2.1. Chemicals and Reagents
Cysteamine (Cat. No. M9768) was purchased from Sigma Aldrich (St. Louis, MO, USA). Sodium phosphate monobasic monohydrate (NaH2PO4) (Cat. No. 389872500), sodium phosphate dibasic (Na2HPO4) (Cat. No. 424375000), phosphate buffer saline (PBS) (Cat. No. 10010031), potassium Ferricyanide (Cat. No. 196785000), sulfo-NHS (N-hydroxysulfosuccinimide) (Cat. No. 24510), N-(3-(dimethylamino)propyl)-N′-ethylcarbodiimide hydrochloride (Cat. No. 22980), sodium chloride (Cat. No. AC447302500), calcium sulfate (CaSO4, Cat. No. AA33301A7), cupric sulfate (CuSO4, Cat. No. 7758998), and magnesium chloride (MgCl2, Cat. No. LC163801) were purchased from Thermo Fisher Scientific (Waltham, MA, USA). MC-LR (Cat. No. ALX-350-012-C100), anti-MC-LR monoclonal antibody (MC10E7, lgG1 type, ALX-804-320-C200), MC-YR (Cat. No. ALX-350-044-C025), and MC-RR (Cat. No. ALX-350-043-C050) were purchased from Enzo Life (Farmingdale, NY, USA). The Abnova™ MC-LR ELISA kit (Cat. No.89028336) was purchased from Thermo Fisher Scientific (Waltham, MA, USA). The ELISA kit was analyzed using a VersaMax microplate reader (Molecular Devices, San Jose, CA, USA).
0.01 M phosphate buffer solution (PB, pH 7.6) was prepared by mixing solutions of NaH2PO4 and Na2HPO4 in Milli-Q water. The electrolyte solution was prepared by mixing 5 mM [Fe(CN)6]3−/4− in 10 mM PBS solution (pH 7.3).
2.2. Anti-MC-LR/MC-LR/Cysteamine/SPCE Biosensor Fabrication
The surface functionalization process of the Anti-MC-LR/MC-LR/Cysteamine/SPCE biosensor is shown in Fig. 1. First, the screen-printed carbon electrode (SPCE) (RRPE1001C, Pine Research Instrumentation Inc., Durham NC, USA) was sufficiently cleaned by pouring a small amount (approximately 10 mL) of acetone (99.7%, Cat. No. 180349, Thermo Fisher Scientific, Waltham, MA, USA) into individual 50 mL graduated polypropylene centrifuge tubes (Cat. No. 14375150, Thermo Fisher Scientific, Waltham, MA, USA), so that the working electrode (WE) was then completely submerged into the solution. With the lid, the tubes were placed in a large beaker with water and a probe-type ultrasonic bath (CL-334, No. 2015060673, Fisher Scientific), making sure none of the tubes were touching either the walls of the beaker or the probe. With a power of 500 watts and a frequency of 20kHz at an amplitude of 25%, the sensors were sonicated for 90 minutes at room temperature. Then, the sensors were removed and rinsed with 70% isopropanol and Milli-Q water. The sensors were placed in a petri dish with a task wiper (Kimwipes) and dried with N2 (99.999%). Next, a cysteamine self-assembled monolayer (SAM) was coated onto the surface of the carbon WE to provide anchoring sites for the MC-LR molecules (dos Santos et al. 2019). For the SAM coating, 3 µL of 18 mM cysteamine, diluted using Milli-Q water, was dropped onto the surface of the WE. The sensors were then placed in a dark and semi-humid environment. After 4 hours, the electrode was rinsed with Milli-Q water and dried with N2. To immobilize the MC-LR onto the SAM layer of the WE surface, the WE was incubated with an activated solution of MC-LR. The MC-LR was activated by using equal portions of 20 mM sulfo-N-hydroxysulfosuccinimide (sulfo-NHS), 80 mM N-(3-(dimethylamino)propyl)-N′-ethylcarbodiimide hydrochloride (EDC), and 10− 2 g L-1 of MC-LR and spun in the centrifuge (Cat. No. 367160, Beckman Counlter, Inc. (Brea, CA)) at room temperature for 30 min at 500 rpm. Using a pipette, 1 µL of the activated MC-LR solution was dropped onto the WE and allowed to immobilize overnight in the refrigerator (4°C). The sensor was then rinsed with Milli-Q water and dried with N2. The fabricated MC-LR detection biosensor was stored in the refrigerator before use (i.e., antibody application to test water samples).
2.3. Electrochemical Characterization of the MC-LR Detection Biosensor
The electrochemical characterization of the functionalization process of the Anti-MC-LR/MC-LR/Cysteamine/SPCE biosensor was conducted using EIS and CV measurements performed by the PalmSens 4 (Houten, Netherlands). Electrochemical characterization was conducted in three-electrode system using a modified MC-LR/Cysteamine/SPCE biosensor as the working electrode (WE), carbon counter electrode, and an Ag/AgCl reference electrode (MI-401, Microelectrodes, Inc., Bedford, NH, USA). The CV and EIS measurements were analyzed for each step of the functionalization process, bare SPCE, cysteamine/SPCE, and MC-LR/cysteamine/SPCE, and compared to ensure electrode surface functionalization. For the EIS measurements, the potential of the DC and AC were set to 0.001V and 0.25V, respectively, and a frequency range of 0.01 to 100,000 Hz. For the CV measurements, the t equilibrium was set to 8 seconds, the E begin was − 1.0 V, E step was 0.01 V, and the scan rate was 0.05 V s− 1.
2.4. Surface Characterization of the MC-LR Detection Biosensor using SEM and XPS
Surface structures of the fabricated biosensors were obtained with an Ultra 55 scanning electron microscopy (SEM) (ZEISS, Oberkochen, Germany). To characterize the surface chemical states, X-ray photoelectron spectroscopy (XPS) was performed with a Thermo Scientific ESCALAB Xi + X-ray Photoelectron Spectrometer Microprobe (Thermo Scientific™, Massachusetts, USA) with a twin-crystal, micro-focusing monochromator and an Al anode. The XPS analysis chamber during measurement was held at a pressure below 1.0×10− 7 torr. All spectra were calibrated to the carbon 1s peak, set to 284.8 eV.
2.5. MC-LR Detection using the Anti-MC-LR/MC-LR/Cysteamine/SPCE Biosensor
The analytical performance of the biosensor was tested using varying known concentrations of MC-LR. Samples containing concentrations of MC-LR from 10− 7 to 10− 4 g L-1 were mixed with equal volumes of 10− 2 g L-1 Anti-MC-LR and PB (pH 7.6) using the centrifuge at room temperature for 30 min at 500 rpm. This allows for sufficient time for the antibody to bind to the MC-LR in the sample. 1 µL of the sample was then dropped onto the working electrode and let sit, in the closed petridish, for an hour at room temperature to allow the available antibody molecules left in solution (not bound to the MC-LR molecules present in the sample) to bind to the MC-LR molecules on the surface of the sensor. The electrode was rinsed with Milli-Q water and then incubated with PB solution for a couple minutes. The sensor was then tested in the redox probe solution (0.005 M [Fe(CN)6]3-/4- in PBS) using the PalmSens 4 to obtain EIS graphs. The total whole process including mixing antibody in the test solution and incubating the sensor took approximately less than 90 mins. The peak values (Zmax) were obtained from the EIS graph to create the standard curve. The EIS Nyquist plot curves were fitted to the appropriate circuit model using EIS Spectrum Analyser software (ABC Chemistry).
The limit of detection (LOD) was calculated using the following equation (Eq. 1) (Lee et al. 2013).
\(LOD=k\frac{{S}_{b}}{b}\) (Eq. 1)
where k is the parameter whose value is suggested to be 3 by the International Union of Pure and Applicable Chemistry (IUPAC), Sb is the standard deviation of blank signals and b is the slope of the calibration curve.
The reproducibility of the Anti-MC-LR/MC-LR/Cysteamine/SPCE biosensor was tested by preparing new test solutions and comparing to the previously developed standard curve.
The monoclonal antibody (anti-MC-LR) (MC10E7, lgG1 type, ALX-804-320-C200, Enzo Life, Farmingdale, NY, USA) was used because of its high sensitivity and selectivity toward MC-LR (Zeck et al. 2001b). MC10E7 is very stable and exhibits low interferences from humic acid, salts, and surfactants/organic solvents present in solutions (Zeck et al. 2001a), proving it well-suited for its application in a blend of water matrices.
2.6. Selectivity and Interference Tests
The selectivity of the biosensor toward MC-LR was evaluated by testing the cross-reactivity with other common MCs (MC-RR and MC-YR) (Wu et al. 2019). Samples were prepared with a fixed concentration of 10− 5 g L− 1 of MC-LR, MC-RR, and MC-YR, as well as a combination sample of three MCs (i.e., MC-LR, MC-RR, and MC-YR).
The interference resilience of the biosensor was also analyzed by adding possible interfering ions commonly found in natural water to the MC-LR test sample. The ions selected for analysis were Na+, Mg2+, Cu2+, Ca2+, Cl−, and SO42−. In fresh surface water, sodium (Na+) concentrations typically range between zero and 100 mg L− 1, magnesium (Mg2+) concentration range is 1–100 mg L− 1, calcium (Ca2+) concentration range is 1–135 mg L− 1, copper (Cu2+) concentration range is 0.5–1,000 µg L− 1, chloride (Cl−) concentration range is 1–100 mg L− 1, and sulfate (SO42−) concentration averages 53 mg L− 1 (CEPA 2008, Elbein 2017, Hunt et al. 2012, Morr et al. 2006, USEPA 2003). Keeping the ion concentrations within the typical ranges of natural surface water, the chemical concentrations selected were 82.4 mg L− 1 NaCl, 143.4 mg L− 1 MgCl2, 3.93 mg L− 1 CuSO4, and 53.7 mg L− 1 CaSO4. 10− 5 g L− 1 of MC-LR was tested with the selected concentration of each chemical. A sample was also tested with a mixture all the chemicals.
2.7. Applications to Real Surface Water Samples
A natural surface water sample was collected from the Daecheong Lake in South Korea experiencing HAB events (Figure S1). The sonication method (Kim et al. 2009) was used to lyse the cyanobacteria, ensuring the MC-LR concentrations in the test sample. A CO-Z ultrasonic cleaner (PS-10A, AC100-120V, 60Hz) was used to sonicate the cyanobacteria for 30 minutes. The sample was then filtered using a glass fiber prefilter (No. APFA04700, MilliporeSigma, Burlington, MA, USA) to remove the larger particles and then a 0.22 µm membrane filter (No. SLGPM33RS, MilliporeSigma, Burlington, MA, USA) was used to remove the remaining particles. The MC-LR concentration was measured using ELISA (EPA method 546) (Steven C. Wendelken 2016) (Supplementary Information (SI) S.1.1.).