA sensor exhibits great advantages over the modified RPA analyzed by gel electrophoresis in terms of simplicity (no heat source and no label required), sensitivity and portability together with allowing quantitative analysis of nucleic acid. The proposed biosensor is of high potential interest for in-field applications for plant pathogen early detection, which would overcome the limitations of classical molecular methods such as PCR.