sensors Article

A Disposable Microfluidic Device with a Screen Printed Electrode for Mimicking Phase II Metabolism Rafaela Vasiliadou 1, *, Mohammad Mehdi Nasr Esfahani 2 , Nathan J. Brown 2 and Kevin J. Welham 1 1 2

*

Department of Chemistry, University of Hull, Cottingham Road, Hull HU6 7RX, UK; [email protected] Department of Engineering, University of Hull, Cottingham Road, Hull HU6 7RX, UK; [email protected] (M.M.N.E.); [email protected] (N.B.) Correspondence: [email protected] or [email protected]

Academic Editors: Jesus Iniesta Valcarcel and Craig E. Banks Received: 9 May 2016; Accepted: 25 August 2016; Published: 2 September 2016

Abstract: Human metabolism is investigated using several in vitro methods. However, the current methodologies are often expensive, tedious and complicated. Over the last decade, the combination of electrochemistry (EC) with mass spectrometry (MS) has a simpler and a cheaper alternative to mimic the human metabolism. This paper describes the development of a disposable microfluidic device with a screen-printed electrode (SPE) for monitoring phase II GSH reactions. The proposed chip has the potential to be used as a primary screening tool, thus complementing the current in vitro methods. Keywords: metabolism; electrochemistry; screen-printed electrodes; mass spectrometry

1. Introduction Metabolism is a physiological process occurring in the liver, which enhances the elimination of both endogenous and exogenous compounds. Metabolites are formed during phase I and phase II metabolic reactions [1,2]. During natural conditions, phase I metabolites conjugate with glutathione (GSH)—a reaction catalysed by glutathione S-transferases (GSTs)—and are detoxified (phase II reactions) [3–5]. However, during phase I reactions, reactive metabolites can form, some of which are toxic to DNA and proteins [6,7]. Currently, in vitro methodologies are expensive and involve lengthy experimental procedures [8]. Direct electrochemistry in electrochemical flow cells can mimic the catalytic activity of CYP 450 in a cost effective manner [9]. In this technique, metabolites are synthesized electrochemically on the working electrode and detected by mass spectrometry (MS) either intact or as phase II GSH-adducts [10], depending on the application. Acetaminophen (APAP) and dopamine (DOPA) are electroactive compounds with well-known metabolic routes that are widely used in the field [11–13]. In 1989, Getek et al. generated and detected GSH conjugates of APAP by coupling online an electrochemical flow cell with a thermospray mass spectrometer. The reactive metabolite N-acetyl-p-benzoquinone imine (NAPQI) was formed in the electrochemical cell and reacted subsequently with GSH in a mixing tee [14]. Lohhmann et al. [15], used a commercially available cell equipped with a glassy carbon working electrode, a Pd/H2 reference electrode and a Pd counter electrode. The isomeric adducts of APAP were obtained and detected by liquid chromatography (LC)/mass spectrometry (MS). More recently, a miniaturized electrochemical cell with high conversional rates was developed by Van den Brink et al. [16]. The microfluidic device had a dual application, either the generation of short lived metabolites or the formation of GSH-adduct via the addition of a T-piece on the effluent of chip. However, flow cells and glass microfluidics are fabricated from expensive materials and require tedious cleaning methods for their reuse. A cost effective polymer chip for electrochemical tagging was fabricated by Roussel et al. [17]. The electrospray emitter was Sensors 2016, 16, 1418; doi:10.3390/s16091418

www.mdpi.com/journal/sensors

Sensors 2016, 16, 1418

2 of 15

developed with a microchannel and an integrated carbon electrode. The electrode surface-to-volume ratio was 50 m−1 providing an efficient electrolytic cell for both electrochemical reactions and mass spectrometric detections. Screen-printed electrodes (SPEs) are commercially available disposable sensors, which can also be readily custom-made. The fabrication process relies on the screen printing technique, involving the deposition of ink or paste onto a ceramic, glass, or plastic substrate [18–21]. SPEs are becoming widely applied in the field of electrochemistry due to their simplicity, portability, low cost, mass production, and miniaturized format [21]. Kauffmann et al. [21] used a carbon SPE instead of a conventional flow cell for electrochemically mimicking the metabolism of APAP. The sensor included all the required electrodes—including the working, reference, and counter electrodes—resulting in an alternative miniaturized cell. The reactive metabolite of APAP, the NAPQI, was generated on the working area of the SPE in the presence of excess GSH concentration. NAPQI was covalently bonded with GSH, leading to the formation of an APAP-GSH phase II adduct. The resulting phase II adduct was detected offline by liquid chromatography (LC)/ESI-MS. Thus, the development of a primary screening tool with an SPE can complement the in vitro methods, saving both time and cost. In the present study, we describe the integration of a commercially available sensor (DS-110: DropSens, Oviedo, Spain) with a serpentine reaction channel into a disposable polymeric chip, for mimicking phase II GSH reactions. A cheaper alternative towards the existing flow cells and glass microfluidic chips without the need of any tedious cleaning processes. In addition, the electrochemical step is separated from the ionization process, as seen in the previous polymeric design [17], ensuring that the generated metabolites are electrochemically driven and not the result of ionization. DOPA was used as a probe compound for monitoring the phase II GSH reactions. According to the literature, dopaminoquinone is the reactive intermediate of DOPA and probably the causal agent of Parkinson’s disease [22]. 2. Experimental Section 2.1. Chemical Reagents DOPA and GSH were purchased from Sigma Aldrich (Gillingham, UK). APAP was obtained by Alfa Aesar (Heysham, UK). Buffers of different pH values were prepared using ammonium acetate (Fisher scientific, Loughborough, UK), sodium phosphate dibasic (Sigma Aldrich), and potassium phosphate monobasic (Sigma Aldrich). The reagents were used as received without any additional purification. All solutions were degassed with pure nitrogen (99.9%) for 15 min prior to electrochemical measurements. The water was deionised using an Elgastat prima 3 reverse osmosis unit (Elga Ltd., High Wycombe, UK). 2.2. Instrumentation The electrochemical measurements were recorded using an SP-50 potentiostat (Bio-Logic, Claix, France) and a PalmSens 3 potentiostat (Alvatek, Tetbury, UK). Metabolites were identified by ESI/MS using a LCQ Classic ion trap mass spectrometer (Thermo Scientific, Hemel Hempstead, UK). A syringe pump MD-1002 pursued from Basi (West Lafagette, IN, USA) was used to introduce sample and GSH solutions into the chip. 2.3. Screen-Printed Electrodes (SPEs) Unmodified SPEs (DropSens) were used for the mimicry of phase II metabolism. A three electrode configuration (Figure 1), composed of a carbon working (4 mm diameter) electrode, a carbon counter electrode, and a silver reference electrode. The dimensions of the sensor were 3.4 cm × 1.0 cm × 0.05 cm (length × width × height). An edge connector DRP-DSC (DropSens) served as an interface between the SPE and potentiostat.

Sensors 2016, 16, 1418 Sensors 2016, 16, 1418

3 of 15 3 of 15

Figure electrode(SPE) (SPE)with with a three electrode configuration. Figure1.1.Unmodified Unmodifiedscreen-printed screen-printed electrode a three electrode configuration.

2.4. Microfluidic Microfluidic Device Device with with an an SPE SPE 2.4. An SPE SPE was was integrated integrated with with aa serpentine serpentine reaction reaction channel channel into into aa polymeric polymeric chip chip for for monitoring monitoring An phase IIII metabolism. metabolism. The The working working principle principle involved involved the the electrochemical electrochemical generation generation of of the the phase phase II phase intermediates within the SPE cell and their subsequent conjugation with GSH in a serpentine intermediates within the SPE cell and their subsequent conjugation with GSH in a serpentine reaction channel. channel. reaction 2.4.1. Time Time of of Electrolysis Electrolysis in in the the SPE SPECell Cell 2.4.1. Investigations on on SPEs SPEs suggested suggested that that 0.5 0.5 min min of of control control potential potential electrolysis electrolysis (CPE) (CPE) was was sufficient sufficient Investigations for GSH-adduct GSH-adductdetection detectioninin [21]. Thus equal or longer durations of electrolysis are required for MSMS [21]. Thus equal or longer durations of electrolysis are required within within the SPE cell. Considering the delicate structure of the integrated chip and the capabilities in MS, the SPE cell. Considering the delicate structure of the integrated chip and the capabilities in MS, the −1 −1 − 1 − 1 the selected range from 2.5·min µL·minto to Thevolume volumeof ofSPE SPEcell cell was was 32 32 µL µL to to selected flowflow raterate range waswas from 2.5 µL 5050 µLµL·min ·min . .The ensure longer longer durations of electrolysis ensure electrolysis (t (t >>0.5 0.5min) min)atatthe theselected selectedflow flowrate raterange, range,Table Table1.1. Table 1. 1. Effect Effect of of flow flow rate rate in inmetabolic metabolicsynthesis synthesisand andGSH GSHconjugation. conjugation. Table Flow Rate Flow Rate −1) (µL·min −1 ) (µL·min 2.5 2.5 5 5 10 10 20 20 30 30 60 60

Time of Electrolysis Time of Electrolysis (min) (min) 12.8 12.8 6.4 6.4 3.2 3.2 1.6 1.6 1.1 1.1 0.6 0.6

Combined Flow Rate Combined Flow Rate −1 (µL··min (µL min−)1 ) 5.0 5.0 10 10 20 20 40 40 60 60 100 100

Time of GSH Conjugation Time of GSH (min) Conjugation (min) 4.8 4.8 2.4 2.4 1.2 1.2 0.6 0.6 0.4 0.4 0.2 0.2

2.4.2. Time Time of of GSH GSH Conjugation Conjugation in in aa Serpentine SerpentineChannel Channel 2.4.2. In aaserpentine serpentinereaction reaction channel, laminar flows of sample test sample andare GSH are expected to In channel, thethe laminar flows of test and GSH expected to diffuse diffuse and then to react covalently. The width of the serpentine channel is a key parameter for and then to react covalently. The width of the serpentine channel is a key parameter for determining determining the time of diffusion mixing. As shown in Equation (1), at a awidth of 0.01 cm, atofa the time of diffusion mixing. As shown in Equation (1), a width of 0.01 cm, diffusion coefficient −6 2 diffusion of 3.0 × a10diffusion cm /s, provided diffusion mixing of 0.28times min. for TheGSH calculated times 3.0 × 10−6coefficient cm2 /s, provided mixing of a0.28 min. The calculated conjugation, −1 to 100 µL·min−1, are shown in − 1 − 1 for GSH conjugation, at the combined flow rate range of 5 µL·min at the combined flow rate range of 5 µL·min to 100 µL·min , are shown in Table 1. At all flow rates, Table 1. At all flow rates, the duration of conjugation is longerfor ordiffusion close to mixing: the time required for the duration of conjugation is longer or close to the time required diffusion mixing: (width)2 t= (1) ( ℎ) (1) = 2x (di f f usion coe f f icient) 2 ( ) 2.4.3. Fabrication of Chip

2.4.3.The Fabrication Chip chip wasoffabricated from three polycarbonate wafers of 3.3 cm × 1.7 cm (length × width), bonded together with silicone and double tape (Figure 32 µL 20 µL serpentine The chip was fabricated from threeadhesive polycarbonate wafers2).ofA3.3 cm SPE × 1.7cell, cma(length × width), reaction channel of 120 mm × 0.01 cm × 20 µm (length × width × depth) and a side channel of bonded together with silicone and double adhesive tape (Figure 2). A 32 µL SPE cell, a 20 µL serpentine reaction channel of 120 mm × 0.01 cm × 20 µm (length × width × depth) and a side channel of 80 mm × 0.04 cm × 20 µm (length × width × depth) were drilled into the middle layer (Figure 2B)

Sensors 2016, 16, 1418

4 of 15

Sensors 2016, 16, 1418

4 of 15

80 mm × 0.04 cm × 20 µm (length × width × depth) were drilled into the middle layer (Figure 2B) by by a computer numerical control (CNC) machine (Daltron, Hampshire, USA). inlets a computer numerical control (CNC) machine (Daltron, NewNew Hampshire, NH,NH, USA). The The inlets and and outlet were drilled in wafer (Figure 2A), and wafer (Figure 2C) served as a base for the chip. The outlet were drilled in wafer (Figure 2A), and wafer (Figure 2C) served as a base for the chip. The full full design of proposed the proposed is shown in Figure design of the chipchip is shown in Figure 2D. 2D.

Figure 2. 2. Fabrication Fabrication process. process. (A) (A) Top Top wafer wafer with with inlets inlets and and outlet; (B) middle wafer containing containing the the Figure outlet; (B) middle wafer 32 µL SPE cell and 20 µL serpentine reaction channel; (C) base layer and (D) full design of the chip. 32 µL SPE cell and 20 µL serpentine reaction channel; (C) base layer and (D) full design of the chip.

2.5. Optimization Optimization of of SPEs SPEs 2.5. Prior to to microfluidic microfluidic investigations, investigations, the the conditions conditions for for DOPA DOPA metabolism metabolism were were optimized optimized on on Prior bare SPEs. A new SPE was used for each experiment, considering the disposable nature of the bare SPEs. A new SPE was used for each experiment, considering the disposable nature of the sensor. For For both both kinetic kinetic and and synthetic synthetic applications, applications, 50 50 µL µL of of sample sample solution solution were were loaded loaded into into the the sensor. three electrode electrode configuration. Oxidation products three configuration. Oxidation products were were subsequently subsequently detected detected by by offline offline ESI/MS. ESI/MS. 2.6. Chip/ESI-MS Chip/ESI-MS 2.6. The integrated integrated chip chip was via transfer transfer capillaries capillaries (Figure (Figure 3). 3). DOPA DOPA The was coupled coupled online online to to ESI/MS ESI/MS via −5 M) in 0.1 M ammonium acetate was infused into the SPE cell at a constant flow rate of (2.5 × 10 −5 (2.5 × 10 M) in 0.1 M ammonium acetate was infused into the SPE cell at a constant flow rate of −1 and electrolyzed for 6.4 min. GSH (5 × 10−5 M) in 0.1 M of ammonium acetate was also ·min−1 55 µL µL·min and electrolyzed for 6.4 min. GSH (5 × 10−5 M) in 0.1 M of ammonium acetate was also infused at the same flow flow rate and conjugated covalently with with the the infused at the same rate towards towards the the serpentine serpentine channel channel and conjugated covalently oxidation products for 2.4 min. It should be emphasized that a new chip was used for each electrolysis. oxidation products for 2.4 min. It should be emphasized that a new chip was used for each

electrolysis.

Sensors 2016, 2016, 16, 16, 1418 1418 Sensors Sensors 2016, 16, 1418

of 15 15 55 of 5 of 15

Figure 3. Online coupling of chip with electrospray ionization/mass spectrometry (ESI/MS). Figure3.3.Online Onlinecoupling couplingof ofchip chipwith withelectrospray electrosprayionization/mass ionization/mass spectrometry spectrometry (ESI/MS). (ESI/MS). Figure

3. Validation of SPEs 3. 3. Validation Validationof ofSPEs SPEs 3.1. Electrochemical Behavior of DOPA 3.1. Electrochemical Electrochemical Behavior of DOPA Cyclic voltammetry was used to study the electrochemical behaviour of DOPA on an SPE. The Cyclic voltammetry thethe electrochemical behaviour of DOPA on an The voltammetrywas wasused usedtotostudy study electrochemical behaviour of DOPA onSPE. an SPE. voltammetric response of DOPA (2.5 × 10−5 M)−in 0.1 M phosphate buffer (pH 5) was recorded over 5 M) voltammetric response of DOPA (2.5 (2.5 × 10−5 0.1 in M0.1 phosphate bufferbuffer (pH 5)(pH was5)recorded over The voltammetric response of DOPA ×M) 10 in M phosphate was recorded the potential range of 0.3–0.6 V and at 0.155 V·s−1 scan rates. A redox pair was obtained with an the rangerange of 0.3–0.6 V and at 0.155 V·s−1V·scan rates.rates. A redox pairpair waswas obtained with an overpotential the potential of 0.3–0.6 V and at 0.155 s−1 scan A redox obtained with oxidative peak at 0.128 V and a reductive peak at −0.08 V (Figure 4). The separation peak potential oxidative peak at 0.128 VV and a reductive an oxidative peak at 0.128 and a reductivepeak peakatat−0.08 −0.08VV(Figure (Figure4). 4).The Theseparation separationpeak peak potential potential (ΔΕρ) was found to be 0.208 V, indicating a quasi-reversible process [23]. (ΔΕρ) indicating aa quasi-reversible quasi-reversible process process[23]. [23]. (∆E$) was found to be 0.208 V, V, indicating

−5 M). Figure Figure 4. 4. Cyclic Cyclic Voltammograms Voltammogramsof of(A) (A)Blank Blankelectrolyte electrolytesolution, solution,(B) (B)DOPA DOPA(2.5 (2.5×× 10 10−5 M). Figure 4. Cyclic Voltammograms of (A) Blank electrolyte solution, (B) DOPA (2.5 × 10−5 M).

Effect of of pH pH 3.2. Effect 3.2. Effect of pH Cyclic voltammograms voltammograms of of DOPA (2.5 10−5−5M) M)were wererecorded recordedin in different different pH pH values values (0.1 (0.1 M Cyclic (2.5 × × 10 Cyclic voltammograms of DOPA (2.5 × 10−5 M)−1were recorded in different pH values (0.1 M −1 ·s scan scanrates. rates. phosphate buffers at pH 1, 2, 3, 4, 5, 6, 7, 8) at 0.1 V V·s phosphate buffers at pH 1, 2, 3, 4, 5, 6, 7, 8) at 0.1 V·s−1 scan rates. 3.2.1. Participation Participation of of Protons Protons 3.2.1. 3.2.1. Participation of Protons A linear linear relationship relationship was was obtained obtained between and pH, pH, suggesting suggesting the the A between the the anode anode potential potential (E (Epa pa)) and A linear relationship was obtained between the anode potential (Epa) and pH, suggesting the participation of of protons protons during during the the electrochemical electrochemical process process (Figure (Figure 5). 5). The The anode anode potential potential shifted shifted to to participation participation of protons during the electrochemical process (Figure 5). The anode potential shifted to − 1 was more negative values as pH increased. The obtained slope 0.0514 V · pH close to the Nernestian −1 more negative values as pH increased. The obtained slope 0.0514 V·pH−1 was close to the Nernestian more negative values as pH increased. The obtained slope 0.0514 V·pH was close to the Nernestian slope of of 0.059 0.059 V·pH V·pH−1−1[24], [24],implying implyingthe theinvolvement involvementofofthe thesame samenumber numberofofprotons protonsand andelectrons. electrons. slope slope of 0.059 V·pH−1 [24], implying the involvement of the same number of protons and electrons.

Sensors 2016, 16, 1418 Sensors Sensors 2016, 2016, 16, 16, 1418 1418

6 of 15 66 of of 15 15

Figure 5. ) with symbol size is Figure of anode potential (E pH. Not visible error bars: when the symbol size Figure 5. 5. Dependence Dependence of ofanode anodepotential potential(E (Epapa pa)) with withpH. pH.Not Notvisible visibleerror errorbars: bars:when whenthe the symbol size larger than the error bar. is larger than the error bar. is larger than the error bar.

3.2.2. 3.2.2. Stability Stability Studies Studies /Ipc )close closeto unity[25] [25]is consideredas asaaacriterion criterionof ofproduct productstability. stability. pa )) close A pa/I /I pcpc totounity unity [25] isisconsidered considered as criterion of product stability. A peak peak current current ratio ratio (I (Ipa (Ipc currents have have the magnitude since since the The The anode anode (I (Ipa pa))) and and cathode cathode (I pc))) currents currents have the same same magnitude the electrogenerated electrogenerated pa pc intermediate shown in Figure 6, the product was intermediate is is not not participating participating in in further further reactions. reactions. As As shown in Figure 6, the product was stable As shown stable over at pH with an IIpa /I over acidic The highest /I 1.08. In more acidic conditions. conditions. The highest stability stability was was obtained obtained at at pH pH555with withan anIpa pa /Ipc pc of of 1.08. In more pc alkaline pa alkaline conditions, conditions, the /Ipc startedto increase,suggesting suggestinginstability instabilityand and further further reactions the IIpa pa/I /I pcpcstarted started totoincrease, increase, suggesting instability and further reactions with with molecules solutions. At pH 8 the greatest instability was seen with an I pa /I pc of 3.2. molecules in At pH /I of 3.2. in solutions. 8 the greatest instability was seen with an Ipa pa/Ipcpc of 3.2.

Figure Figure 6. 6. Product Product stability stability as as aaa function function of of pH. pH. Not Not visible visible error error bars: bars: when when the symbol size is larger Figure 6. Product stability as function of pH. Not visible error bars: when the the symbol symbol size size is is larger larger than the error bar. than the error bar. than the error bar.

3.2.3. 3.2.3. Current Current 3.2.3. Current The the is directly The concentration concentration of of the electrogenerated electrogenerated product product directly proportional proportional to to the the obtained obtained The concentration of the electrogenerated product is is directly proportional to the obtained current [26]. Herein, the highest signals for both IIpa IIpc observed at pH 44 (Figure 7). For pH current [26]. Herein, the highest signals for both pa and and pc were were observed at pH (Figure 7). current [26]. Herein, the highest signals for both Ipa and Ipc were observed at pH 4 (Figure 7). For For pH pH values of 1–3 and 5–8, signals were low, implying less product formation. The obtained peak values of 1–3 and 5–8, signals were low, implying less product formation. The obtained peak values of 1–3 and 5–8, signals were low, implying less product formation. The obtained peak currents currents were in with stability studies since at pH product was unstable currents wereagreement in good good agreement agreement with the thestudies stability studies since pH 88 the the product wasand unstable were in good with the stability since at pH 8 theatproduct was unstable lower and lower currents were generated. and lower currents were generated. currents were generated. 3.3. Effect of of Scan Scan Rate Rate 3.3. Effect Useful Useful information information regarding regarding the the electrochemical electrochemical reaction reaction mechanisms mechanisms can can be be acquired acquired from from −5 −5 − 5 the potential scan rate. Therefore, the electrochemical response of DOPA (2.5 × 10 M) in 0.1 scan rate. rate. Therefore, the the electrochemical electrochemical response response of of DOPA DOPA(2.5 (2.5×× 10 10 M) the potential scan M) in in 0.1 M M −1 −1 −1 −1 1 to 0.155 −1 by phosphate different scan rates from 0.02 V·s V·s by buffer (pH 5) was investigated atat different scan rates from 0.02 V·sV·to to 0.155 V·s V by·scyclic cyclic phosphate buffer buffer(pH (pH5) 5)was wasinvestigated investigatedat different scan rates from 0.02 s−0.155 voltammetry (Figure 8A). results illustrated quasi-reversible kinetics [27,28]. E pc shifted voltammetry (Figure(Figure 8A). The The results illustrated quasi-reversible kinetics [27,28].[27,28]. Epa pa and andEE E shifted cyclic voltammetry 8A). The results illustrated quasi-reversible kinetics Epc papcand to to more more positive positive and and more more negative negative values, values, respectively respectively (Figure (Figure 8B). 8B). Also, Also, the the Δ Δερ ερ increased increased with with the the

Sensors 2016, 2016, 16, 16, 1418 1418 Sensors

77 of of 15 15

increase in scan rate (Figure 8C) while a linear relationship was obtained between the Ip and square Sensors 2016, 16, 1418 7 of 15 root (SQRT) of scan rate and (Figure 8D). shifted to more positive more negative values, respectively (Figure 8B). Also, the ∆ε$ increased with increase the increase in rate scan(Figure rate (Figure 8C)awhile linear relationship was between obtainedthe between the Ip and in scan 8C) while linear arelationship was obtained Ip and square root (SQRT) of scan rate (Figure 8D). 8D). square root (SQRT) of scan rate (Figure

Figure 7. Effect of current a functionof pH.(A) (A) Anode Anode (B) Cathode current. Not Not visible as as function ofofpH. pH. (A) Anodecurrent. current. (B) Cathode current. Figure 7. Effect of current as aa function current. (B) Cathode current. visible error bars: when the symbol size is larger than the error bar. bar. error bars: when the symbol size is larger than the error bar.

Figure 8. Scan rate investigations. (A) Cyclic voltammograms of DOPA from 0.02 V·s−1 to 0.155 V·s−1; (B) EPc and Epa variation with scan rate; (C) Variation of Δερ with scan rate and (D) Linear relationship of Ipa and Ipc with square root (SQRT) scan rate; Not visible error bars: when the symbol size is larger ratebar. investigations. (A) (A)Cyclic Cyclicvoltammograms voltammogramsofofDOPA DOPAfrom from 0.02 to0.155 0.155VV·s Figure 8. the Scan rate investigations. 0.02 V·V·s s−1−1to ·s−−11;; than error

(B) EPc Pc and and EEpa variationwith withscan scanrate; rate; (C) (C) Variation Variation of Δ ∆ερ withscan scanrate rateand and(D) (D)Linear Linear relationship relationship pavariation ε$with pa and of Ipa and IIpcpcwith withsquare squareroot root(SQRT) (SQRT)scan scanrate; rate; Not Not visible visible error error bars: bars: when when the the symbol symbol size is larger bar. than the error bar.

Sensors Sensors 2016, 2016, 16, 1418

88 of of 15 15

The transport mode of electroactive material towards the electrode surface is determined by the The transportcurrent mode of electroactive material towards theSlopes electrode is determined by the plot of logarithm (Ip) versus logarithm scan rate (V). closesurface to 1 suggest an adsorption plot of logarithm current (Ip) versus logarithm scan rate (V). Slopes close to 1 suggest an adsorption process, below 0.5 a diffusion process and values between 0.5 and 1, a mixture of both diffusion and process, below 0.5 a A diffusion process and was values between 0.5 and a mixtureofofIpa both adsorption [27–29]. linear relationship found between the 1, logarithm anddiffusion Ipc with and the adsorption [27–29]. A linear relationship was found between the logarithm of I and I with pa pc logarithm of scan rate (V), Equations (2) and (3), and the obtained slopes were 0.52 and the 0.6 logarithm of scan rate (V),anEquations (2) and (3), and the obtained slopes were 0.52 and 0.6 respectively, respectively, illustrating adsorption-diffusion controlled process: illustrating an adsorption-diffusion controlled process: log = 0.52 log − 2.5, R = 0.994 (2) 2 log I pa = 0.52 log V − 2.5, R = 0.994 (2) log = 0.60 log − 2.4, R = 0.995 (3) pc a=good 0.60 agreement log V − 2.4,with R2 =previous 0.995 published data of DOPA (3) These particular findings log are Iin in solid These electrodes [30–32]. As concluded, DOPA was oxidized electrochemically on of theDOPA surface of a particular findings are in a good agreement with previous published data in solid bare SPE into the corresponding dopaminoquinone, via the transfer of two electrons and electrodes [30–32]. As concluded, DOPA was oxidized electrochemically on the surface of a bare two SPE protons. into the corresponding dopaminoquinone, via the transfer of two electrons and two protons.

3.4. 3.4. Cyclic Cyclic Voltammetry Voltammetry in in the the Presence Presence of GSH Oxidationof ofDOPA DOPA(2.5 (2.5×× 10 10−−55 M) Oxidation M)in in the the presence presence of of various various GSH GSH concentrations concentrations from from 00 M M to to − 4 −4 2.5 was studied byby cyclic voltammetry. AsAs seen in in Figure 9, the voltammetric response of 2.5 ××10 10 MM was studied cyclic voltammetry. seen Figure 9, the voltammetric response DOPA changed after the the additions of GSH, illustrating the formation of a GSH-adduct and, as a of DOPA changed after additions of GSH, illustrating the formation of a GSH-adduct and, consequence, the trapping of dopaminoquinone. Ipc Ireduced with the as a consequence, the trapping of dopaminoquinone. with theincrease increaseofofGSH, GSH, reaching reaching pc reduced almost the highest highest GSH GSH concentration concentration (2.5 (2.5 ×× 10 10−−44 M). almost an irreversible system at the M). The The particular particular electrochemical responseexplained explained the consumption of dopaminoquinone through GSH electrochemical response the consumption of dopaminoquinone through GSH conjugation. conjugation. pa increased andtoEmore pa shifted to more positive values, further In addition, IIn increasedIand Epa shifted positive values, further implying theimplying formationthe of pa addition, formation of a GSH-adduct metabolite, which was also electroactive. a GSH-adduct metabolite, which was also electroactive.

−5 −5 as a function of GSH addition. (A) DOPA (2.5 × Figure × 10 Figure 9. 9. Cyclic Cyclicvoltammetry voltammetryofofDOPA DOPA(2.5 (2.5 × 10M) M) as a function of GSH addition. (A) DOPA −5 −5 −5 M GSH. − − 5 − 5 5 MIn M). (B) In the presence of 5 × 10 M GSH. (C) In the(C) presence 1.24 × 10 the 10 (2.5 × 10 M). (B) In the presence of 5 × 10 M GSH. In the of presence of 1.24 × 10 (D) GSH. −4 M GSH. (E) −5 M). − 4 − 5 GSH (5 × 10 presence of 2.5 × 10 (D) In the presence of 2.5 × 10 M GSH. (E) GSH (5 × 10 M).

3.5. 3.5. Potential Potential Optimization Optimization The The oxidative oxidative potential potential was was optimized optimized from from 0.17 0.17 V V to to 0.52 0.52 V, V, over over aa mass mass transport transport limit limit with with increasing steps of 0.05 V, as shown in Figure 10. The selection of potentials was based on the increasing steps of 0.05 V, as shown in Figure 10. The selection of potentials was based on the obtained −5 M of GSH. The DOPA-GSH adduct obtained voltammograms in in Figure 9, in theof presence 5 of × 10 5M voltammograms in Figure 9, the presence 5 × 10−of GSH. The DOPA-GSH adduct (m/z 459) (m/z 459) was detectable at all potentials and the parent compound optimum was detectable at all potentials and the parent compound was depleted.was Thedepleted. optimumThe potential was potential was found to be 0.27 V, since this is when higher yields were obtained, as indicated by the found to be 0.27 V, since this is when higher yields were obtained, as indicated by the intensities intensities from MS. from MS.

Sensors 2016, 16, 1418 Sensors Sensors 2016, 2016, 16, 16, 1418 1418

of 15 999 of of 15 15

Figure 10. 10. Potential Potential optimization optimization for for DOPA DOPA electrochemical Figure Figure 10. Potential optimization for DOPA electrochemical metabolism metabolism on on aa bare bare SPE. SPE.

Exhaustive Electrolysis 3.6. 3.6. Exhaustive Electrolysis yields, thethe optimized potential waswas heldheld constant for 30 min In order to generate higher metabolite yields, optimized potential constant for 30 In order orderto togenerate generatehigher highermetabolite metabolite yields, the optimized potential was held constant for 30 over the same conditions. At theAt end ofend electrolysis, a brownish colour change obtained in solution, min the conditions. the of aa brownish colour change was in min over over the same same conditions. At the end of electrolysis, electrolysis, brownish colour was change was obtained obtained in which was not seen potential or during cyclic voltammetry. The brownish was solution, which was seen potential or cyclic The brownish solution, which wasinnot not seen in inoptimization potential optimization optimization or during during cyclic voltammetry. voltammetry. Thecolour brownish attributed to attributed aattributed new metabolite with metabolite m/z of 301, which wasof found prolonged electrolysis. As higher colour to with 301, which was after colour was was to aa new new metabolite with m/z m/z of 301,after which was found found after prolonged prolonged electrolysis. As of were this further reactions yields of dopaminoquinone produced, this allowed further reactions to occur. Intermediates electrolysis. As higher higher yields yieldswere of dopaminoquinone dopaminoquinone were produced, produced, this allowed allowed further reactions to to occur. Intermediates with dark colours have been reported in literature, as a part of DOPA phase with dark colours have been reported in literature, as a part of DOPA phase 1 metabolism [22]. occur. Intermediates with dark colours have been reported in literature, as a part of DOPA phase 11 metabolism [22]. The disulphide also confirming of The glutathione (GSSG) was also (GSSG) obtained, confirming the ability of SPEsthe to ability generate metabolism [22].disulphide The glutathione glutathione disulphide (GSSG) was was also obtained, obtained, confirming the ability of SPEs multiple metabolites extended electrolysis. The multiple metabolites in extended times in of The of expected DOPA-GSH adduct DOPA-GSH was formed SPEs to to generate generate multiple metabolites in electrolysis. extended times times of electrolysis. The expected expected DOPA-GSH adduct formed to the metabolite in lowerwas intensities, compared to the newcompared metabolite of 301) (Figure (m/z 11). adduct was formed in in lower lower intensities, intensities, compared to(m/z the new new metabolite (m/z of of 301) 301) (Figure (Figure 11). 11).

Figure 11. (A) Colour change electrolysis; (B) Mass of 11.(A) (A) Colour change after electrolysis; (B)spectrum Mass spectrum spectrum of DOPA DOPA electrochemical electrochemical Figure 11. Colour change afterafter electrolysis; (B) Mass of DOPA electrochemical metabolism metabolism at 0.27 V. metabolism at 0.27 V. at 0.27 V.

3.7. Metabolism 3.7. Proposed Proposed Reaction Reaction Mechanism Mechanism for for Electrochemical Electrochemical Metabolism Metabolism 3.7. Proposed Reaction Mechanism for Electrochemical According to m/z, According to to the the obtained obtained m/z, m/z, at at the optimized potential of 0.27 V, DOPA was oxidized oxidized via via the the According the obtained atthe theoptimized optimizedpotential potentialof of0.27 0.27V, V, DOPA DOPA was was oxidized via the transfer two electrons and two reactive Scheme transfer ofof oftwo two electrons and protons two protons protons into the thedopaminoquinone, reactive dopaminoquinone, dopaminoquinone, Scheme 1. 1. transfer electrons and two into the into reactive Scheme 1. Subsequently, Subsequently, the phase I metabolite followed three main pathways leading to DOPA-GSH adduct, Subsequently, the phase I metabolite followed three main pathways leading to DOPA-GSH adduct, the phase I metabolite followed three main pathways leading to DOPA-GSH adduct, GSSG, and DOPA GSSG, and DOPA dimer. The pathway involved covalent of GSSG, The andfirst DOPA dimer. Thethefirst first pathway involved the covalent conjugation conjugation of the the dimer. pathway involved covalent conjugation of thethe electrogenerated dopaminoquinone electrogenerated dopaminoquinone with GSH, forming the expected phase II DOPA-GSH adduct electrogenerated dopaminoquinone with GSH, forming the expected phase II DOPA-GSH adduct with GSH, forming the expected phase II DOPA-GSH adduct (m/z 459). In the second pathway, a series (m/z 459). pathway, aa series of steps were taken, leading (m/z 459). In In the the second second pathway,steps series of electrochemical electrochemical and chemical steps were taken, leading of electrochemical and chemical were taken, leading and to a chemical dimer (m/z 301) with a brownish to a dimer (m/z 301) with a brownish colour. In particular, dopaminoquinone was converted to a dimer (m/z 301) dopaminoquinone with a brownish colour. In particular, dopaminoquinone was converted colour. In particular, was converted chemically into leucodopaminochrome chemically into leucodopaminochrome via intramolecular cyclization, and chemically into leucodopaminochrome via intramolecular cyclization, and subsequently via intramolecular cyclization, and subsequently electrochemically oxidised throughsubsequently the transfer electrochemically oxidised through the transfer of two electrons and two protons to electrochemically oxidised through thecorresponding transfer of two electrons and two protons process to the the of two electrons and two protons to the aminochrome. A polymerisation corresponding aminochrome. A process between parent DOPA corresponding aminochrome. A polymerisation polymerisation process between aminochrome and parent DOPA between aminochrome and parent DOPA compound generated theaminochrome phase I DOPAand dimer metabolite. compound generated the phase II DOPA The side were compound generated thein phase DOPA dimer dimer metabolite. The side products products were in in an an agreement agreement The side products were an agreement withmetabolite. the relevant literature in electrochemical [31] and with the relevant literature in electrochemical [31] and conventional metabolism [22,33]. Finally, with the relevant literature in electrochemical [31] and conventional metabolism [22,33]. Finally, the the

Sensors 2016, 16, 1418

10 of 15

Sensors 2016, 16, 1418

10 of 15

conventional metabolism [22,33]. Finally, the third pathway involved a catalytic process among the third pathway involved a catalytic among theofdopaminoquinone GSH, of causing the dopaminoquinone with GSH, causingprocess the regeneration the DOPA and the with formation the GSSG regeneration of the DOPA and the formation of the GSSG (m/z 612). (m/z 612).

Scheme 1. Proposed reaction for the electrochemical metabolism of DOPA at 0.27 V. Scheme 1. Proposed reaction for the electrochemical metabolism of DOPA at 0.27 V.

4. Microfluidic Device Device with with an an SPE SPE 4. Microfluidic 4.1. 4.1. Functionality Functionality of of Chip Chip The functionality of the chip in generating phase II GSH-adducts was initially investigated using The functionality of the chip in generating phase II GSH-adducts was initially investigated APAP, considering its simple and well-studied electrochemical metabolism. using APAP, considering its simple and well-studied electrochemical metabolism. 4.1.1. Cyclic Voltammetry on Chip 4.1.1. Cyclic Voltammetry on Chip During validation studies, cyclic voltammetry was used to determine the potential range for During validation studies, cyclic voltammetry was used to determine the potential range for CPE. Cyclic voltammograms of APAP (1 × 10−−33 M) in 0.1 M ammonium acetate were recorded on the CPE. Cyclic voltammograms of APAP (1 × 10 M) in 0.1 M ammonium acetate were recorded on the 1 to 50 µL chip at stop flow conditions and at various flow rates from 2.5 µL·min− ·min−1 . As shown chip at stop flow conditions and at various flow rates from 2.5 µL·min−1 to 50 µL·min−1. As shown in in Figure 12, a well-defined quasi-reversible pair was obtained, illustrating the capability of chip to Figure 12, a well-defined quasi-reversible pair was obtained, illustrating the capability of chip to determine a potential range. Shifts in potentials have been obtained mainly in higher flow rates due to determine a potential range. Shifts in potentials have been obtained mainly in higher flow rates due an enhanced material adsorption. to an enhanced material adsorption.

Sensors 2016, 16, 1418 Sensors 2016, 16, 1418

11 of 15 11 of 15

−3 M). (A) Off chip; (B) Stop flow; Figure 12. 12. Cyclic Cyclicvoltammetry voltammetryininthe thepresence presenceofofAPAP APAP(1(1× × 10M). Figure 10−3 (A) Off chip; (B) Stop flow; (C) −1 −1 −1 ; (F) 50 µL −1 . (C)µL·min 2.5 µL·−1 min 5 µL−1·min 20 µL −1;·min −1. ·min ; (D) ;5(D) µL·min ; (E) 20; (E) µL·min (F) 50 µL·min 2.5

4.1.2. Electrolysis Electrolysis and and GSH GSH Conjugation Conjugation on Chip 4.1.2. The functionality functionality of of the the chip chip in in generating generating phase phase II II GSH-adducts GSH-adducts was was investigated investigated offline, offline, using using The 5 M) in 0.1 M ammonium acetate (pH 7.4) was infused into −5 − APAP as aa drug drug probe. probe. APAP APAP(2.5 (2.5××1010 M) in 0.1 M ammonium acetate (pH 7.4) was infused into the the surface of SPE electrolysed min, Thepotential potentialwas wasnot notoptimized; optimized;however, however, itit surface of SPE andand electrolysed for for 6.46.4 min, at at 0.50.5 V.V.The oxidative peak in was operated over over aa mass mass transport transportlimit, limit,since sinceititwas was0.1 0.1VVhigher higherthan thanthe theobtained obtained oxidative peak cyclic voltammetry. Subsequently, the effluent was collected and analysed ESI/MS.by The APAP-GSH in cyclic voltammetry. Subsequently, the effluent was collected and by analysed ESI/MS. The adduct (m/z adduct 456.84) was successfully, seen in Figure confirming the13, complete mimicry APAP-GSH (m/zgenerated 456.84) was generatedassuccessfully, as13, seen in Figure confirming the of the phase I (dehydrogenation) and phase II (GSH conjugation) reactions. complete mimicry of the phase I (dehydrogenation) and phase II (GSH conjugation) reactions. 4.1.3. Online Online Coupling Coupling 4.1.3. CPE in in the the potential potential range range of of 0.4 0.4 V V to to 0.75 0.75 V V was was recorded, recorded, in in 0.05 0.05 V V steps steps to to determine determine the the CPE optimum potential for APAP-GSH formation. The conditions applied were the same as in offline optimum potential for APAP-GSH formation. The conditions applied were the same as in offline coupling. The Theoxidation oxidationpotential potentialwas was0.35 0.35V, V, as as determined determined by by cyclic cyclic voltammetry; voltammetry; however, however, the the coupling. investigated potential range was at least 0.05 V higher to ensure minimum interference from kinetics. investigated potential range was at least 0.05 V higher to ensure minimum interference from At the oxidative the electrochemical reaction is not completed and a strong is kinetics. At the peak oxidative peak the electrochemical reaction is notyet completed yet competition and a strong created between the electron transfer and mass transport. Thus, maximum metabolite yields can be competition is created between the electron transfer and mass transport. Thus, maximum metabolite obtained higher potentials, where the electrochemical leads gradually depletion and the yields canatbe obtained at higher potentials, where the reaction electrochemical reactiontoleads gradually to system is dominated by mass transport. As in transport. Figure 14, the adducts andAPAP-GSH GSSG were depletion and the system is dominated byseen mass As APAP-GSH seen in Figure 14, the

adducts and GSSG were formed, suggesting the capability of the chip to mimic two metabolic pathways, simultaneously over flow conditions. The optimum potential for the APAP-GSH adducts was found to be 0.5 V.

Sensors 2016, 16, 1418

12 of 15

formed, suggesting the capability of the chip to mimic two metabolic pathways, simultaneously over Sensors 2016, 12 Sensors 2016, 16, 16, 1418 1418The optimum potential for the APAP-GSH adducts was found to be 0.5 V. 12 of of 15 15 flow conditions.

Figure offline to mass spectrometry (MS). Figure 13. 13. Microfluidic Microfluidic device device coupled coupled offline offline to to mass mass spectrometry spectrometry(MS). (MS). Figure 13. Microfluidic device coupled

Figure electrochemical metabolism. Figure 14. 14. Potential Potential optimizations optimizations of of acetaminophen acetaminophen (APAP) (APAP) electrochemical electrochemicalmetabolism. metabolism.

4.1.4. Optimization for the Generation of DOPA-GSH Adduct 4.1.4. Potential Potential Optimization Optimization for for the the Generation Generation of of DOPA-GSH DOPA-GSHAdduct Adduct 4.1.4. Potential Experiments on an SPE confirmed the capability of the particular sensors to mimic multiple Experiments on on an an SPE SPE confirmed confirmed the the capability capability of of the the particular particular sensors sensors to to mimic mimic multiple multiple Experiments DOPA metabolites. The great instability of dopaminoquinone at pH 7.4 permitted the generation of DOPA metabolites. The great instability of dopaminoquinone at pH 7.4 permitted the generation of DOPA metabolites. The great instability of dopaminoquinone at pH 7.4 permitted the generation of leucodopaminochrome. For this reason, the pH of DOPA solution was decreased to pH 7, aiming for leucodopaminochrome. For was decreased to to pHpH 7, aiming for leucodopaminochrome. Forthis thisreason, reason,the thepH pHofofDOPA DOPAsolution solution was decreased 7, aiming the total conversion of into phase II In the of the the totaltotal conversion of dopaminoquine dopaminoquine into the the required phasephase II adduct. adduct. In addition, addition, the time timethe of for conversion of dopaminoquine intorequired the required II adduct. In addition, electrolysis was also reduced significantly to 6.4 min in order to prevent the dominance of electrolysis was also reduced significantly to 6.4 min in order to prevent the dominance of time of electrolysis was also reduced significantly to 6.4 min in order to prevent the dominance leucodopaminochrome over GSH-adduct. As seen in Figure 15, the DOPA-GSH adduct (m/z 459) leucodopaminochrome over GSH-adduct. adduct (m/z (m/z 459) 459) of leucodopaminochrome over GSH-adduct.AsAsseen seenininFigure Figure15, 15, the the DOPA-GSH DOPA-GSH adduct was successfully formed at all potentials and the generation of the phase I dimer was avoided. was successfully formed at all potentials and the generation of the phase I dimer was avoided. was successfully formed at all potentials and the generation of the phase I dimer was avoided. The maximum DOPA-GSH adduct formation However, the optimum The maximum maximum DOPA-GSH DOPA-GSH adduct adduct formation formation was was obtained obtained at at 0.37 0.37 V. V. However, However, the the optimum optimum The was obtained at 0.37 V. potential on the bare SPE was found to be at 0.27 V. This observation is due to the surface chemistry potential on on the the bare bare SPE SPE was was found found to to be be at at 0.27 0.27 V. V.This This observation observation is is due due to to the the surface surface chemistry chemistry potential having changed over flow conditions, leading to strong adsorption phenomena. The potential having changed over flow conditions, leading to strong adsorption phenomena. The potential having changed over flow conditions, leading to strong adsorption phenomena. The potential difference did not limit the functionality of the chip, since the DOPA-GSH was successfully differencedid didnot not limit functionality the since chip,the since the DOPA-GSH was successfully difference limit the the functionality of theofchip, DOPA-GSH was successfully generated generated the metabolism DOPA was fulfilled. The obtained generated and the aim aim of of monitoring monitoring the phase phase II IIof metabolism of DOPA The wasobtained fulfilled.ion Theintensities obtained and the aimand of monitoring the phase II the metabolism DOPA wasof fulfilled. ion intensities for the DOPA-GSH adduct were increased from 0.17 V to 0.37 V, and then decreased. ion intensities for the DOPA-GSH adduct were increased from 0.17 V to 0.37 V, and then decreased. for the DOPA-GSH adduct were increased from 0.17 V to 0.37 V, and then decreased. Considering that Considering Considering that that SPEs SPEs have have aa limited limited potential potential application application up up to to 11 V, V, as as the the potentials potentials were were approaching approaching the the potential potential limit, limit, the the functionality functionality and and the the electrochemical electrochemical stability stability of of the the chip chip gradually gradually started started to to decline. decline. Additionally, Additionally, the the catalytic catalytic product product of of GSSG GSSG was was also also obtained obtained as as in in the the bare bare SPE; SPE; the the spectra spectra at at the the optimized optimized potential potential of of 0.37 0.37 V V is is shown shown in in Figure Figure 16. 16. The The ion ion intensities intensities

Sensors 2016, 16, 1418

13 of 15

SPEs have a limited potential application up to 1 V, as the potentials were approaching the potential limit, 2016, the functionality and the electrochemical stability of the chip gradually started to decline. Sensors 16, 1418 13 of 15 Additionally, the catalytic product of GSSG was also obtained as in the bare SPE; the spectra 13 at of the Sensors 2016, 16, 1418 15 for the microfluidic compared with investigations on bare SPEs, due optimized potentialdevice of 0.37were V is lower shownwhen in Figure 16. The ionearlier intensities for the microfluidic device forthe the microfluidic device were when compared with earlier on bare SPEs, due to continuous dilution in a serpentine channel well as investigations adsorption were lower when sample compared withlower earlier investigations onasbare SPEs, due to theeffects. continuous sample to the continuous sample dilution in a serpentine channel as well as adsorption effects. dilution in a serpentine channel as well as adsorption effects.

Figure 15. Potential optimization of DOPA electrochemical metabolism within the microfluidic Figure of DOPA electrochemical metabolism withinwithin the microfluidic device. Figure 15. 15.Potential Potentialoptimization optimization of DOPA electrochemical metabolism the microfluidic device. device.

Figure atat the optimized potential of of 0.37 V. V. Figure16. 16.Mass Massspectra spectraasasobtained obtainedfrom fromthe themicrofluidic microfluidicdevice device the optimized potential 0.37 Figure 16. Mass spectra as obtained from the microfluidic device at the optimized potential of 0.37 V.

5. 5. Conclusions Conclusions 5. Conclusions A A disposable disposable microfluidic microfluidic device device with with an an SPE SPE was was developed developed and and successfully successfully mimicked mimicked the the A disposable microfluidic device with an SPE was developed and successfully the phase II metabolism. In particular, an SPE cell was combined with a serpentine channel that phase II metabolism. In particular, an SPE cell was combined with a serpentine channelmimicked that allowed allowed phase II metabolism. In particular, cell was combined with aAserpentine channel that allowed the generation anddetoxification detoxification ofSPE dopaminoquinone, respectively. A metabolic profile was the generation and of an dopaminoquinone, respectively. metabolic profile was obtained the generation and detoxification of dopaminoquinone, respectively. A metabolic profile was obtained in less than 10 min without the use of any tedious cleaning processes. The total cost for each in less than 10 min without the use of any tedious cleaning processes. The total cost for each chip was obtained in less than 10 min without the use of any tedious cleaning processes. The total cost for each chip only 5 GBP, providing a cheaper alternative to the current in vitro The proposed only was 5 GBP, providing a cheaper alternative to the current in vitro studies. Thestudies. proposed chip has the chip was only 5 GBP, providing a cheaper alternative to the current in vitro studies. The proposed chip has the potential to be used as a primary screening tool in pharmaceutical and clinical research, potential to be used as a primary screening tool in pharmaceutical and clinical research, and thereby chipthereby has the complement potential to methodologies. be as a methodologies. primary screening tool in pharmaceutical and clinical research, and theused current complement the current and thereby complement the current methodologies.

Acknowledgments: This study has not received any form of financial support. Acknowledgments: This study has not received any form of financial support. Acknowledgments: This study has not received any form of financial support. Author Contributions: R.V. designed and performed the experiments, the data analysis, created and edited the manuscript and illustration. M.M.N.E.and andperformed N.J.B. designed and fabricated the chip. K.J.W. contributed to the Author Contributions: R.V. designed the experiments, the data analysis, created and edited the study concept andillustration. design, offering supervision together withand technical and administrative support throughout manuscript and M.M.N.E. and N.J.B. designed fabricated the chip. K.J.W. contributed to the the entire duration project. study concept and of design, offering supervision together with technical and administrative support throughout the entire duration of project. Conflicts of Interest: The authors declare no conflict of interest.

Sensors 2016, 16, 1418

14 of 15

Author Contributions: R.V. designed and performed the experiments, the data analysis, created and edited the manuscript and illustration. M.M.N.E. and N.J.B. designed and fabricated the chip. K.J.W. contributed to the study concept and design, offering supervision together with technical and administrative support throughout the entire duration of project. Conflicts of Interest: The authors declare no conflict of interest.

References 1. 2. 3. 4. 5. 6.

7. 8.

9. 10.

11.

12.

13. 14.

15. 16.

17.

18.

Xu, C.; Li, Y.T.C.; Kong, A.N.T. Induction of phase I, II and III drug metabolism/transport by xenobiotics. Arch. Pharm. Res. 2005, 28, 249–268. [CrossRef] [PubMed] Iyanagi, T. Molecular mechanism of phase I and phase II drug-metabolizing enzymes: Implications for detoxification. Int. Rev. Cytol. 2007, 260, 35–112. [PubMed] Jancova, P.; Anzenbacker, P.; Anzenbacherova, E. Phase II metabolizing enzymes. Biomed. Pap. 2010, 154, 103–116. [CrossRef] Hayes, J.D.; Flanagan, J.U.; Jowsey, I.R. Glutathione transferases. Annu. Rev. Pharmacol. Toxicol. 2005, 45, 51–88. [CrossRef] [PubMed] Alfafara, C.G.; Kanda, A.; Shioi, T.; Shimizu, H.; Shioya, S.; Suga, K. Effect of amino-acids on glutathione production by sacharromyces-cerevisiaie. Appl. Microbiol. Biotechnol. 1992, 36, 538–540. [CrossRef] Rajasekaran, M.; Abirami, S.; Chen, C. Effects of Single Nucleotide Polymorphisms on Human N-Acetyltransferase 2 Structure and Dynamics by Molecular Dynamics Simulation. PLoS. ONE 2011, 6, 1–12. [CrossRef] [PubMed] Srivastava, A.; Maggs, J.L.; Antoine, D.J.; Williams, D.P.; Smith, D.A.; Park, B.K. Role of reactive metabolites in drug-induced hepatotoxicity. In Adverse Drug Reactions; Springer: Heidelberg, Germany, 2010; pp. 165–194. Lohmann, W.; Karst, U. Generation and identification of reactive metabolites by electrochemistry and immobilized enzymes coupled on-line to liquid chromatography/mass spectrometry. Anal. Chem. 2007, 79, 6831–6839. [CrossRef] [PubMed] Baumann, A.; Karst, U. Online electrochemistry/mass spectrometry in drug metabolism studies: Principles and applications. Expert Opin. Drug Metab. Toxicol. 2010, 6, 715–731. Van Leeuwen, S.M.; Blankert, B.; Kauffmann, J.M.; Karst, U. Prediction of clozapine metabolism by on-line electrochemistry/liquid chromatography/mass spectrometry. Anal. Bioanal. Chem. 2005, 382, 742–750. [CrossRef] [PubMed] Plattner, S.; Pitter, F.; Brouwer, H.J.; Oberacher, H. Formation and Characterization of Covalent Guanosine Adducts with Electrochemistry—Liquid Chromatography–Mass Spectrometry. J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 2012, 883–884, 198–2014. [CrossRef] [PubMed] Lohmann, W.; Hayen, H.; Karst, U. Covalent Protein Modification by Reactive Drug Metabolites Using Online Electrochemistry/Liquid Chromatography/Mass Spectrometry. Anal. Chem. 2008, 80, 9714–9719. [CrossRef] [PubMed] Regino, M.C.S; Brajiter-Toth, A. An electrochemical cell for on-line electrochemistry/mass spectrometry. Anal. Chem. 1997, 69, 5067–5072. [CrossRef] Getek, T.A.; Korfmacher, W.A.; Mcrae, T.A.; Hinson, J.A. Utility of solution electrochemistry mass spectrometry for investigating the formation and detection of biologically important conjugates of acetaminophen. J. Chromatogr. A 1989, 479, 245–256. [CrossRef] Lohmann, W.; Karst, U. Simulation of detoxification of paracetamol using on-line electrochemistry/liquid chromatography/mass spectrometry. Anal. Bioanal. Chem. 2006, 386, 1701–1708. [CrossRef] [PubMed] Van den Brink, F.T.G.; Buter, L.; Odjik, M.; Olthuis, W.; Karst, U.; van den Berg, A. Mass spectrometric detection of short–lived drug metabolites generated in an electrochemical microfluidic chip. Anal. Chem. 2015, 87, 1527–1535. [CrossRef] [PubMed] Roussel, C.; Dayon, L.; Lion, L.; Rohner, T.C.; Josserand, J.; Ropssier, J.S.; Jensen, H.; Girault, H.H. Generation of Mass Tags by the Inherent Electrochemistry of Electrospray for Protein Mass Spectrometry. J. Am. Soc. Mass Spectrom. 2004, 15, 1767–1779. [CrossRef] [PubMed] Miserere, S.; Ledru, S.; Ruille, N.; Criveau, S.; Boujtita, M.; Bedioui, F. Biocompatible carbon-based screen-printed electrodes for the electrochemical detection of nitric oxide. Electrochem. Commun. 2006, 8, 238–244. [CrossRef]

Sensors 2016, 16, 1418

19. 20. 21.

22.

23. 24.

25.

26. 27. 28. 29.

30. 31. 32.

33.

15 of 15

Mistry, K.K.; Layek, K.; Mahapatra, A.; RoyChaudhurib, C.; Sahab, H. A review on amperometric-type immunosensors based on screen-printed electrodes. Analyst 2014, 139, 2289–2311. [CrossRef] [PubMed] Khaled, E.; Hassan, H.N.A.; Habib, I.H.I.; Metelka, R. Chitosan modified screen-printed carbon electrode for sensitive analysis of heavy metals. Int. J. Electrochem. Sci. 2010, 5, 158–167. Kauffmann, J.M.; Antwerpen, P.V.; Sarakbi, A.; Feier, B.; Tarik, S.; Aydogmus, Z. Utility of Screen Printed Electrodes for in Vitro Metabolic Stability Assays: Application to Acetaminophen and its Thioconjugates. Electroanalysis 2011, 23, 2643–2650. [CrossRef] Monoz, P.; Melendez, C.; Paris, I.; Sequra-Aquilar, J. Molecular and neurochemical mechanism dopamine oxidation to 0-quinone in Parkinson disease pathogenesis. In Toxicity and Autophage in Neurodegenerative Disorders; Springer: Basel, Switzerland, 2015; pp. 205–223. Eklund, J.C.; Alan, B.M.; Alden, J.A.; Gompton, R.G. Perspectives in modern voltammetry: Basic Concepts and Mechanistic Analysis. Adv. Phys. Org. Chem. 1999, 2, 1–120. Mahanthesha, K.R.; Kumara-Swamy, B.E.; Vasantakumar-Pai, K.; Chandra, U.; Sherigara, B.S. Cyclic Voltammetric Investigations of Dopamine at Alizarin Modified Carbon Paste Electrode. Int. J. Electrochem. Sci. 2010, 5, 1962–1971. Nematollahi, D.; Shayani-Jam, H.; Alimoradi, M.; Niroomand, S. Electrochemical oxidation of acetaminophen in aqueous solutions: Kinetic evaluation of hydrolysis, hydroxylation and dimerization processes. Electrochim. Acta 2009, 54, 7407–7515. [CrossRef] Joseph, B.; Justice, J.R. Introduction to in Vivo Voltammetry. In Voltammetry in the Neurosciences; Adams, R.N., Justice, J.B., Eds.; Humana Press: New York, NY, USA, 1987; pp. 15–17. Bagheri, A.; Hosseini, H. Electrochemistry of raloxifene on a glassy carbon electrode and its determination in pharmaceutical formulation and human plasma. Bioelectrochemistry 2012, 88, 164–170. [CrossRef] [PubMed] Hasan, M.M.; Hossain, M.E.; Mamum, M.A.; Ehsan, M.Q. Study of redox behavior of Cd(II) and interaction of Cd(II) with proline in the aqueous medium using cyclic voltammetry. J Saud. Chem. Soc. 2012, 16, 145–151. Ndangli, P.M.; Jijana, A.N.; Olowu, R.N.; Mailu, S.N.; Ngece, F.R.; Williams, A.; Waryo, T.T.; Baker, P.G.L. Impedimetric Response of a Label-Free Genosensor Prepared on a 3-Mercaptoprionic Acid Capped Gallium Selenide Nanocrystal Modified Gold Electrode. Int. J. Electrochem. 2011, 6, 1438–1453. Ahuja, S.; Jespersen, N. Modern Instrumental Analysis; Elsevier: Kidlington, UK, 2006. Ke, N.J.; Lu, S.H.; Cheng, S.H. A strategy for the determination of dopamine at a bare glassy carbon electrode: p-Phenylenediamine as a nucleophile. Electrochem. Commun. 2006, 8, 1514–1520. [CrossRef] Liu, C.Y.; Liu, Z.Y.; Peng, R.; Zhong, Z.C. Quasireversible Process of Dopamine on Copper-Nickel Hydroxide Composite/Nitrogen Doped Graphene/Nafion Modified GCE and Its Electrochemical Application. J. Anal. Methods Chem. 2014, 2014, 1–7. [CrossRef] [PubMed] Spina, M.B.; Cohen, G. Dopamine turnover and glutathione oxidation: Implications for Parkinson disease. Proc. Natl. Acad. Sci. USA 1989, 86, 1398–1400. [CrossRef] [PubMed] © 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license (http://creativecommons.org/licenses/by/4.0/).