Letter pubs.acs.org/ac
Novel Electrochemical Methodology for Activity Estimation of Alkaline Phosphatase Based on Solubility Difference Kosuke Ino,*,† Yusuke Kanno,† Toshiharu Arai,† Kumi Y. Inoue,†,‡ Yasufumi Takahashi,§ Hitoshi Shiku,† and Tomokazu Matsue*,†,§ †
Graduate School of Environmental Studies, Tohoku University, Sendai 980-8579, Japan R&D Center of Excellence of Integrated Microsystems, Tohoku University, Sendai 980-8579, Japan § WPI-Advanced Institute for Materials Research, Tohoku University, Sendai 980-8577, Japan ‡
S Supporting Information *
ABSTRACT: We propose a novel electrochemical detection system for alkaline phosphatase (ALP) activity using the difference in water and oil solubilities between the substrate, ferrocene ethyl phosphate ester (FcEtOPO32−), and the enzymatic product, ferroceneethanol (FcEtOH). In this system, water droplets containing ALP and FcEtOPO32− were placed on a Pt disk microelectrode and surrounded by a mineral oil. By the ALP-catalyzed reaction, FcEtOPO32− was converted to FcEtOH, which was then transferred to the mineral oil from the water droplets with FcEtOPO32− remaining in the water droplets. After partitioning FcEtOH from the water droplets, FcEtOPO32− was detected at the Pt disk microelectrode to estimate the ALP activity. Using this novel system, the ALP activity of embryoid bodies was successfully detected. We believe that the present system will be widely applicable to ALP-based bioassays.
E
droplets via horseradish peroxidase (HPR).8 Thomas et al. reported electrochemical detection of a microbead-based immunoassay in small volumes.9 In their study, p-aminophenol (PAP) generated by alkaline phosphatase (ALP) was detected. ALP is one of the most commonly used labels in biochemical applications due to its high turnover number and low cost. This enzyme hydrolyzes orthophosphoric monoesters into alcohols. The electrochemical detection method has been widely used for detecting ALP activity using both p-aminophenyl phosphate (PAPP, substrate) and PAP (enzymatic product). PAPP and PAP are simultaneously detected even when they are both present in the same solution because their redox potentials are different. As another electroactive substrate, Benoit et al. used ferrocene ethyl phosphate ester (FcEtOPO32−), which leads to ferroceneethanol (FcEtOH) by ALP-catalyzed hydrolysis.10 Since FcEtOPO32− possesses similar oxidation potential as FcEtOH, they used a Nafion film electrode to separate FcEtOH from FcEtOPO32− in the solution.10 In the present study, we describe a novel electrochemical method for detecting ALP activity in droplets using FcEtOPO32− without separation membranes. In the present system, the difference in water and oil solubilities between the substrate, FcEtOPO32−, and the enzymatic product, FcEtOH, was utilized for partition. Water droplets containing ALP and
nzymes have frequently been used as biological markers for bioassays, such as the immunoassay and reporter gene assay because they can selectively convert appropriate substrates to products exhibiting amplified color changes, fluorescent emissions, or electrochemical signals.1−3 Recently, the droplet-based detection method has attracted attention for the construction of more flexible systems that allow handling of numerous samples or reagents as compartmentalized droplets for chemical and biological analyses.4,5 Small droplets are also useful for effective mass transfer and chemical reaction at the droplet interface because the surface-to-volume ratio in droplets decreases as the droplet volume decreases. In order to construct droplet-based bioassay systems, the development of a reliable and sensitive detection method is important. Among the various detection methods, electrochemical methods have been applied for droplet systems because electrochemistry is useful for biochemical assays of ultrasmall environments and a wide range of specialized microcells has been designed for reliable electroanalysis of microliter samples. Han et al. reported an electrochemical method in droplet-based microfluidic devices to detect H2O2.6 In the study, droplets containing catalase and H2O2 were fabricated in oil using a microfluidic device, and H2O2 was electrochemically detected using a Pt electrode in the microfluidic device to measure rapid enzymatic kinetics. Lindsay et al. developed a droplet-based digital magnetofluidics system.7 They used the magnetic force in air for washing, merging, and detecting droplets, and dopamine and glucose were electrochemically detected. Zhang et al. reported amperometric detection of oligonucleotides in © 2012 American Chemical Society
Received: May 24, 2012 Accepted: August 30, 2012 Published: August 30, 2012 7593
dx.doi.org/10.1021/ac301429n | Anal. Chem. 2012, 84, 7593−7598
Analytical Chemistry
Letter
FcEtOPO32− were placed on a microelectrode and surrounded by a mineral oil. FcEtOPO32− was hydrolyzed by ALP to yield FcEtOH, which was then transferred to the mineral oil from the water droplets with FcEtOPO32− remaining in the water droplets. After partitioning FcEtOH from the water droplets, FcEtOPO32− in the water droplets was electrochemically detected to estimate ALP activity in the water droplet. To the best of our knowledge, the present report is the first to demonstrate the partition of an enzyme and an enzymatic product based on the solubility difference in water and oil for electrochemical detection. We applied the novel electrochemical detection system to ALP activity on embryoid bodies (EBs).
■
MATERIALS AND METHODS Reagents. FcEtOPO3Na2 was synthesized by the APRO life Science Institute (Japan) according to a previous study.10 ALP was obtained from Oriental Yeast Co., Ltd. (Japan). The ALP activity was 6760 U/mg protein. FcEtOPO3Na2 and ALP were dissolved into a 50 mM Tris-HCl solution (pH 9.5) containing 2.0 mM MgCl2. To prepare 1 mM FcEtOH, a large amount of ALP (20 μg) was added to 1 mL of FcEtOPO32− solution (1 mM) and incubated in a tube for more than 1 h. The solution after the incubation was used as the 1 mM FcEtOH solution. Device Fabrication. A device with Pt disk microelectrodes was fabricated by a conventional photolithography method.11 Briefly, Ti/Pt was sputtered onto a glass substrate (Matsunami Glass Ind., Ltd., Japan), and SU-8 (2 μm thick, SU-8 2002, Microchem Co., Newton, MA) was spin-coated, followed by fabrication of disk-shaped holes on the substrate (first SU-8 layer). The resulting Pt microdisks (15 μm diameter) were used as working electrodes. Ti/Pt was then sputtered on the SU-8 for a quasi-reference/counter electrode. Then, SU-8 walls (second SU-8 layer) were fabricated to trap droplets (Supplementary Figure 1, Supporting Information). The twoelectrode system comprising the Pt disk microelectrode and the Pt reference/counter electrode was employed for electrochemical analysis of droplets. The first SU-8 layer limited the active working electrodes, and the second SU-8 layer formed walls to trap water droplets easily. Since the SU-8 layer was hydrophobic, the SU-8 walls prevented the water droplets from slipping out of the sensor areas (Supplementary Figure 2, Supporting Information). Electrochemical Detection. The Tris-HCl buffer containing 1 mM FcEtOPO32− and ALP was mixed in a tube, and subsequently, water droplets (2 μL) of the buffer solution were placed on the Pt disk microelectrode and surrounded by a mineral oil (Sigma-Aldrich, St. Louis, MO, USA) within 1 min after mixing. The volume of the mineral oil was 50 μL, 25 times as large as that of the water droplets. No chemical was added to the mineral oil. The droplets were incubated for a while to allow the ALP reaction to proceed in the droplets. The mineral oil was used for partitioning FcEtOH (a reaction product) from FcEtOPO32− (an enzyme substrate) in the water droplets (Figure 1). During the incubation, ALP hydrolyzed FcEtOPO32− to yield FcEtOH, which was transferred to the mineral oil. After partitioning, FcEtOPO32− in the water droplets was electrochemically detected using the device. EB Formation and Detection of Their ALP Activity. Mouse embryonic stem (ES) cells (129/Sv, DS Pharma Biomedical Co., Ltd., Japan) were cultured, and EBs were formed by the hanging drop method according to a previous report.12−14 Briefly, the ES cell suspension was introduced on
Figure 1. Schematic illustration of ALP detection using FcEtOPO32− and the oil/water phase. (A) FcEtOPO32− is hydrolyzed by ALP, and FcEtOH is produced. (B) The droplet including ALP and FcEtOPO32− is mounted on a Pt disk microelectrode and a Pt quasi-reference electrode (Supplementary Figure 1, Supporting Information). The water droplets are then covered with a mineral oil. After the ALP reaction, the enzymatic products (FcEtOH) are transferred to the oil phase. After incubation, FcEtOPO32− is electrochemically detected to estimate ALP activity through the loss of FcEtOPO32−.
the cover of a culture dish to form droplets (20 μL) containing the ES cells (1000 cells). The droplets were then hung from the dish cover, and the ES cells were incubated for 2 days to form EBs. Four EBs were added to the 5 μL Tris-HCl buffer containing FcEtOPO32− and incubated in a tube for an hour at 37 °C under a humidified atmosphere to prevent evaporation. After incubation, 2 μL of the supernatant was collected from the suspension, placed on the device, and surrounded by the mineral oil for electrochemical detection of FcEtOPO32−. After covering the water droplets with the mineral oil, the water droplets were incubated without mixing.
■
RESULTS AND DISCUSSION Electrochemical Detection of FcEtOPO32− and FcEtOH. Figure 2 shows the cyclic voltammograms of 1.0 mM FcEtOPO32− and FcEtOH in water droplets without the mineral oil. The onset potentials for oxidation of FcEtOH and FcEtOPO32− at the Pt disk microelectrode are similar, which suggests that a partition process of FcEtOPO32−/FcEtOH is indispensable for electrochemical enzyme assays utilizing ALP. The plateau current of FcEtOPO32− was approximately twothirds that of FcEtOH. We estimated the diffusion coefficients of FcEtOPO32− and FcEtOH by cyclic voltammetry using a conventional Pt microdisk electrode (20 μm diameter), and they were found to be 3.7 × 10−10 and 6.2 × 10−10 m2/s, respectively at 25 °C. The surface-to-volume ratio in droplets is important because mass transfer from water droplets to a mineral oil occurs 7594
dx.doi.org/10.1021/ac301429n | Anal. Chem. 2012, 84, 7593−7598
Analytical Chemistry
Letter
large enough to negate the consumption of FcEtOH and FcEtOPO32− by electrochemical measurements. If the Pt disk microelectrode (working electrode) is close to the Pt band electrode (quasi-reference electrode), the products at the Pt band electrode may interfere with the reaction at the Pt disk electrode. We investigated the influence of the distance between the two electrodes on the electrochemical responses. Cyclic voltammograms of the 2 μL droplet containing 1 mM FcEtOPO32− at the devices with a 500 or 3000 μm distance showed almost the same behavior (Supplementary Figure 4, Supporting Information). Thus, the influence of electrode arrangement on the electrochemical response was negligible in the time scale of the voltammetric measurements if the distance was 500 μm. Partition of FcEtOPO32− and FcEtOH. Figure 3 shows the electrochemical signals from water droplets containing FcEtOH and FcEtOPO32− surrounded by the mineral oil. The decrease in the signal for FcEtOH with time demonstrates that FcEtOH transferred to the mineral oil from the water droplets (Figure 3A). In contrast, FcEtOPO32− remained in the water droplet (Figure 3B). These results suggest that the substrate, FcEtOPO32−, can be separated from the enzymatic product,
Figure 2. Cyclic voltammograms of the water droplets without the mineral oil. The droplet contained 1 mM FcEtOPO32− or 1 mM FcEtOH (1 mM FcEtOPO32− and 20 μg/mL ALP). The scan rate was 60 mV/s.
through the water/oil interface.15 When using small droplets, the mixing process may be unnecessary during the ALPcatalyzed reaction. In this study, we used 2 μL droplets because this droplet size ensured rapid mass transfer while the size was
Figure 3. (A) Cyclic voltammograms of the 2 μL droplet containing 1 mM FcEtOH (1 mM FcEtOPO32− and 20 μg/mL ALP) in the mineral oil. (B) Cyclic voltammograms of the 2 μL droplet containing 1.0 mM FcEtOPO32− in the mineral oil. (C) Cyclic voltammograms of the 2 μL droplet containing 0, 0.10, 0.25, 0.50, and 1.0 mM FcEtPO32− after a 5 min addition of the mineral oil. The scan rate was 60 mV/s. (AII and BII) The electrochemical signals of the cyclic voltammograms at 0.25 V (AI and BI) were plotted for time-course analyses. 7595
dx.doi.org/10.1021/ac301429n | Anal. Chem. 2012, 84, 7593−7598
Analytical Chemistry
Letter
droplet are concentrated by evaporation. We utilized this preconcentration process for signal amplification. The preconcentration process was done by leaving the droplet at 50 °C for 3 min, followed by sealing with the mineral oil. The electrochemical signal, after the preconcentration process, was 5.9 times higher than that before the preconcentration process because FcEtOPO32− in the water droplets was concentrated during the process (Figures 3B and 5). When 10 ng/mL ALP was included in the water droplets, 64% FcEtOPO32− was converted by concentrated ALP while 57% FcEtOPO32− was converted by ALP without the preconcentration process (Figure 5). Since the heating process for the evaporation process improves ALP activity, the effects of the preconcentration process on signal amplification are complicated. However, it is known that the evaporation process is effective for concentrating the components in the water droplets (Figures 3B and 5). This droplet system is also useful for signal amplification. Detection of ALP Activity on EBs. ALP activity on EBs was successfully detected (Figure 6). Since the supernatant contained some proteins released from EBs incubated in the Tris-HCl buffer, the cyclic voltammograms contained undesired background currents (Figure 6B). However, the onset potential and E1/2 appeared to be similar to those of the control measurements without cells (Figures 2−5). The background correction was made using the voltammograms obtained after the 20 min incubation as a background. Because ALP activity on the EBs is used to check the differentiation level of ES cells,16 the detection system can be used for the qualitycontrolled differentiation induction culture. In our previous study, ALP activity on EBs was electrochemically detected using both PAPP (substrate) and PAP (enzymatic product).14,17 Although the PAPP/PAP pair is useful for electrochemical ALP detection, PAP signal decays very quickly at high pHs (9.0−10.0) due to its susceptibility to air oxidation at high pHs which is suitable for alkaline phosphatase activity.18 In contrast, FcEtOPO32− and FcEtOH are stable at high pHs, compared to PAPP and PAP. Therefore, the assay using FcEtOPO32− and FcEtOH might be more quantitative than that using PAPP and PAP. Since ALP is used for various biochemical assays, we believe that the novel electrochemical detection system is useful for the biochemical assay based on ALP activity. The ALP activity of the EBs was also optically detected using p-nitrophenyl phosphate (PNPP) as a substrate. After the 30 min incubation, the ALP on a single EB converted PNPP to approximately 2 nmol p-nitrophenol (PNP) (Supporting text and Supplementary Figure 3, Supporting Information). The droplet-based assay using electrochemical detection indicated that most FcEtOPO32− was converted to FcEtOH in the water droplets as shown in Figure 6. Thus, a single EB is able to produce more than 0.5 nmol FcEtOH during a 1 h incubation. Although it is difficult to compare the optical method with the electrochemical/partitioning method because the incubation time and initial cell concentration were different, the ALP activities estimated from both methods appeared to be similar. The present water droplet-based assay using the partition between water and oil phases can minimize the use of expensive reagents and shorten the analysis time due to rapid separation between the substrate and product by partitioning. Since electrochemical reactions do not proceed in a mineral oil phase, it is particularly suitable if one wants to obtain direct information only on the water droplets, excluding information
FcEtOH, using the water/oil interface. The electrochemical signal from FcEtOPO32− was proportional to the concentration of FcEtOPO32− (Figure 3BII). Therefore, ALP activity can be estimated from the loss of FcEtOPO32− in the droplet after the ALP reaction. The oxidation current of FcEtOPO32− signals slightly increased in the first 2 min (Figure 3BI). However, further incubation of the water droplets on the device with the mineral oil gradually decreased the electrochemical signals and reached approximately 93% of the original signal after 6 h. The slight initial increase and gradual decrease of the signal may be correlated with flattening of the water droplets (Supplementary Figure 2, Supporting Information). In the initial stage, flattening may mix the components in the water droplets and the FcEtOPO32− concentration was observed to increase after addition of the mineral oil. The decrease of the signal intensity may be caused by the flatter droplet shape after 6 h. Since the change in the first hour was negligible (less than 2%), the electrochemical detection was performed within an hour after incubation. FcEtOH was partitioned sufficiently within an hour. We performed partition experiments of FcEtOH and FcEtOPO32− between water and oil using a conventional electrochemical method. After partition, the concentrations of these species were determined by cyclic voltammetry. The partition coefficients of FeEtOH and FcEtOPO32− in the mineral oil/water were found to be 6.1 and
