Article Cite This: Anal. Chem. 2018, 90, 13491−13497
pubs.acs.org/ac
Rapid and Sensitive Detection of Aspergillus niger Using a SingleMediator System Combined with Redox Cycling Jungwook Kwon,† Eun-Min Cho,‡ Ponnusamy Nandhakumar,† Sung Ik Yang,*,‡ and Haesik Yang*,‡ †
Department of Chemistry and Chemistry Institute for Functional Materials, Pusan National University, Busan 46241, Korea Department of Applied Chemistry, Kyung Hee University, Yongin 17104, Korea
‡
Downloaded via UNIV OF SOUTH DAKOTA on December 20, 2018 at 04:56:26 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
S Supporting Information *
ABSTRACT: Rapid and sensitive mold detection is becoming increasingly important, especially in indoor environments. Common mold detection methods based on doublemediated electron transfer between an electrode and molds are not highly sensitive and reproducible, although they are rapid and simple. Here, we report a sensitive and reproducible detection method specific to Aspergillus niger (A. niger), based on a single-mediator system combined with electrochemical-chemical (EC) redox cycling. Intracellular NAD(P)H-oxidizing enzymes in molds can convert electro-inactive hydroxy-nitro(so)arenes into electro-active hydroxy-aminoarenes. Since the membrane and wall of A. niger is well permeable to both a substrate (4-nitro-1-naphthol) and a reduced product (4-amino-1-naphthol) in tris buffer (pH 7.5) solution, the electrochemical signal is increased in the presence of A. niger due to two reactions: (i) enzymatic reduction of the substrate to the reduced product and (ii) electrochemical oxidation of the reduced product to an oxidized product. When a reducing agent (NADH) is present in the solution, the oxidized product is reduced back to the reduced product and then electrochemically reoxidized. This EC redox cycling significantly amplifies the electrochemical signal. Moreover, the background level is low and highly reproducible because the substrate and the reducing agent are electro-inactive at an applied potential of 0.20 V. The calculated detection limit for A. niger in a common doublemediator system consisting of Fe(CN)63− and menadione is ∼2 × 104 colony-forming unit (CFU)/mL, but the detection limit in the single-mediator system combined with EC redox cycling is ∼2 × 103 CFU/mL, indicating that the newly developed single-mediator system is more sensitive. Importantly, the detection method requires only an incubation period of 10 min and does not require a washing step, an electrode modification step, or a specific probe.
R
oxygen, but rapidly with the organic mediator and the electrode. A system consisting of Fe(CN)63− and menadione is the most common of double-mediator systems (Figure 1a).1,2,7−10 NAD(P)H-oxidizing enzymes (NOEs), such as diaphorase, reduce menadione to menadiol in the presence of NAD(P)H inside the cell, and menadiol then transfers electrons to Fe(CN)63− in the solution and finally to the electrode. NAD(P)H is continuously supplied through catabolism of digested nutrients such as glucose, which allows high, steady-state electrochemical signals. In this doublemediator system, the electrochemical signal is mainly dependent on the amount of menadiol produced within the cell. Diaphorase enzymes can convert hydroxy-nitro(so)arenes into hydroxy-aminoarenes in the presence of NAD(P)H.11−13 Such a conversion can also occur within cells containing NOEs (Figure 1b). When hydroxy-aminoarenes are electrochemically oxidized, quinone imines are produced. A quinone imine can be reduced back to a hydroxy-aminoarene by a reducing agent (without reduction of the substrate hydroxy-nitro(so)arene), and the hydroxy-aminoarene is then electrochemically
apid and easy electron transfer between an electrode and cells has played a pivotal role in (i) detection of viable microorganisms, (ii) monitoring of cellular dynamics, (iii) assessment of cell viability and cytotoxicity, and (iv) microbial electrocatalysis for microbial fuel and electrolysis cells.1−4 Mediated electron transfer using small electron mediator(s) is commonly employed for this purpose, because direct electron transfer between an electrode and cells can only be used in limited circumstances.5,6 Both oxidized and reduced forms of an electron mediator must pass through hydrophobic regions of cell membrane (and wall) to obtain a high mediated electron-transfer rate without using harsh permeabilization conditions.1,2 The reduced form of the electron mediator must transfer the electron(s) rapidly to the electrode or cells before it is oxidized by the oxygen dissolved in the solution to obtain high electron-transfer efficiency. In general, it is difficult to meet these two requirements by using a single-mediator system. Thus, a double-mediator system comprising an oxygen-resistant organometallic mediator and a lipophilic organic mediator has commonly been used.1,2,7 Both the oxidized and the reduced forms of the organic mediator pass through the hydrophobic regions of cell membrane (and wall), and the oxidized form rapidly reacts with the redox enzymes inside the cell. The organometallic mediator reacts slowly with © 2018 American Chemical Society
Received: July 31, 2018 Accepted: October 26, 2018 Published: November 7, 2018 13491
DOI: 10.1021/acs.analchem.8b03417 Anal. Chem. 2018, 90, 13491−13497
Article
Analytical Chemistry
Here, we present a rapid and sensitive mold detection method (a single-mediator system combined with redox cycling) that is based on two key ideas: (i) intracellular NOEs can be used for the catalytic conversion of hydroxynitro(so)arene into hydroxy-aminoarene and (ii) the electrochemical oxidation signal of hydroxy-aminoarene can be increased by EC redox cycling in the presence of a reducing agent. For this purpose, three hydroxy-nitro(so)diarenes were tested as substrates, and NADH and H3N-BH3 were tested as reducing agents. Optimal conditions for high signal amplification were determined, and they were applied to a comparative study of detecting five molds. Finally, the detection limit for the single-mediator system combined with EC redox cycling was compared with that for the common double-mediator system consisting of Fe(CN)63− and menadione.
■
EXPERIMENTAL SECTION Chemicals and Solutions. Hydroxylamine hydrochloride, H2O2, NH4OH, ethylenediaminetetraacetic acid, 4-nitroso-1naphthol (4-NO-1-N), NADH (β-nicotinamide adenine dinucleotide reduced dipotassium salt), ammonia-borane (H3N-BH3), menadione, K3Fe(CN)6 [potassium ferricyanide(III)], glucose, tris(hydroxymethyl)-aminomethane (tris), phosphate-buffered saline (PBS), and all the reagents used for the optimization experiments were purchased from Sigma Aldrich Co. 4-Nitro-1-naphthol (4-NO2-1-N) and 1-nitro-2naphthol (1-NO2-2-N) were obtained from Tokyo Chemical Industry Co., Ltd. Phosphate-buffered saline solution (PBS, pH 7.4) contained 10 mM phosphate, 0.138 M NaCl, and 2.7 mM KCl. Tris buffer solutions (pH 7.5 and 9.0) were prepared using 50 mM tris and 1.0 M HCl. ITO electrodes were purchased from Corning Co. (Daegu, Korea). A. niger (KCCM strain 11239) and Aspergillus fumigatus (A. fumigatus, KCCM strain 60331) were obtained from the Korean Culture of Microorganisms. Penicillium citrinum (P. citrinum, KCTC strain 46042) and Alternaria alternata (A. alternata, KCTC strain 46457) were obtained from the Korean Collection for Type Cultures. Cladosporium cladosporioides (C. cladosporioides) was collected from childcare centers. The molds were cultured on potato dextrose agar in Petri dishes incubated for up to 1−2 weeks, depending on the species and until spores were visible, at 25 °C. Afterward, spores were collected using 5 mL of deionized water and filtered using cell strainer (SPL Life Science) to remove any unwanted debris (hyphae and agar). The concentration of conidia was determined by cell counting with a hemocytometer. Procedure of A. niger Detection. For the detection of A. niger, 100 μL of tris buffer (pH 7.5) containing 10 mM NADH, 100 μL of tris buffer (pH 7.5) containing 50 mM glucose, 500 μL of tris buffer (pH 7.5) containing 0.2 mM 4NO2-1-N, and 200 μL of tris buffer (pH 7.5) were mixed with 100 μL of sample containing different concentrations of a mold. The mixed solution (1.0 mL) was injected into a vessel of an electrochemical cell. The resulting concentrations of NADH, glucose, and 4-NO2-1-N were 1.0, 5.0, and 0.1 mM, respectively. Preparation of Electrodes and Electrochemical Measurements. The ITO electrodes were pretreated with a 5:1:1 solution of H2O, H2O2 (30%), and NH4OH (30%) at 70 °C for 1 h. Teflon electrochemical cells were assembled with a sensing electrode, an Ag/AgCl (3 M NaCl) reference electrode, and a Pt counter electrode. The exposed geometric area of the sensing electrodes was approximately 0.28 cm2.
Figure 1. Schematic of electrochemical mold detection based on (a) a double-mediator system consisting of Fe(CN)63− and menadione and (b) a single-mediator system combined with electrochemical-chemical (EC) redox cycling. Schematic of unwanted side reactions: (c) direct oxidation of NADH at an electrode, (d) direct reaction between a substrate and NADH, and (e) electrode-mediated substrate reduction.
reoxidized. This electrochemical−chemical (EC) redox cycling can significantly increase electrochemical signals.13−16 Hydroxy-nitro(so)diarenes are more hydrophobic than hydroxynitro(so)monoarenes, allowing higher permeability of cell membranes (and walls). Hydroxy-aminodiarenes are easily electrochemically oxidized at low overpotentials, even on low electrocatalytic indium tin oxide (ITO) electrodes.13,17,18 Moreover, hydroxy-aminodiarenes remain stable when a reducing agent such as NADH is present in solution, although they can be readily oxidized by dissolved oxygen in the absence of a reducing agent.13,17,18 Therefore, if both a hydroxynitro(so)diarene and its reduced form (hydroxy-aminodiarene) pass easily through cell membranes (and walls), the hydroxynitro(so)diarene can be used as an organic mediator for a single-mediator system that does not require an organometallic mediator. In this system, the electrochemical signal is dependent on the amount of hydroxy-aminodiarene produced in a solution as well as within cells. This system offers the possibility of obtaining a higher electrochemical signal than the double-mediator system consisting of Fe(CN) 6 3− and menadione. Mold detection, especially in indoor environments, is becoming increasingly important because molds pose a severe threat to human health.19,20 Although many methods for mold detection have been developed, they are still far from being rapid and sensitive. Time-consuming culture-based detection, immunoassay, and DNA amplification-based detection are not suitable for rapid detection.21−26 Simple mold detection based on mediated electron transfer between an electrode and cells could be a good solution for rapid and sensitive detection if its sensitivity is improved. Aspergillus niger (A. niger) is one of the most common mold species found in mesophilic environments such as soil and enclosed air environments.27,28 A. niger can infect the lungs and ears of patients with weak immune systems, causing aspergillosis. Therefore, rapid and sensitive detection of A. niger is of great importance. 13492
DOI: 10.1021/acs.analchem.8b03417 Anal. Chem. 2018, 90, 13491−13497
Article
Analytical Chemistry
Figure 2. (a) Chemical structures of three hydroxy-nitro(so)diarenes. (b) Cyclic voltammograms obtained at a scan rate of 20 mV/s at bare ITO electrodes after an incubation period of 10 min in tris buffer (pH 7.5) containing (i) 5.0 mM glucose and 1.0 mM NADH; (ii) 5.0 mM glucose, 1.0 mM NADH, and 0.1 mM 4-NO2-1-N; and (iii) 5.0 mM glucose, 1.0 mM NADH, 0.1 mM 4-NO2-1-N, and 105 CFU/mL A. niger. (c) Chronocoulograms obtained at 0.20 V at bare ITO electrodes after an incubation period of 10 min in tris buffer (pH 7.5) containing (i) 5.0 mM glucose, 0.1 mM 4-NO2-1-N, and 1.0 mM NADH; (ii) 5.0 mM glucose, 0.1 mM 4-NO2-1-N, 1.0 mM NADH, and 105 CFU/mL A. niger; (iii) 5.0 mM glucose, 0.1 mM 4-NO2-1-N, 1.0 mM H3N-BH3; (iv) 5.0 mM glucose, 0.1 mM 4-NO2-1-N, 1.0 mM H3N-BH3, and 105 CFU/mL A. niger; and (v) 5.0 mM glucose, 0.1 mM 4-NO2-1-N, and 105 CFU/mL A. niger. (d) Histogram of the charge values measured at 100 s from the chronocoulograms obtained at 0.20 V at ITO electrodes after an incubation period 10 min in tris buffer (pH 7.5) containing (i) 5.0 mM glucose and 1.0 mM reductant and (ii) 5.0 mM glucose, 1.0 mM reductant, and 105 CFU/mL A. niger. (e) Histogram of the charge values obtained in three different buffers containing 0.1 mM 4-NO2-1-N, 1.0 mM NADH, and 5.0 mM glucose. The error bars represent one standard deviation of the measured values.
(vi) the reaction between the substrate and the reducing agent should be very slow, and (vii) direct oxidation of the reducing agent should be slow at the electrode. To meet all the requirements, hydroxy-nitro(so)diarenes were considered as possible substrates, and NADH and H3N-BH3 were considered as possible reducing agents, based on the facts (i) that hydroxy-aminodiarenes are more hydrophobic than hydroxyaminomonoarenes, (ii) that hydroxy-nitro(so)arenes are rapidly reduced by NOEs, (iii) that hydroxy-aminodiarenes are stable in the presence of a strong reducing agent, (iv) that hydroxy-aminodiarenes are rapidly oxidized even at a bare ITO electrode, (v) that EC redox cycling involving hydroxyaminodiarene and NADH (or H3N-BH3) is fast, (vi) that the direct reaction between hydroxy-nitro(so)diarene and NADH (or H3N-BH3) is slow, and (vii) that the direct oxidation of NADH and H3N-BH3 is slow at an ITO electrode.13,17,18
Incubation was performed for 10 min at room temperature, and electrochemical measurements were conducted using a CHI 708C (CH Instruments, Austin, TX, U.S.A.) for 100 s. The total detection time was
