Detection of the Plant Pathogen Pseudomonas Syringae pv. Lachrymans on Antibody-Modified Gold Electrodes by Electrochemical Impedance Spectroscopy

The present work describes an impedimetric immunosensor for Pseudomonas syringae pv. lachrymans (Psl) detection. This pathogen infects many crop species causing considerable yield losses, thus fast and cheap detection method is in high demand. In the assay, the gold disc electrode was modified with 4-aminothiophenol (4-ATP), glutaraldehyde (GA), and anti-Psl antibodies, and free-sites were blocked with bovine serum albumin (BSA). Sensor development was characterized by cyclic voltammetry (CV) and antigen detection by electrochemical impedance spectroscopy (EIS) measurements. Seven analyzed strains of Psl were verified as positive by the reference method (PCR) and this immunoassay, proving sensor specificity. Label-free electrochemical detection was in the linear range 1 × 103–1.2 × 105 CFU/mL (colony-forming unit) with an R2 coefficient of 0.992 and a detection limit (LOD) of 337 CFU/mL. The sensor did not interfere with negative probes like buffers and other bacteria. The assay was proven to be fast (10 min detection) and easy in preparation. The advantage was the simplicity and availability of the verified analyte (whole bacteria) as the method does not require sample pretreatment (e.g., DNA isolation). EIS biosensing technique was chosen as one of the simplest and most sensitive with the least destructive influence on the probes compared to other electrochemical methods.


Introduction
Pseudomonas syringae species divide into 60 subtypes (pathovars) basing on pathogenic characters, and lachrymans is one of them [1,2]. These Gram-negative bacteria are responsible for diseases of many crop species causing spot, speck, and blight diseases [3]. The main hosts are cucumbers, tomato, apples, olives, oats, rice, flowers, and more [4]. Angular leaf spot (ALS) is a widespread cucumber disease caused by this phytopathogen, limiting its open-field production [5][6][7]. The symptoms are water-soaked lesions on the leaves, and when they become necrotic contribute to the reduced often offering higher sensitivities thanks to advanced transducer technology [42]. Antibody-based biosensors can detect both matrix or surface proteins of the target pathogen. In the case of external proteins, practically no sample preparation is needed, which is the main advantage over genosensors.
From electrochemical measurement methods, electrochemical impedance spectroscopy (EIS) is favorable thanks to real-time reaction monitoring, various data generation (about the electrode structure, and chemical and psychical changes during analyte binding), being label-free, and having low-current requirements and a nondestructive impact on the analyte [43].
In this work, we want to present an alternative sensor for DNA-based sensors detecting Psl robustly and accurately. The label-free electrochemical detection of Psl on gold electrodes modified with anti-Psl antibodies was not previously reported. The immunosensor was proven to be fast (10 min detection) and easy in preparation. The advantage is the simplicity and availability of the verified analyte (whole bacteria). The assay does not require DNA isolation or any sample pretreatment. The electrochemical impedance spectroscopy method was chosen as it is not destructive for the analyte.

Materials and Reagents
All chemicals were used without further purification. Aqueous solutions were made using double-distilled sterile water (ddH 2 O). Ethanolic and aqueous solutions were deaerated using argon from Air Products (Poland) and stored in 4 • C conditions. For electrochemical measurements, potassium hexacyanoferrate(II), potassium hexacyanoferrate(III), and potassium chloride were purchased from Chempur (PiekaryŚląskie, Poland). The working gold electrodes were purchased from Mineral (Poland), the reference silver chloride electrode was purchased from Mineral (Łomianki-Sadowa, Poland), and a platinum sheet was the counter electrode. For electrodes preparation and modification, 99.8% ethanol and sulfuric acid were provided by Chempur (PiekaryŚląskie, Poland); phosphate-buffered saline (PBS) tablets, 97% 4-ATP, 25% GA, and Bovine Serum Albumin (BSA) were provided from Sigma Aldrich (Munich, Germany). The anti-Psl polyclonal antibody was purchased from Creative Diagnostics (USA). For bacteria growth, Yeast extract -Peptone -Glucose -Agar (LPGA) medium, peptone, yeast extract, and agar were provided by Sigma Aldrich (Munich, Germany). Seven strains of Psl were purchased from CIRM-CFBP (Beaucouzé, France) and are described in Table 1. Genomic DNA was isolated with the ExtractMe DNA Bacteria kit from Blirt (Gdansk, Poland). The primers were synthesized by Sigma Aldrich (Munich, Germany), and for qPCR amplification, the SensiFAST SYBR No-ROX kit from Bioline (London, UK) was used.

Bacterial Samples Verification and Quantification
The Psl was identified in samples by qPCR according to Gazdik et al. [27] and based on a sequence of the nfrB gene. A 2 µL of sample was added to reaction mixture consisting of 5 µL 2×SensiFast Sybr No-Rox Mix, 1 µL of each primer (10 µM): Pal-nfrB-fwd (5 -GTA CGG TAT GCA AGC GAT CTTC-3 ) and Pal-nfrB-rev (5 -CCG AAA CAG GTA ACG GT-3 ) [27], and 1 µL of water. Quantification of bacteria was based on a standard curve made of three-fold dilutions of genomic DNA of the reference strain Psl 1729 and is presented in Figure 1. The genomic DNA was isolated according to the ExctractMe Bacterial DNA kit protocol. in order to obtain overnight cultures. The overnight cultures were centrifuged 6500 RPM, 10 min. Bacterial pellets were resuspended in 1 mL sterile ddH2O, and OD600 was measured. The bacterial inoculums were adjusted to OD600 approx. 0.1, which roughly equals 0.7 × 10 7 CFU/mL (colonyforming unit).

Bacterial Samples Verification and Quantification
The Psl was identified in samples by qPCR according to Gazdik et al. [27] and based on a sequence of the nfrB gene. A 2 μL of sample was added to reaction mixture consisting of 5 μL 2×SensiFast Sybr No-Rox Mix, 1 μL of each primer (10 μM): Pal-nfrB-fwd (5′-GTA CGG TAT GCA AGC GAT CTTC-3′) and Pal-nfrB-rev (5′-CCG AAA CAG GTA ACG GT-3′) [27], and 1 μL of water. Quantification of bacteria was based on a standard curve made of three-fold dilutions of genomic DNA of the reference strain Psl 1729 and is presented in Figure 1. The genomic DNA was isolated according to the ExctractMe Bacterial DNA kit protocol.

Apparatus
Cyclic voltammetry and EIS measurements were carried out on PalmSens4 potentiostat (Netherlands) with PSTrace 5.6 software. For bacteria preparation, Thermo Scientific Pico 17 (USA) was used, and OD600 was measured on Biosan DEN-1 Biogenet (Poland) densometer. The qPCR reaction was carried out in Bio-Rad CFX Connected thermocycler (USA). The DNA concentration was measured by Thermo Scientific NanoDrop One.

Electrochemical Measurements
All electrochemical measurements were performed using a three-electrode system. Gold discs of 3-mm diameter (ca. 0.0707 cm 2 surface area) served as working electrodes, silver chloride electrode (Ag/AgCl/3 M KCl) was used as a reference electrode, and a platinum sheet was a counter electrode. Cyclic voltammetry (CV) measurements were carried for gold modification steps evaluation ( Figure  S1 in the Supplementary Material), and EIS measurements were carried for both gold modification steps evaluation and bacteria detection. The supporting electrolyte was 5 mM K3[Fe(CN)6]/K4[Fe(CN)6] in 0.1 M KCl aqueous solution and used for both CV and EIS experiments. Before use, the electrolyte was deaerated using argon stream. CV data were collected in the voltage window of -0.5 to +0.5 V at the scan rate of 100 mV/s, always in triplicate. The impedance spectra were recorded in frequencies between 10 kHz and 0.5 Hz (21 steps). The applied potential value was

Apparatus
Cyclic voltammetry and EIS measurements were carried out on PalmSens4 potentiostat (Netherlands) with PSTrace 5.6 software. For bacteria preparation, Thermo Scientific Pico 17 (USA) was used, and OD 600 was measured on Biosan DEN-1 Biogenet (Poland) densometer. The qPCR reaction was carried out in Bio-Rad CFX Connected thermocycler (USA). The DNA concentration was measured by Thermo Scientific NanoDrop One.

Electrochemical Measurements
All electrochemical measurements were performed using a three-electrode system. Gold discs of 3-mm diameter (ca. 0.0707 cm 2 surface area) served as working electrodes, silver chloride electrode (Ag/AgCl/3 M KCl) was used as a reference electrode, and a platinum sheet was a counter electrode. Cyclic voltammetry (CV) measurements were carried for gold modification steps evaluation ( Figure S1 in the Supplementary Materials), and EIS measurements were carried for both gold modification steps evaluation and bacteria detection. The supporting electrolyte was 5 mM K 3 [Fe(CN) 6 ]/K 4 [Fe(CN) 6 ] in 0.1 M KCl aqueous solution and used for both CV and EIS experiments. Before use, the electrolyte was deaerated using argon stream. CV data were collected in the voltage window of −0.5 to +0.5 V at the scan rate of 100 mV/s, always in triplicate. The impedance spectra were recorded in frequencies between 10 kHz and 0.5 Hz (21 steps). The applied potential value was formal potential (E f ) +55 mV (held for 45 s before measurement) and a sinusoidal potential of 10 mV amplitude. All experiments were done in triplicate. Measurement data were analyzed using EIS Spectrum Analyser [44].

Preparation of Electrodes
Before use, the gold electrodes were rinsed with ethanol and dried in the argon stream. Further, they were cleaned with CV (30 scans in 0.5 M H 2 SO 4 in the potential range from −0.1 to +1.5 V with the scan rate of 100 mV/s) and tested by CV and EIS measurements to ensure the acceptable electrochemical parameters (Z' of 150 Ω in EIS and redox peaks ∆E of 110 mV in CV).

Fabrication of Electrochemical Immunosensor
After electrodes cleaning, they were rinsed with ethanol and incubated in 0.1 M ethanolic solution of 4-Aminothiophenol (4-ATP) for 17 h in 4 • C conditions. The unbound particles were flushed away with pure ethanol, and the gold surface was dried in an argon stream. In the next modification step, the 2.5% aqueous solution of GA was placed on the electrode and left for 15 min in a dark place (room temperature). The excess of GA was profusely rinsed with ddH 2 O, and the surface was dried using argon. On the gold electrodes modified with 4-ATP-GA linker, the anti-Psl antibodies diluted 1000-times (according to the producer information) in 1 × PBS pH 7.4 were placed for 1.5 h and incubated in 4 • C conditions. After this step, the electrodes were gently rinsed with a small amount of 1 × PBS pH 7.4 and left wet to avoid the structure disruption. The last step of modification included free-sites blockage. The 1% aqueous solution of BSA was placed on antibodies-modified gold and left for 30 min incubation in 4 • C conditions. The sensor prepared according to the above procedure was flushed gently with 1 × PBS pH 7.4 and stored in stable conditions (100% relative humidity, 4 • C) before use.

The Sensing Principle of Biosensor
The immunosensor preparation procedure is presented in Scheme 1. In the first modification step, the thiol groups of 4-ATP were chemisorbed on the gold surface in a structurally well-defined layer. This spontaneous but stable organization of molecules on solid structures is called self-assembled monolayer (SAM). On the top of SAM, the amine groups of 4-ATP remained free. The second modification step involved the amide bond formation between 4-ATP and GA. GA is a cross-linking reagent that binds amine groups. After modification, the formed surface was aldehyde-ended, which can further bind proteins. The next step was anti-Psl antibodies linkage on the top of Au/4-ATP/GA. The last step was for the exclusion of any unspecific reactions between electrode free-sites and antigen. The neutral, non-reactive BSA protein was used as a blocking reagent. All detailed information about the modification condition is described in the Materials and Methods section. formal potential (Ef) +55 mV (held for 45 s before measurement) and a sinusoidal potential of 10 mV amplitude. All experiments were done in triplicate. Measurement data were analyzed using EIS Spectrum Analyser [44].

Preparation of Electrodes
Before use, the gold electrodes were rinsed with ethanol and dried in the argon stream. Further, they were cleaned with CV (30 scans in 0.5 M H2SO4 in the potential range from −0.1 to +1.5 V with the scan rate of 100 mV/s) and tested by CV and EIS measurements to ensure the acceptable electrochemical parameters (Z' of 150 Ω in EIS and redox peaks ΔE of 110 mV in CV).

Fabrication of Electrochemical Immunosensor
After electrodes cleaning, they were rinsed with ethanol and incubated in 0.1 M ethanolic solution of 4-Aminothiophenol (4-ATP) for 17 h in 4 °C conditions. The unbound particles were flushed away with pure ethanol, and the gold surface was dried in an argon stream. In the next modification step, the 2.5% aqueous solution of GA was placed on the electrode and left for 15 min in a dark place (room temperature). The excess of GA was profusely rinsed with ddH2O, and the surface was dried using argon. On the gold electrodes modified with 4-ATP-GA linker, the anti-Psl antibodies diluted 1000-times (according to the producer information) in 1 × PBS pH 7.4 were placed for 1.5 h and incubated in 4 °C conditions. After this step, the electrodes were gently rinsed with a small amount of 1 × PBS pH 7.4 and left wet to avoid the structure disruption. The last step of modification included free-sites blockage. The 1% aqueous solution of BSA was placed on antibodiesmodified gold and left for 30 min incubation in 4 °C conditions. The sensor prepared according to the above procedure was flushed gently with 1 × PBS pH 7.4 and stored in stable conditions (100% relative humidity, 4 °C) before use.

The Sensing Principle of Biosensor
The immunosensor preparation procedure is presented in Scheme 1. In the first modification step, the thiol groups of 4-ATP were chemisorbed on the gold surface in a structurally well-defined layer. This spontaneous but stable organization of molecules on solid structures is called selfassembled monolayer (SAM). On the top of SAM, the amine groups of 4-ATP remained free. The second modification step involved the amide bond formation between 4-ATP and GA. GA is a crosslinking reagent that binds amine groups. After modification, the formed surface was aldehydeended, which can further bind proteins. The next step was anti-Psl antibodies linkage on the top of Au/4-ATP/GA. The last step was for the exclusion of any unspecific reactions between electrode freesites and antigen. The neutral, non-reactive BSA protein was used as a blocking reagent. All detailed information about the modification condition is described in the Materials and Methods section.

Electrochemical Characterization of Biosensor
The fabrication process of Psl biosensor was characterized by CV and EIS measurements. CV records indicated the charge transfer changes, and EIS records showed the resistance changes on the electrode surface. According to Figure 2a, the CV spectra of the bare Au electrode showed its reversible behavior towards active redox species used in the electrolyte ([Fe(CN) 6 ] 3-/4-). The potential difference (∆E) between reduction and oxidation reactions was 110 mV with current heights of 120 mA and 132 mA for cathodic and anodic peak, respectively. The 4-ATP chemisorption resulted in the drastic decrease of this characteristic, proving the electron transfer blockage between the electrolyte and the electrode. No characteristic maximum of the cathodic current peak was notable, and the anodic peak moved towards higher potentials. The following stages of protein linkage to the gold platform resulted in the successive lowering of anodic currents. Simultaneously, the EIS spectra (Figure 2b) for the bare gold electrode showed the standard behavior of the electrochemical system. The Nyquist plot (Z'(Z") function) indicated both the interfacial electron-transfer kinetics and redox probe diffusion, which determined the general electrode process. This characteristic is defined as Faradaic impedance, where the curve is semicircle in the higher frequencies' region (kinetics) and straight in the lower frequencies (diffusion) [45]. The SAM of 4-ATP significantly changed the EIS characteristics, increasing the impedance values but remaining in kinetics-diffusion rate. In the next steps, the linkage of GA, similarly, the increase of impedance was observed. These changes were relatively small compared to the next steps. Proteins immobilization (anti-Psl and BSA) showed a drastic rise in the semicircle diameter.

Electrochemical Characterization of Biosensor
The fabrication process of Psl biosensor was characterized by CV and EIS measurements. CV records indicated the charge transfer changes, and EIS records showed the resistance changes on the electrode surface. According to Figure 2a, the CV spectra of the bare Au electrode showed its reversible behavior towards active redox species used in the electrolyte ([Fe(CN)6] 3-/4-). The potential difference (∆E) between reduction and oxidation reactions was 110 mV with current heights of 120 mA and 132 mA for cathodic and anodic peak, respectively. The 4-ATP chemisorption resulted in the drastic decrease of this characteristic, proving the electron transfer blockage between the electrolyte and the electrode. No characteristic maximum of the cathodic current peak was notable, and the anodic peak moved towards higher potentials. The following stages of protein linkage to the gold platform resulted in the successive lowering of anodic currents. Simultaneously, the EIS spectra ( Figure 2b) for the bare gold electrode showed the standard behavior of the electrochemical system. The Nyquist plot (Z'(Z'') function) indicated both the interfacial electron-transfer kinetics and redox probe diffusion, which determined the general electrode process. This characteristic is defined as Faradaic impedance, where the curve is semicircle in the higher frequencies' region (kinetics) and straight in the lower frequencies (diffusion) [45]. The SAM of 4-ATP significantly changed the EIS characteristics, increasing the impedance values but remaining in kinetics-diffusion rate. In the next steps, the linkage of GA, similarly, the increase of impedance was observed. These changes were relatively small compared to the next steps. Proteins immobilization (anti-Psl and BSA) showed a drastic rise in the semicircle diameter. The received EIS data from the electrode modification process was fitted to the Randles circuit (Scheme 2). The Re[CPE(RctZW)] equivalent electrical circuit expresses electrolyte-electrode interfacial and consists of electrolyte resistance (Re) combined parallel with the constant phase element (CPE) and an impedance Faradaic reaction (charge transfer resistance, Rct). Diffusion in the low frequencies' region is presented as the Warburg element (ZW) [46]. The received EIS data from the electrode modification process was fitted to the Randles circuit (Scheme 2). The R e [CPE(R ct Z W )] equivalent electrical circuit expresses electrolyte-electrode interfacial and consists of electrolyte resistance (R e ) combined parallel with the constant phase element (CPE) and an impedance Faradaic reaction (charge transfer resistance, R ct ). Diffusion in the low frequencies' region is presented as the Warburg element (Z W ) [46].
The R ct parameter is the most important for the impedance signal to analyze [47], thus this parameter was chosen as the evaluation of bacteria detection. The percentage of R ct changes were calculated from the equation: where R ct 1 is for bare Au, and R ct 2 is for the modification step. Calculated data is presented in Table 2.
The data confirmed the modification success as the charge-transfer process was impeded, suggesting the increase of a biolayer thickness. The chemisorption of 4-ATP resulted in a 66% increase of R ct compared to the bare electrode. The GA linkage resulted in a 132% increase of R ct compared to the bare electrode. Subsequent increases derived from protein immobilization, giving a strong increase of 491% and 713% for anti-Psl and BSA, respectively. The Rct parameter is the most important for the impedance signal to analyze [47], thus this parameter was chosen as the evaluation of bacteria detection. The percentage of Rct changes were calculated from the equation: where Rct 1 is for bare Au, and Rct 2 is for the modification step. Calculated data is presented in Table  2. The data confirmed the modification success as the charge-transfer process was impeded, suggesting the increase of a biolayer thickness. The chemisorption of 4-ATP resulted in a 66% increase of Rct compared to the bare electrode. The GA linkage resulted in a 132% increase of Rct compared to the bare electrode. Subsequent increases derived from protein immobilization, giving a strong increase of 491% and 713% for anti-Psl and BSA, respectively.

Psl Bacteria Identification by qPCR
Psl identification was performed by qPCR amplification according to the method described in the Experimental section. The pathogen was detected in all samples, as shown in Table 3. Seven Psl strains and positive control (17 ng DNA) were checked as positive, three negative controls were confirmed as negative. By determining the standard curve (Figure 1), the number of viable bacteria in the tested samples was calculated. The study showed bacterial availability at the level of 0.47 to 1.17 × 10 9 CFU/mL.

Psl Bacteria Identification by qPCR
Psl identification was performed by qPCR amplification according to the method described in the Experimental section. The pathogen was detected in all samples, as shown in Table 3. Seven Psl strains and positive control (17 ng DNA) were checked as positive, three negative controls were confirmed as negative. By determining the standard curve (Figure 1), the number of viable bacteria in the tested samples was calculated. The study showed bacterial availability at the level of 0.47 to 1.17 × 10 9 CFU/mL.

Psl Bacteria Detection on Developed Electrochemical Immunosensor
The developed sensing system was applied for the detection of Psl. Firstly, the reference Psl strain 1729 was incubated on the Au/4-ATP/GA/anti-Psl/BSA electrode, and impedance spectra were recorded  Figure 3 presents the sensor response on Psl 1729. The strong change of EIS characteristic was observed after 5 min incubation. The baseline (BSA-modified electrode) drifted between electrodes due to nonrepetitive gold structures, thus the sensor response was presented in the relative impedance change. The percentage of R ct changes were calculated from the equation: where R ct S is for sample, and R ct B is for Au/4-ATP/GA/anti-Psl/BSA. nonrepetitive gold structures, thus the sensor response was presented in the relative impedance change. The percentage of Rct changes were calculated from the equation: where Rct S is for sample, and Rct B is for Au/4-ATP/GA/anti-Psl/BSA. The charge-transfer resistance value increased 281.8% over the Au/4-ATP/GA/anti-Psl/BSA measurement due to the presence of Psl. The second measurement after 10 min incubation resulted in the following Rct increase of 334.52%. The third spectra after 15 min incubation practically remained unchanged compared to 10 min incubation. The results suggest that the binding of Psl cells to the antibodies creates a barrier for the electrochemical process. The redox probe had partly hindered access to the electrode surface resulting in a notable increase in the Rct value. Moreover, similar spectra for 10 and 15 min incubation suggest that after 10 min the system was in equilibrium, which means that all antibodies active sites were saturated with antigen. The 10 min incubation was chosen as detection time, and the percentage change of the Rct parameter was chosen as the sensor response towards samples.
.  The charge-transfer resistance value increased 281.8% over the Au/4-ATP/GA/anti-Psl/BSA measurement due to the presence of Psl. The second measurement after 10 min incubation resulted in the following R ct increase of 334.52%. The third spectra after 15 min incubation practically remained unchanged compared to 10 min incubation. The results suggest that the binding of Psl cells to the antibodies creates a barrier for the electrochemical process. The redox probe had partly hindered access to the electrode surface resulting in a notable increase in the R ct value. Moreover, similar spectra for 10 and 15 min incubation suggest that after 10 min the system was in equilibrium, which means that all antibodies active sites were saturated with antigen. The 10 min incubation was chosen as detection time, and the percentage change of the R ct parameter was chosen as the sensor response towards samples.
The next step of immunosensing system tests involved different concentrations of Psl for linear response and LOD determination. The series of eight bacterial concentrations expressed in CFU/mL were checked. Figure 4a displays the impedance changes obtained with Au/4-ATP/GA/anti-Psl/BSA electrode incubation with different concentrations of Psl. There is a progressive increase in sensor response with the increased Psl amount. The results indicate a significant impedance difference between Psl concentrations due to the increasing number of Psl cells which bind to immunosensor. At a concentration of 100 CFU/mL, the impedance change from a bacterial sample was on the negative control level. The smallest positive R ct change was for 1 × 10 3 CFU/mL (14.68%) and the highest for 1.2 × 10 5 CFU/mL (334.52%). No more increase was observed with the subsequent incubations with a higher concentration of Psl. The linear relationship between the impedance change (R ct change) and Psl concentration describes the equation: with an R 2 coefficient of 0.992 and is presented in Figure 4b.
control level. The smallest positive Rct change was for 1 × 10 3 CFU/mL (14.68%) and the highest for 1.2 × 10 5 CFU/mL (334.52%). No more increase was observed with the subsequent incubations with a higher concentration of Psl. The linear relationship between the impedance change (Rct change) and Psl concentration describes the equation: with an R 2 coefficient of 0.992 and is presented in Figure 4b.
(a) (b) The sensitivity was established as the lowest quantity of Psl detectable in the assay (LOD). LOD was calculated from the impedance signal deriving from negative control (nontarget Haemophilus influenzae bacteria (Hin, results described below)). The Hin Rct change signal was multiplied by three giving LOD of 11.7%, which equals 377 CFU/mL of Psl. Table 4 presents in detail the comparison of various detection methods of Pseudomonas syringae species mentioned in this article with the proposed electrochemical biosensor. According to the standard PCR method, which detects bacterial cells, this assay is less sensitive than real-time qPCR [27]. The real-time qPCR LOD is 1.3 times lower, but these values are very similar, in the same order of magnitude. However, the developed biosensor is more sensitive than PCR variation-loop-mediated isothermal amplification (LAMP) [28] and detects 29.7 times less Pseudomonas syringae cells. The last step of sensor evaluation was repeatability/reproducibility and selectivity studies. For the repeatability determination, one Au/4-ATP/GA/anti-Psl/BSA sensor was freshly prepared and used for the triplicate measurement of reference Psl strain 1729 (concentration of 10 4 CFU/mL). For the reproducibility determination, three separately prepared Au/4-ATP/GA/anti-Psl/BSA sensors The sensitivity was established as the lowest quantity of Psl detectable in the assay (LOD). LOD was calculated from the impedance signal deriving from negative control (nontarget Haemophilus influenzae bacteria (Hin, results described below)). The Hin R ct change signal was multiplied by three giving LOD of 11.7%, which equals 377 CFU/mL of Psl. Table 4 presents in detail the comparison of various detection methods of Pseudomonas syringae species mentioned in this article with the proposed electrochemical biosensor. According to the standard PCR method, which detects bacterial cells, this assay is less sensitive than real-time qPCR [27]. The real-time qPCR LOD is 1.3 times lower, but these values are very similar, in the same order of magnitude. However, the developed biosensor is more sensitive than PCR variation-loop-mediated isothermal amplification (LAMP) [28] and detects 29.7 times less Pseudomonas syringae cells. The last step of sensor evaluation was repeatability/reproducibility and selectivity studies. For the repeatability determination, one Au/4-ATP/GA/anti-Psl/BSA sensor was freshly prepared and used for the triplicate measurement of reference Psl strain 1729 (concentration of 10 4 CFU/mL). For the reproducibility determination, three separately prepared Au/4-ATP/GA/anti-Psl/BSA sensors were used for the analysis of the same probe (Psl 1729, 10 4 CFU/mL). The percentage changes of impedance in the presence of Psl 1729 on one sensor resulted in the average relative standard deviation (RSD) value of 1.74%. For three separate sensors, the average RSD resulted in 3.45%. For the selectivity studies, the series of negative controls were tested to exclude any unspecific responses. The negative PBS buffer used for bacteria dilution, LPGA bacterial medium, and nontarget Hin bacteria were chosen.
All the samples were separately incubated on electrodes, and the series of measurements were recorded, similarly to the detection of positive Psl 1729. Figure 5 presents the sensor response towards all negative samples.
were used for the analysis of the same probe (Psl 1729, 10 4 CFU/mL). The percentage changes of impedance in the presence of Psl 1729 on one sensor resulted in the average relative standard deviation (RSD) value of 1.74%. For three separate sensors, the average RSD resulted in 3.45%. For the selectivity studies, the series of negative controls were tested to exclude any unspecific responses. The negative PBS buffer used for bacteria dilution, LPGA bacterial medium, and nontarget Hin bacteria were chosen. All the samples were separately incubated on electrodes, and the series of measurements were recorded, similarly to the detection of positive Psl 1729. Figure 5 presents the sensor response towards all negative samples. From Figure 5, all the negative samples did not change the character of impedance spectra. Negligible Rct increase was observed after 5 min incubation; thus, samples were further incubated in 5 min cycles until the stability of the system. In all cases, the repeatable EIS was after 10 min. similarly to positive Psl 1729 strain, confirming the chosen detection time. The Rct increases were 1.3% and 2.3%, and for PBS and LPGA bacterial medium, respectively, compared to Au/4-ATP/GA/anti-Psl/BSA. The highest Rct increase was 3.9% for nontarget Hin bacteria. For selectivity determination, except for negative control, seven PCR-confirmed strains of Psl were used as positive samples and examined on immunosensor. Detailed results of all tested samples were presented in the percentage Rct change in Figure 6. From Figure 5, all the negative samples did not change the character of impedance spectra. Negligible R ct increase was observed after 5 min incubation; thus, samples were further incubated in 5 min cycles until the stability of the system. In all cases, the repeatable EIS was after 10 min. similarly to positive Psl 1729 strain, confirming the chosen detection time. The R ct increases were 1.3% and 2.3%, and for PBS and LPGA bacterial medium, respectively, compared to Au/4-ATP/GA/anti-Psl/BSA. The highest R ct increase was 3.9% for nontarget Hin bacteria. For selectivity determination, except for negative control, seven PCR-confirmed strains of Psl were used as positive samples and examined on immunosensor. Detailed results of all tested samples were presented in the percentage R ct change in Figure 6.
The 11.7% of R ct change (LOD) was established as a threshold for the distinction between positive and negative samples in this assay. Three negative controls did not exceed this threshold, where all seven Psl strains confirmed by PCR were also tested as positive in the immunoassay giving visibly more significant impedance change. The strongest interaction with the anti-Psl antibodies had strain no. 1729 (334.52%), and the weakest strain no. 2275 (29.48%). According to obtained results, the sensor was highly specific for the detection of Psl bacteria and not giving false-positive results for PBS buffer, medium, and other bacteria. The 11.7% of Rct change (LOD) was established as a threshold for the distinction between positive and negative samples in this assay. Three negative controls did not exceed this threshold, where all seven Psl strains confirmed by PCR were also tested as positive in the immunoassay giving visibly more significant impedance change. The strongest interaction with the anti-Psl antibodies had strain no. 1729 (334.52%), and the weakest strain no. 2275 (29.48%). According to obtained results, the sensor was highly specific for the detection of Psl bacteria and not giving false-positive results for PBS buffer, medium, and other bacteria.

Discussion
This work presented the design and characterization of anti-Psl antibodies immobilization onto gold electrodes for impedimetric detection of Psl bacteria. The development of the platform consisted of the 4-ATP self-assembled monolayer, glutaraldehyde crosslinker, antibodies, and BSA proteins. Bacterial detection relied on the impedance changes (increases) caused by Psl cells binding onto the Au/4-ATP/GA/anti-Psl/BSA surface. The assay could detect Psl at concentrations as low as 337 CFU/mL with a linear detection range 10 3 -1.2 × 10 5 CFU/mL (R 2 = 0.992). The sensitivity was in the same order of magnitude as the standard method (LOD for real-time qPCR was 1.3 times lower). Simultaneously, the assay was 30 times more sensitive than the LAMP method, which is an isothermal alternative for PCR. The sensor was proved to detect seven Psl strains. The specificity was evaluated using negative controls: PBS buffer, LPGA medium, and nontarget H. influenzae bacteria. There were no significant impedance changes when probes were incubated (all did not exceed the LOD of 11.7% Rct change), which evidenced the sensing specificity towards Psl. The total detection time was 10 min.
The ease of assay preparation, label-free detection, and miniaturization possibilities qualifies it for the portable biosensing system. There is a continuous demand for simple analytical methods for the determination of many biochemical analytes. Assays based on antibody-antigen interactions are promising thanks to high specificities and sensitivities. The in-field monitoring of many crops for Psl infection risk is possible due to the accessibility of target analyte (whole bacteria cells). The chosen EIS detection method was fast, nondestructive for samples, and required low currents (battery mode). In the future, authors want to examine Psl directly from infected species, e.g., cucumber leaves, and implement the developed method on disposable electrodes.

Discussion
This work presented the design and characterization of anti-Psl antibodies immobilization onto gold electrodes for impedimetric detection of Psl bacteria. The development of the platform consisted of the 4-ATP self-assembled monolayer, glutaraldehyde crosslinker, antibodies, and BSA proteins. Bacterial detection relied on the impedance changes (increases) caused by Psl cells binding onto the Au/4-ATP/GA/anti-Psl/BSA surface. The assay could detect Psl at concentrations as low as 337 CFU/mL with a linear detection range 10 3 -1.2 × 10 5 CFU/mL (R 2 = 0.992). The sensitivity was in the same order of magnitude as the standard method (LOD for real-time qPCR was 1.3 times lower). Simultaneously, the assay was 30 times more sensitive than the LAMP method, which is an isothermal alternative for PCR. The sensor was proved to detect seven Psl strains. The specificity was evaluated using negative controls: PBS buffer, LPGA medium, and nontarget H. influenzae bacteria. There were no significant impedance changes when probes were incubated (all did not exceed the LOD of 11.7% R ct change), which evidenced the sensing specificity towards Psl. The total detection time was 10 min.
The ease of assay preparation, label-free detection, and miniaturization possibilities qualifies it for the portable biosensing system. There is a continuous demand for simple analytical methods for the determination of many biochemical analytes. Assays based on antibody-antigen interactions are promising thanks to high specificities and sensitivities. The in-field monitoring of many crops for Psl infection risk is possible due to the accessibility of target analyte (whole bacteria cells). The chosen EIS detection method was fast, nondestructive for samples, and required low currents (battery mode). In the future, authors want to examine Psl directly from infected species, e.g., cucumber leaves, and implement the developed method on disposable electrodes.