Development of an Electrochemical Immunosensor for Phakopsora pachyrhizi Detection in the Early Diagnosis of Soybean Rust

A ferrugem da soja é uma doença que ocorre nas folhas da soja e é considerada muito agressiva, pois reduz a qualidade do grão final. A identificação precoce do fungo na plantação previne várias perdas para o agricultor bem como para as culturas vizinhas. Neste trabalho, um imunossensor foi desenvolvido sem a utilização de marcadores usando medidas de impedância para detectar a ferrugem asiática nos extratos das folhas contaminadas nos estágios iniciais da doença, antes do aparecimento dos sintomas. O anticorpo anti-micélio do fungo Phakopsora pachyrhizi (agente causador da doença) foi imobilizado em substrato de ouro via formação de monocamadas autoorganizadas de tióis e ligações covalentes com a biomolécula. O imunossensor apresentou um limite de detecção de 385 ng mL. A otimização das condições experimentais e o bloqueio da superfície para minimizar ligações não-específicas foram realizados. Este estudo proporcionou uma nova perspectiva da utilização do método para o diagnóstico precoce da doença.


Introduction
Asian soybean rust is a fungal disease caused by Phakopsora pachyrhizi which is a virulent pathogen that can quickly defoliate plants, reduce pod set, pod fill, seeds and, thus, quality, reducing crop yields by 10 to 80%. 1 The disease is disseminated through spores (urediniospores) transported by the wind and spreads rapidly, causing loss of foliar area and a severe reduction in grain yield 2 and, consequently, severe economic losses.
The appearance of the disease is recent in countries that now have large production of soybeans, such as Brazil and United States, but the infection cost was estimated at approximately $1.2 billion for past harvests, 3 a large percentage of these losses resulting from inappropriate use of fungicides.
Symptoms begin on the lower leaves of the plant as small lesions on the undersides of the leaves that increase in size and change from gray to brown. 4 The biggest problem is in the early detection of the fungus infestation because it is performed by a visual method.However, when the fungus is readily visible, the culture is already infected.The lesions can also be misinterpreted as non aggressive infections, which are normally found on the crop.The method used to control the pathogen is fungicide application, thus precise and early diagnosis minimizes spore dissemination, decreasing the costs of control and avoiding a continuous environmental contamination. 5hus, farmers need an early identification method that could help them to control disease infestation.][8] Immunosensors are based on the use of an antibody that reacts specifically with a substance (antigen) to be tested.Immobilization of the receptor (e.g., an antigen) on a substrate is convenient for applications of molecular bio-recognition for the detection of a target molecule (e.g., its antibody) present in solution.The specificity of antigen-antibody interactions allows the development of immunosensor devices for clinical diagnostics, environmental monitoring, etc. 9 Immunosensors offer several advantages such as limited hands-on time, high-throughput screening, improved sensitivity, real-time analysis, possibility of quantification and label-free detection (i.e., antibodies do not need to be labelled with fluorophores, radioisotopes, colloidal gold particles or enzymes for detection of binding events or signal enhancement).Therefore, immunosensors can be successfully applied in agricultural, food, environmental, pharmaceutical chemistry and clinical applications. 10,11here are various techniques to evaluate the configuration of immunocomplexes and, among them, Electrochemical Impedance Spectroscopy (EIS) to monitor the binding of antigen-antibody can be emphasized.This is performed in the presence of a redox probe.The formation of immunocomplexes disturbs the double-charged layer at the electrode/electrolyte interface resulting in an increase of thickness and an insulation character to the electrode surface in relation to the redox probe added to the solution.These changes affect the capacitance and the electron transfer at the electrode interface.The great advantage in using EIS as a transducer system is the ability to carry out an immunoassay without needing a tracer, thereby reducing analysis time. 12n this paper, an electrochemical impedance immunosensor for the detection of the Phakopsora pachyrhizi fungus was developed aiming at the diagnosis of Asian rust on soybean leaves.The sensor was constructed employing the mycelium extracted from spores of contaminated leaves, making it possible to identify the disease at its earliest stages, i.e., before the appearance of symptoms.

Apparatus
All electrochemical experiments were carried out in a conventional three-electrode cell at room temperature using a saturated calomel electrode (SCE) and Pt wire as reference and counter electrodes, respectively.The geometric area of the gold working electrode (Metrohm 6.1204.020,Switzerland) was 0.07 cm 2 .The impedance measurements were performed using a PGSTAT 30 model from AUTOLAB (Eco Chemie, Netherlands) interfaced with a personal computer.

Antigen
Urediniospores from naturally infected greenhouse soybean plants were obtained from the collection of Embrapa Londrina, PR, Brazil.The spores were collected using a mechanical harvester, and 300 mg of spores was germinated in 300 mL deionized water in a sterile 13 in × 9 in Pyrex baking dish for 24 h at 22 o C.After that the culture was harvested by centrifugation at 6000xg for 15 min, the pellet was suspended in PBS and frozen at -20 o C, after protein quantification.

Polyclonal antiserum
A female New Zealand rabbit weighing 2 kg was subcutaneously inoculated three times at 14-day intervals with aliquots of Phakopsora pachyrhizi suspension (0.5 mL) containing 200-300 μg of protein emulsified with an equal volume of Freund's complete adjuvant for the first and incomplete adjuvant for the two subsequent injections.Ten days after the final injection, blood was collected through cardiac puncture and the sera were kept at −20 o C. Titration and specificity of the collected serum were determined by indirect enzyme-linked immunosorbent assay (ELISA).Preimmune serum was used as a negative control.

Preparation of the self-assembled monolayer (SAM)
The cleaning of the bare gold surface is critically important for self-assembled monolayer formation and should be accomplished systematically.
The gold surface was first polished with a 0.3 μm alumina slurry.After, the electrode was washed with a large amount of deionized water and then sonicated in pure ethanol for 5 min.In a second step, the electrode was cleaned by immersion in a piranha solution (1:3 mixture of 30% H 2 O 2 /conc.H 2 SO 4 ) for 10 min.Then, the electrode was washed with copious amounts of water.Finally, the gold electrode was electrochemically cleaned in 0.5 mol L -1 H 2 SO 4 solution, cycling the potential between -0.1 and 1.4 V (vs.SCE) for 25 scans. 13The final cyclic voltammogram was compared to those reported by Finklea et al. 14 in order to warrant a clean gold surface to be used each time.The electrode was then immersed for 18 h in 10 mmol L -1 cysteamine solution (CYSTE), freshly prepared in ethanol.The modified electrode (Au-SAM) was further rinsed with ethanol to remove physically adsorbed molecules and was used immediately for antibody coupling.

Antibody coupling
For the antibody coupling, 4 mL of the serum containing 47.6 mg mL -1 of antibody was diluted in 200 mL of 2 mmol L -1 EDC solution plus 200 mL of 5 mmol L -1 NHS solution and 600 mL of 0.1 mol L -1 PBS buffer solution at pH 7.4.This solution was stored in a refrigerator for 2 hours.This step is used to active the carboxylic groups of antibodies by the formation of NHS ester.After that, the SAM modified electrodes were immersed in the antibody solution and stored at 4 °C for 90 min.The unbound antibodies were removed from the electrode surface (Au-SAM-Ab) using the method cited by Geng et al. 15 Finally, the immunosensor was treated with 0.1% of BSA-PBS solution for 30 min to block the nonspecific sites.Then, the electrodes were washed with deionized water (Au-SAM-Ab-BSA).

Impedance measurements
For impedance studies, a sine wave with 10 mV of amplitude was applied to the electrode over the formal potential of the redox couple (0.2 V).Impedance spectra were collected in the frequency range varying from 100 kHz to 100 mHz.Electrochemical impedance spectra were fitted using an Equivalent Circuit contained on the FRA software AUTOLAB (Eco Chemie, Netherlands) and the resistance of charge transfer (R ct ) values were determined in a pH 7.4 PBS buffer using 5 mmol L -1 of Fe(CN) 6 3−/4− as redox probe.Different concentrations of an extract of contaminated soybean leaves (antigen) were injected to study antibody-antigen interactions.

SDS gel electrophoresis
SDS-PAGE according to Laemmli 16 was performed using 12% acrylamide and 5% stacking gels containing 0.1% SDS (reducing conditions).Protein bands were stained with Coomassie brilliant blue R-250.A broad-range protein molar mass marker (Sigma Chemical, Saint Louis, MO, USA) was used for the estimation of protein size.

Preparation of the immunosensor
The immunosensor construction and posterior antigen (extracted from contaminated leaves) binding are presented in Figure 1.Firstly, a process of functionalization of the gold surface was conducted through the formation of a cysteamine monolayer based on the strong Au-thiolate bond.Then, the antibodies with previously activated carboxylic groups were immobilized on the surface.Finally, a specific binding event occurred between the antimycelium and the antigen of soybean leaves.
Electrochemical impedance spectroscopy was an effective tool to investigate the modified electrode with different coating layers in the presence of Fe(CN) 6 3−/4− in PBS buffer solution.At the higher frequencies, squeezed semicircles represent an electron transfer-limited process, followed by a diffusionally limited electron transfer process at the lower frequencies. 17The diameter of this semicircle in the Nyquist plot, which exhibits the electron transfer resistance (R ct ) of the layer, can be used to describe the interface properties of the electrode for each immobilization step. 18,19As shown in Figure 2, the gold surface modification by cysteamine produces a SAM with a high number of vacancies and, therefore, a small barrier to probe the electron transfer.This behavior of the cysteamine-SAM was expected and verified using other techniques. 20When the anti-mycelium and BSA molecules were adsorbed onto the SAM, there is an increase in the R ct values because the proteins generate an insulating layer on the modified surface, which makes the redox reaction of Fe(CN) 6 3−/4− more difficult. 21,22ndle's equivalent circuit was adopted to model the physiochemical process occurring at the gold electrode surface: 23 R s (C[R ct W]) where R s is the resistance of the solution, R ct is the resistance to charge-transfer, W is the Warburg impedance and C is a double layer capacitance.

Antibody immobilization conditions
To provide high sensitivity and good repeatability in the measurements, it is important to find suitable linker compounds to enable packing a high density of antibody on the gold electrode.For this, the incubation time between the anti-mycelium and SAM modified surface was optimized.The electrode surface was coated with 20 μL of 0.8 μg mL -1 antibody solution at pH 7.4 for different times.Table 1 shows the results.It is possible to observe that an incubation time shorter than 120 min seems to be insufficient for antimycelium immobilization with adequate coverage of the surface.The impedance becomes constant after 120 min and this time was selected.
Prior to anti-mycelium immobilization, the carboxylic groups of the antibodies are activated, using EDC and NHS solution.However, the formation of NHS ester can promote cross-linking between the antibodies, producing aggregates.This formation could decrease the free sites on the antibody needed for interaction with the antigen.An experiment was carried out to evaluate aggregate formation after site activation of antibodies by using electrophoresis in polyacrylamide and the results are shown in Figure 3.The  electrophoresis profiles in lines 3 and 4 are very similar, considering the presence of a 155 kDa protein-band (black arrow) referring to the antibody molecules formed by heavy and light chains, which are not possible to observe when aggregates were formed.Even if antibody aggregates are formed, the electrophoresis confirms that there are free antibodies to bind with the surface and, therefore, with the antigens.Moreover, the immobilized antibody on the sensor surface interacts with the antigen with good sensitivity.The concentration of anti-mycelium incubated onto the electrode surface determines the amount of antigen bound to the immunosensor and, therefore, it is an important parameter for optimization.The antigen / antibody reaction is a phenomenon that is denominated the equivalence zone, that is, only when the antibody and antigen are in ideal dilution can a significant interaction occur between them.In the region of an excess of antibody there is less reaction.Observing Figure 4, it is possible to verify that at concentrations above 0.19 mg mL -1 the response becomes almost constant and this concentration was selected for subsequent experiments.

Nonspecific interactions
Usually, nonspecific adsorption is a major problem in label-free immunosensing, since it cannot be distinguished if the increase of R ct was caused by an immunoreaction.One efficient alternative to minimize this effect is the blocking of free reactive sites by a selective layer.In this work 0.05% (m/v) gelatin, 0.1% (m/v) bovine serum albumin (BSA) and 10 mmol L -1 glycine were tested as blocking solutions after antibody immobilization.The impedance responses were monitored for each blocking reagent: (i) immunosensor blocking and antigen adsorption and (ii) immunosensor blocking and healthy soybean leaf extract addition at the same concentration, used as a control.This extract was not reacted with the immobilized anti-mycelium.According to Table 2, the best blocking agent was the BSA solution because the R ct values are very different for contaminated and healthy leaves, indicating that BSA blocks the free sites on the surface, leaving only the available recognition sites for antigen binding.

Optimal immuno-reaction time between anti-mycelium and antigen
The dependency of the impedance shift on the reaction time between the antigen and anti-mycelium is shown in Figure 5.The immunosensor was coated with 20 μL of 1.5 μg mL -1 of pH 7.4 antigen solution for different times.The results show that a reaction time shorter than 8 min seems to be insufficient for complete immunoreaction.Above this time, the impedance response is slightly higher, but there is saturation in the response after the second addition of antigen because the complex of the anti-mycelium and soybean leaf extract became unstable.In this way, an interaction time of 8 min was selected for subsequent studies.

Immunosensor performance
Evaluating of the immunoreaction was carried out by exposing the biosensor Au-SAM-Ab-BSA to various concentrations of antigen.The corresponding Nyquist plots of impedance spectra are shown in Figure 6.It was found that the diameter of the Nyquist circle increased on adding antigen, demonstrating that the analyte interacts with the immobilized antibodies.
Figure 7 shows the curve obtained using the biosensor response as a function of antigen concentration.A linear relationship between the ∆R ct and the concentration of contaminated leaf extract was found in the range 0.35-3.5 μg mL -1 , having a sensitivity of 292 W μg -1 mL.The correlation coefficient was 0.995 (n = 10).The detection limit of the proposed immunosensor was measured to be 315 ng mL -1 .A recently published paper has described Surface Plasmon Resonance monitoring of the Phakopsora pachyrhizi fungus mycelium 24 and the detection limit found was 800 ng mL -1 .This information confirms the higher detectability of the proposed method using EIS for the early diagnosis of soybean rust.
The reproducibility of the biosensor for mycelium of Phakopsora pachrizi was investigated with intra and interassays precision.The intra-assay of the immunosensor was evaluated with three replicates prepared independently under the same experimental conditions and the inter-assay used the same device for six measurements.The intra and inter-assay variation coefficient obtained for 1.5 μg mL -1 of antigen were 6.2 and 5.5%, respectively, indicating acceptable precision and fabrication reproducibility.

Conclusions
In this paper, a label-free impedimetric immunosensor for the detection of mycelium of Phakopsora pachrizi was developed by immobilizing anti-mycelium antibodies on self-assembled cysteamine monolayers via covalent coupling on a gold working electrode, presenting good sensitivity.Therefore, this method has potential application to diagnose Asian soybean rust in the initial stages of infection, allowing fast and efficient control against the disease dissemination.

Figure 1 .
Figure 1.Schematic representation of antibody immobilization by covalent linkage to a gold surface modified with cysteamine.

Figure 2 .
Figure 2. Nyquist plot of sequential immobilization steps onto gold electrodes towards a functional immunosensor in 5 mmol L -1 Fe(CN) 6 3−/4− in pH 7.4 PBS solution for the frequency range from 100 kHz to 100 mHz with 10 mV of amplitude: SAM formed on gold surface (Au-SAM); antimycelium immobilization (Au-SAM-Ab) and after BSA agent blocking adsorption (Au-SAM-Ab-BSA).

Figure 4 .
Figure 4. Effect of concentration of anti-mycelium incubated onto the electrode surface in the immunosensor response using 1.5 μg mL -1 of antigen in 5 mmol L -1 Fe(CN) 6 3-/4-in pH 7.4 PBS solution in the frequency range from 100 kHz to 10 mHz with 10 mV of amplitude.

Figure 5 .
Figure 5.The dependency of the impedance shift on the reaction time between the antigen and anti-mycelium using 1.5 μg mL -1 of antigen in 5 mmol L -1 Fe(CN) 6 3−/4− in pH 7.4 PBS solution for the frequency range from 100 kHz to 10 mHz with 10 mV of amplitude.

Figure 6 .
Figure 6.Nyquist plots for impedance measurements in 5 mmol L -1 Fe(CN) 6 3−/4− in pH 7.4 PBS solution for the frequency range from 100 kHz to 1900 mHz with 10 mV of amplitude for the immunosensor after incubation with different concentrations of antigen from 0.35 to 3.5 μg mL -1 .

Figure 7 .
Figure 7. R ct measurements obtained for the immunosensor in different concentrations of antigen in 5 mmol L -1 Fe(CN) 6 3−/4− in pH 7.4 PBS solution for the frequency range from 100 kHz to 10 mHz with 10 mV of amplitude.

Table 2 .
Effect of the blocking agent on the impedance response using an Au-SAM-Ab immunosensor for the antigen with healthy leaf extract used as control in 5 mmol L -1 Fe(CN) 63−/4− in pH 7.4 PBS solution for the frequency range from 100 kHz to 10 mHz with 10 mV of amplitude