An Enzyme-Based Biosensor for the Detection of Organophosphate Compounds Using Mutant Phosphotriesterase Immobilized onto Reduced Graphene Oxide

Research Centre for Chemical Defence, National Defence University of Malaysia, Sungai Besi Camp, Kuala Lumpur 57000, Malaysia Enzyme and Microbial Technology Research Center, Faculty of Biotechnology and Biomolecular Sciences, Universiti Putra Malaysia, Serdang 43400, Malaysia Center for Defence Foundation Studies, National Defence University of Malaysia, Kem Perdana Sungai Besi, Kuala Lumpur 57000, Malaysia Centre for Tropicalisation, National Defence University of Malaysia, Sungai Besi Camp, Kuala Lumpur 57000, Malaysia


Introduction
e use of agricultural and domestic pesticides is a common practice among farmers and plant breeders for the purpose of protecting their crops from pests so that high-quality produce with high marketability can be cultivated. Increases in the use of pesticides, however, have raised public concerns over the possibility of leachate contamination of such pesticides in water sources, food sources, and the environment as a whole [1]. Organophosphate (OP) compounds are examples of a group of toxic compounds that are presently used as pesticides. Apart from being a major constituent in many brands of pesticide, OP compounds such as paraoxon, parathion, and malathion are also used to make plasticizers and other insecticides and even misused as chemical warfare agents [2][3][4][5]. OP compounds are known to be inhibitors towards the enzyme acetylcholine esterase (AChE), an enzyme that is crucial for the functional operation of the nervous system. Inhibition of the enzyme AChE has neurotoxic effects on humans and animals as it disrupts the transfer of neurotransmitters, which will result in paralysis and ultimately cause death [6,7]. Cases involving OP compound pollution have been reported over the past few years and these numbers are increasing at an alarming rate [1,8,9]. e development of portable sensors utilizing functional materials, polymers, and biomolecules is seen to ease the process of analysis through the use of more simplified methods [10]. One manner of performing such detection is through the use of enzymes as bioreceptors for the desired compound. e use of enzymes in biosensors is deemed to be favorable due to its specificity and selectivity towards specific substrates. Organophosphate hydrolase (OPH) is an enzyme capable of hydrolyzing the P-O bonds of OP compounds, thus releasing the radicals that contribute to the toxicity of the compound and thereby rendering it neutralized [11,12].
Phosphotriesterase (PTE), a type of OPH, has been shown to be able to hydrolyze OP compounds such as paraoxon and parathion exceptionally well [13,14]. Tsai et al. have highlighted a specific PTE variant known as mutant H257Y/L303 T (YT-PTE), which exhibits enhanced hydrolysis affinity towards a series of toxic enantiomers of the OP compound [15]. In addition to high affinity towards various OP compound enantiomers, they also exhibit enhanced kinetic value resulting in a faster hydrolysis rate of OP compound than other mutant variants and wild-type PTE [15]. is enzyme is a suitable candidate for the development of a biosensor utilizing an on-site detection method where the OPH will act as a bioreceptor that is fabricated onto a screen-printed electrode to allow for the electrochemical detection of specific OP compounds [16]. In addition to the detection ability against OP compounds, using PTE as a biosensor is a technique that can be done in a relatively short amount of time and both the electrode and enzyme are relatively inexpensive [17].
In various works in the literature, it has been described that the detection of OP compounds utilizing enzymes has been conducted via either an electrochemical approach [18,19] or an optical approach [20,21]. In the electrochemical approach, OPH is usually immobilized onto nanomaterial supports such as carbon nanotubes (CNTs), carbon black, mesoporous carbon, and graphene. e use of these carbon-based nanomaterial supports is appealing since these materials are relatively inexpensive, have good conductive properties, and are easy to handle [22]. e immobilization of OPH onto conductive nanomaterials facilitates electron transfer within the sensor system since biomolecules such as enzymes on their own are generally nonconductive in nature [23]. Nowadays, the use of graphene oxide (GO) as nanomaterial support has gained some interest among researchers due to its structural and electrochemical properties. Although GO can be tailored to have many functional groups on the surface that can efficiently bind any biomolecule [24], the presence of oxide layers could reduce the electrochemical sensing performance in terms of redox activities [25]. As a result, in order to eliminate oxide layers, the GO has to go through the reduction process, forming reduced graphene oxide (rGO). rGO has gained interest in electrochemical sensing due to its unique properties, including high surface area [26], low manufacturing cost [27], good biocompatibility [28], outstanding mechanical flexibility [29], and high thermal and electrical conductivity [30][31][32]. e utilization of rGO as a biorecognition also has been widely used for electrochemical sensing, such as cortisol hormone [33,34], cancer biomarker [35], dopamine [36], glucose [37], human T-lymphotropic virus-1 (HTLV-1) [38], and DNA [39]. However, earlier studies have not yet reported on the combination of YT-PTE with rGO as a new biorecognition for OP detection, which is seen as an interesting work to explore. Hence, this work reports the immobilization of phosphotriesterase YT mutant (YT-PTE) onto reduced graphene oxide (rGO) as a promising bioreceptor, with high affinity and high specificity, for use in the electrochemical detection of OP compounds.

Expression and
Purification of PTE. E. coli BL21 harboring pET51b/YT-PTE was grown in autoinduction media following the method of Studier [40] with glucose as the primary source of carbon and α-lactose as the inducer. e starting culture was inoculated in 400 mL of autoinduction media supplemented with 50 μg/mL ampicillin and grown in an incubator shaker at 37°C, 150 rpm for 3 h. e temperature was then reduced to 25°C and the culture was further incubated for 24 h. Next, the culture was centrifuged at 4°C, 10 000 rpm for 10 mins, resuspended in 0.1 M Tris-Cl buffer (pH 9.3), and sonicated thrice with intermittent timing of 30 secs to obtain the soluble crude PTE. Purification of the PTE was conducted using anion exchange chromatography XK16/20 column (GE Healthcare, USA) packed with Q-sepharose resin. e column was equilibrated with 0.1 M Tris-Cl buffer (pH 9.3) prior to loading of the crude enzyme and was washed five times to remove any unbound proteins. e bound protein was then eluted with 0.1 M Tris-Cl buffer (pH 9.3) supplemented with 0.5 M NaCl.

Assay of Phosphotriesterase Activity.
e enzyme activity of PTE was measured using paraoxon as the substrate following the method of Laothanachareon et al. [41] with slight modification. 100 μL of 1 mM paraoxon was mixed with 890 μL of PBS solution prior to the addition of 10 μL of the enzyme in a microcentrifuge tube. e mixture was briefly mixed by vortexing and was incubated at 30°C for 10 minutes. e reaction was terminated by adding 200 μL of absolute ethanol. e supernatant of the mixture was collected for absorbance reading at A 400 . A unit of phosphotriesterase activity was defined as the production of one mol of p-nitrophenol (pNP) per minute under standard conditions.

Construction of the Sensing Electrode SPCE/rGO/YT-PTE.
A 1 mg/mL of reduced graphene oxide (rGO) suspension was dispersed to homogeneity in PBS buffer solution by ultrasonication for 10 mins. An equal amount of 5 mM EDC-NHS solution as the cross-linker was added and the mixture was incubated at 4°C with mild shaking for 1 h. e mixture was then centrifuged at 10 000 rpm for 5 mins and the supernatant was discarded. e pallet of rGO obtained was washed with PBS buffer for a few times to remove any residual EDC-NHS. Different concentrations of PTE (2, 4, 6, 8, and 10 mg/mL) were added to the mixture and further incubated at 4°C with mild shaking for 16 h. ese concentrations of YT-PTE represented different rGO to enzyme ratios which were 1 : 2, 1 : 4, 1 : 6, 1 : 8, and 1 : 10, respectively. After the incubation period, the mixture was once again centrifuged at 10 000 rpm for 10 min and the pellet of immobilized YT-PTE onto rGO was collected for fabrication onto a blank SPCE. 5 μL of the immobilized YT-PTE was drop-casted onto the working electrode of the SPCE and was allowed to air dry at 4°C overnight. e construct of the SPCE/rGO/YT-PTE is illustrated in Scheme 1.

Measurement Procedure.
e incubation time of the substrate is crucial for electrochemical analysis as sufficient incubation time for the reaction to occur between the enzyme and the substrate is needed to ensure a significant peak current to be detected for analysis. e optimum incubation time for the biosensor was determined by incubating 1 mM paraoxon on the biosensor platform for 5, 10, 15, and 20 minutes. e time when the peak current reached saturation point was chosen to be the optimum incubation time for the biosensor. It was found that 10 minutes was the optimal incubation time.
en, 10 μL of 1 mM paraoxon was dropped onto the surface of the modified working electrode as the substrate and was allowed to react with the immobilized PTE for 10 min. After this reaction time, 90 μL of 5 mM Fe 3 (CN) 6 solution supplemented with 0.1 M KCl as the electrolyte was dropped onto the surface of the modified electrode, and the electrochemical analysis was recorded using cyclic voltammetry (CV) and differential pulse voltammetry (DPV). e potential range of CV and DPV analysis was set in the range of −1.0 to 1.0 V and −0.2 to 0.4 V with a scan rate of 50 mV/s and 10 mV/s, respectively.

Characterization of the Modified Electrode.
In recent time, amine-functionalized rGO has been shown to exhibit better water dispersal [42], reinforce stability, and improve affinity towards redox reaction [43]. Apart from that, it provides numerous amine functional groups for the formation of amide bond linkages during the interaction with EDC-NHS cross-linkers [23]. Enzyme cross-linking typically occurs within the amine group of the amino acid residues; nevertheless, the formation of linkages can also found within the carboxylic group of the amino acid residues [44,45]. e addition of EDC-NHS cross-linkers activates the amino group of rGO prior to covalent attachment to the carboxylic group of the enzyme, as illustrated in Scheme 2. is interaction forms a strong amide bond, which held the enzyme onto the support securely and provided structural support for the enzyme. e formation of immobilized YT-PTE onto the rGO was confirmed by morphological analysis using a Field Emission Scanning Electron Microscopy (FESEM). FESEM analysis was conducted on two samples, SPCE/rGO and SPCE/rGO/YT-PTE, and the difference in their morphologies was compared and recorded. e morphology of SPCE/rGO appeared to be flaky and have a crumpled appearance as shown in Figure 1(a). Immobilization of YT-PTE onto rGO changed the morphology of SPCE/rGO/YT-PTE into a more compact and smoother appearance as shown in Figure 1(b). is morphology is attributed to the formation of an immobilized YT-PTE layer on top of rGO, which inadvertently indicated the success of immobilization of the enzyme onto rGO surfaces [46].
Electrochemical analysis of conductivity and resistivity of SPCE/rGO/YT-PTE and SPCE/rGO/YT-PTE was conducted using cyclic voltammetry (CV) and Electrochemical Impedance Spectroscopy (EIS). rGO was fabricated onto SPCE to enhance the conductivity of the biosensor and to act as a carrier for immobilizing the YT-PTE [47]. e 2D structure of rGO enables the fast flow of electrons between the Fe 3 (CN) 6 analyte and the working electrode, owing to its good conductivity characteristics. As was expected, the enzyme YT-PTE, which is a nonconductive biomolecule, and its immobilization onto the working electrode increased the resistance against electron flow. e resistivity of YT-PTE was analyzed with CV and EIS and this is illustrated in Figures 2(a) and 2(b), respectively.
CV analysis of the biosensor, as compared to a bare SPCE, displayed an increase of the CV anodic peak by 58.2% when rGO was fabricated onto the SPCE in contrast with YT-PTE immobilized onto rGO, which recorded an increment of only 8.0%. e EIS analysis measured the frequency of electron transfer, which is then illustrated as a semicircle-shaped Nyquist plot. e Nyquist plot from the

Mechanism of Electrochemical Detection.
In this work, we utilized the enzymatic reaction between YT-PTE and paraoxon to the basis for the development of a biosensor that is able to determine the level of paraoxon present. In the presence of paraoxon, immobilized YT-PTE hydrolyzes the P-O bond in paraoxon to the corresponding diethyl phosphate and p-nitrophenol (pNP) as shown in Scheme 3.
Since the amount of pNP generated is proportionate to the amount of OP compound present in the sample [6,12,48], the level of paraoxon can thus be quantified by the electrochemical oxidation signal of pNP that couples with Fe 3 (CN) 6 redox molecules as the form of signal amplification. In general, as the concentration of paraoxon increases, more pNP molecules can be generated and oxidized, thus resulting in an increase in the electrochemical current signal present.

Optimization of the Biosensor.
e performance of the biosensor depends on the enzyme activity of YT-PTE. e amount of support to enzyme ratio (rGO : YT-PTE) needs to be optimized to ensure maximum enzyme activity recovery. e amount of rGO used in this study was fixed at 1 mg/mL, following the method described in previous works of literature [23,53]. Different concentrations of YT-PTE (2, 4, 6, 8, and 10 mg/mL) were mixed and allowed to be immobilized onto rGO. Figure 3(a) shows the recovered activity of the YT-PTE after being immobilized onto various concentrations of rGO. It can be seen that the recovered activity of immobilized PTE increased when higher concentrations of PTE were used, up to 1 : 8 support to enzyme ratio, and it shows the optimum recovered activity of up to 90%, which eventually decreased at the 1 : 10 ratio. Maximum loading of enzyme depended on the structure and conformation of the nanomaterial support. e use of EDC-NHS as a cross-linker formed secure linkages, which held YT-PTE securely on the rGO despite the fact that it is difficult to adhere the enzyme onto a 2D planar configured nanomaterial. Inadvertently, the simple planar configuration of rGO was found to not have a detrimental effect upon YT-PTE since up to 90% of enzyme activity was recovered after the enzyme underwent the immobilization processes. e use of a nonbiocompatible redox solution, Fe 3 (CN) 6 , had a detrimental effect on the immobilized enzyme. is solution disrupts the structural conformation of the enzyme, especially its active sites, thus rendering it denatured [54,55].
is caused major concern since the detection mechanism of this sensor relies on the hydrolysis of OP compounds by the enzyme. For that reason, the substrate was allowed to react with the enzyme first prior to the addition of the redox solution for electrochemical analysis of the hydrolysis product. e incubation time for hydrolysis of the OP compound was investigated at 5 min intervals. As shown in Figure 3(b), the peak current detected for pNP increased with increasing the incubation time from 0 mins to 10 mins. e peak current was recorded at a plateau level when the time reached 10 mins, which signaled the completion of the hydrolysis reaction. erefore, 10 mins was chosen as the optimal substrate incubation time prior to the addition of the redox solution for electrochemical analysis.

Electrocatalytic Analysis of SPCE/rGO/PTE.
e performance of SPCE/rGO/YT-PTE was investigated in a series of paraoxon dilutions, which were measured with differential pulse voltammetry (DPV). e effect of different concentrations of paraoxon was tested using concentrations ranging between 1 mM and 0.5 μM. e DPV peak resulted in an anodic peak about +0.1 V (Figure 4(a)), which corresponds to pNP oxidation in the Fe 3 (CN) 6 redox solution. In general, as the concentration of paraoxon rises, the oxidation peak rises, which occurred at +0.1 V, correlating with the fact that the amount of pNP generated is proportional to the amount of paraoxon found in the sample [6,12,48]. While the emergence of an anodic oxidation peak of pNP is often very small, the incorporation of rGO as a support for the immobilization of YT-PTE greatly enhanced this oxidation signal [56]. Figure 4(b) represents the relationship between the logarithm concentration of paraoxon and the peak current, which was illustrated by the linear expression y � −1.99E − 06x + 6.9E − 06 with an R 2 value of 0.9151. e calibration curve demonstrated a linear relationship between the -log 10 concentration of paraoxon and the DPV peak current signal. e limit of detection (LOD) of the biosensor was calculated at 0.11 μM. e LOD calculated in this work was equivalent to the LOD of other OPH-based OP biosensors reported in the literature (see Table 1). e prominent advantages of YT-PTE were that it had high sensitivity towards paraoxon in particular and further modifications such as the incorporation of nanoparticles or metal oxides could be done to increase the potential sensitivity of the biosensor [58].

Evaluation of the Biosensor.
e reproducibility of the SPCE/rGO/YT-PTE biosensor was determined by fabricating the immobilized YT-PTE onto different SPCEs prior to electrochemical analysis using 1 mM paraoxon as the substrate. e analysis was conducted weekly over a period of five weeks. As shown in Figure 5 selectivity of the fabricated electrode was tested using 1 mM of different substrates: paraoxon, parathion, malathion, chlorpyrifos, and diazinon. Based on Figure 5(b), paraoxon displayed the highest current signal followed by parathion.
is is due to the nature of YT-PTE, which had a higher hydrolysis affinity towards phosphotriesterases such as paraoxon and parathion [59,60]. e higher affinity towards these two compounds could be attributed to the structural homogeneity that each other shared, as shown in Figure 6. e only difference between these two compounds was the presence of a P-S bond in parathion as compared to P-O bonds in paraoxon [16]. Consequently, weaker current signals were detected with malathion, chlorpyrifos, and diazinon due to the different phosphorus bond types found with these OP compounds.

Application of the Biosensor in a Real Water Sample.
Since OP compounds are commonly used for various agricultural purposes, as mentioned earlier, these  compounds are often found leached into nearby water sources [61,62]. is poses a serious threat as the contamination can directly affect public health, both human and animal. e detection of OP compounds in real water samples using the SPCE/rGO/YT-PTE biosensor was tested by spiking different water samples: lake water, drain water, and soil run-off water, with 1 mM of paraoxon. As shown in Figure 7, the current signal of distilled water (a sample with minimal or no interferences) spiked with paraoxon exhibited is 4.72 ± 0.985 μA, while for the soil run-off water sample, the current signal was 4.17 ± 0.465 μA. It can be clearly seen that the values of the current signal reduce only around 11%. Moreover, in lake water and drain water, the current signal reduces to 1.79 ± 0.502 μA and 1.50 ± 0.434 μA, respectively. It is expected due to high interference from environmental changes such as salinity, temperature, and pH that will affect the enzyme activity. e detection of current signals in these samples varied as the different water samples harbored different interference levels towards the biosensor. Nevertheless, the changes in the current signal with and without the presence of paraoxon can still be obtained. It was thus ascertained that the SPCE/rGO/YT-PTE biosensor system could be utilized for the detection of OP compounds in various water samples.

Conclusion
In general, YT-PTE immobilized onto rGO was able to retain up to 90% of its enzyme activity as compared to the free enzyme. Subsequently, the incorporation of rGO as the support for immobilizing YT-PTE aided in the electron transfer flow within the biosensor. e constructed biosensor was also shown to be able to detect the oxidation signal of pNP, which is a product of enzymatic hydrolysis of paraoxon. e electrochemical sensor utilizing rGO and YT-PTE in this work recorded a linear range of 1 mM-0.05 μM with a detection limit measured at 0.11 μM. In addition to that, the biosensor showed selectivity towards different OP compounds, especially paraoxon and parathion. Immobilization of YT-PTE onto rGO resulted in a good bioreceptor, which can easily be fabricated onto SPCE for fast and portable field detection of OP compounds in real environmental water samples.

Data Availability
e data used to support the findings of this study are included within the article.