Research Article Rapid determination of superoxide free radical in hepatocellular carcinoma cells by MCE with LIF A method for determination of superoxide free radical (O��2 ) based on MCE with LIF was developed. Fluorescent reagent 2-chloro-1, 3-dibenzothiazolinecyclohexene, which was synthesized in our laboratory, was employed as the labeling reagent, the highest deri- vatization efficiency was obtained in 20 mM HEPES buffer (pH 7.4) for 10 min at 371C. Optimal determination of O��2 was achieved on a glass microchip, using 50 mM HEPES buffer (pH 7.4). Under the optimized conditions, linearity of response was obtained in the range of 4.0� 10�7–1.0� 10�5M, the detection limit (S/N5 3) was 0.15 mM, the RSDs of migration time and peak area were 2.6 and 3.8%, respectively. Interference experiment was investigated and the result indicates that 1000-fold molar excess of hydrogen peroxide does not interfere with the determination of O��2 in complex system. Finally, the method has been successfully applied to determine O��2 in hepatocellular carcinoma cells as well as phorbol 12-myristate 13-acetate stimulated RAW264.7 macrophages. The average recoveries were 97.3 and 98.6%, respectively. Keywords: Hepatocellular carcinoma cells / Macrophages RAW264.7 / MCE with LIF / Superoxide free radical / 2-Chloro-1,3-dibenzolinecyclohexene DOI 10.1002/elps.200800421 1 Introduction Superoxide anion radical (O��2 ) is a short-lived and highly reactive free radical in biological system. In a normal cellular environment, it mediates signal transduction and defenses against viral or bacterial attack [1]. In the case of overproduction, O��2 can lead to oxidative damage of proteins, DNA, and lipid peroxidation [2]. It has been reported that O��2 inhibits enzymes including glutathione peroxidase, catalase, and creatine kinase [3]. O��2 also serves as a precursor to other reactive oxygen species (ROS). For example, O��2 converts to hydrogen peroxide (H2O2) spontaneously or under action of superoxide dismutase; O��2 reacts with nitric oxide to form the powerful oxidant peroxynitrite, which can cause many diseases related to inflammatory processes and autoimmune diabetes [4, 5]. Therefore, rapid, sensitive detection and quantification of intracellular O��2 is critically important in understanding its physiological functions and pathogenesis of various diseases associated with ROS. Although several methods to detect O��2 such as electron spin resonance [6], electrochemistry [7], fluorescence spec- trometry [8], and HPLC [9] have been developed, there were some drawbacks for the above-mentioned methods, includ- ing the large sample volume, the long analyzing time, inconvenience to operate, or cost. CE also has obtained some achievements in analyzing total ROS [10] or single ROS [11, 12]. CE combined with LIF has been introduced to detect O��2 in rat skeletal muscle mitochondria, after deri- vatization with fluorescence reagent hydroethidine (HE) [12]. Unfortunately, HE has low selectivity to O��2 , mmol level of H2O2 may interfere with the determination of O �� 2 , while the content of H2O2 in biological system is far higher than mmol level due to accumulation in the biological system, and the long analysis time is unsuitable for rapid trapping of O��2 . Meanwhile, the injection and separation procedures in CE analysis were quite tedious and inconvenient using one- dimensional structure. Since 1990s, microfluidic chip or lab-on-a-chip provides a new technology platform for the research in chemistry, Xin Liu Qingling Li Xiaocong Gong Hongmin Li Zhenzhen Chen Lili Tong Bo Tang College of Chemistry, Chemical Engineering and Materials Science, Engineering Research Center of Pesticide and Medicine Intermediate Clean Production, Key Laboratory of Molecular and Nano Probes, Ministry of Education, Shandong Normal University, Jinan, P. R. China Received June 30, 2008 Revised August 19, 2008 Accepted September 4, 2008 Abbreviations: DBZTC, 2-chloro-1, 3-dibenzothiazoline- cyclohexene; HE, hydroethidine; HepG2 cells, hepatocellular carcinoma cells; H2O2, hydrogen peroxide; O��2 , superoxide radical; PMA, phorbol 12-myristate 13-acetate; ROS, reactive oxygen species; Tiron, 4,5- Dihydroxy-1,3-benzenedisulfonic acid disodium salt; XA, xanthine; XO, xanthine oxidase Correspondence: Professor Bo Tang, College of Chemistry, Chemical Engineering and Materials Science, Engineering Research Center of Pesticide and Medicine Intermediate Clean Production, Key Laboratory of Molecular and Nano Probes, Ministry of Education, Shandong Normal University, Jinan, 250014, P. R. China E-mail:
[email protected] Fax: 186-531-86180017 & 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com Electrophoresis 2009, 30, 1077–1083 1077 biology, microengineering, and other related microsystem fields [13, 14]. MCE offers many attractive benefits such as reducing sample requirements and reagent volumes, which can reduce overall cost and shorter analyzing time. Different functions could be integrated on a single microchip, which is an important step toward maintaining a completely closed system, thereby reducing contamination and eliminating human intervention and error [15, 16]. Among the several detection techniques employed in microchip analysis, LIF detection method is most easily adapted to the dimensions of microchips [17]. The coherence and low divergence of a laser beammake it easy to focus on very small analyte volumes and obtain much high irradiation, resulting in one of the most sensitive and powerful means of any detection systems [17, 18]. However, much attention has been focused on detection of total ROS using MCE coupled with LIF [19–22]. As different reactive species always coexist in the reactive environment and single ROS has its own unique physiolo- gical activity [23, 24], detection of single ROS is far more significant for further insight on its action mechanisms in biological processes. Currently, only Zhu et al. [25] employed HE as the labeling reagent and achieved O��2 determination on microchip using LIF detector. Despite the contribution, it is still urgent to develop an MCE-LIF method with better selectivity for O��2 determination. In our group, a new fluorescence reagent 2-chloro-1, 3-dibenzothiazolinecyclo- hexene (DBZTC) has been synthesized, which is highly sensitive and selective toward determination of O��2 with no interference from a 500-fold molar excess of H2O2 [26]. In this work, using fluorescence reagent DBZTC, a simple, rapid and sensitive O��2 determination was devel- oped by MCE with LIF detection technique. Interference experiment was investigated with MCE-LIF to test the feasibility of the method. After optimizing the derivatization conditions and electrophoresis parameters, the method was applied to determine O��2 in hepatocellular carcinoma (HepG2) cells and RAW264.7 macrophages. Finally, it reached simple, rapid, and sensitive O��2 detection on microchip. 2 Materials and methods 2.1 Chemicals and reagents All the chemicals were of analytical reagent grade or HPLC reagent grade. All aqueous solutions were prepared with doubly distilled water (18.2O cm); DBZTC was synthesized in-house, and DBZTC oxide was synthesized and purified according to the literature procedure [26]; HEPES, H2O2 (30% aqueous solution), Xanthine oxidase (XO), and phorbol 12-myristate 13-acetate (PMA) were obtained from Sigma (St. Louis, MO, USA); 4,5-Dihydroxy-1,3-benzenedi- sulfonic acid disodium salt (Tiron) and Xanthine (XA) was purchased from Shanghai Reagent. The stock solution (1.00 mM) of DBZTC and DBZTC oxide were prepared with dimethylsulfoxide and stored at 41C in darkness. These stock solutions were diluted to 5.0� 10�4M before use. The XA solution (1.00 mM) was prepared with 1.0� 10�2M NaOH; The XO solution of (1.00 U/mL) was prepared in 2.30 mM (NH4)2SO4, 1.0� 10�2M sodium salicylate biology buffer, stored at 2–81C; PMA was prepared in DMSO at a concentration of 1.0 mg/mL and stored at �201C before use; The stock solution (100 mL) of H2O2 (0.30 M) was freshly prepared by diluting H2O2 (30%, 3.4 mL) with water; HEPES, phos- phate, and borate buffer were prepared with doubly distilled water, and pH of the solutions were adjusted by the addition of appropriate amounts of hydrochloric acid or sodium hydroxide to a desired pH; Before use, all solutions were filtered through a 0.22 mm polypropylene filter film. All experiments were performed at room temperature (25721C). 2.2 Fluorescence spectra Fluorimetric spectra were measured with an Edinburgh FLS 920 spectrofluorimeter (Edinburgh Instruments, UK), fitted with a xenon lamp, in a quartz cuvette (1.0 cm optical path) as the container. Spectrometer slits were set for 3.5 nm band-pass. For recording the emission spectra, the excitation wavelength was set at 473 nm, with spectral bandwidth (10 nm), while the emission wavelength was scanned at a specified scan rate from 495 to 680 nm. A solution of 10 mM of DBZTC and 10 mM of DBZTC with 20 mM XA/20mU XO, was prepared, respectively, for the fluorescence analysis. 2.3 Cell culture and sample preparation HepG2 cells and RAW264.7 macrophages (purchased from the American Type Culture Collection, Manassas, USA), were cultured in DMEM containing 10% fetal bovine serum, 1% penicillin, and 1% streptomycin at 371C in a 5% CO295% air incubator MCO-15AC (SANYO). Cell viability was determined by the trypan-blue exclusion assay. When cells were in a logarithmic growth phase, a proportion of HepG2 cells were incubated with O��2 scavenger Tiron (100 mM) for 1 h at 37711C, another proportion of HepG2 cells were not incubated with Tiron. Similarly, a proportion of the macrophage cells were incu- bated with Tiron (100 mM) for 1 h prior to probe loading, another portion of macrophage cells were stimulated with PMA (2.0 ng/mL) at 37711C for 12 h. Then, each group of cells were incubated with DBZTC (10 mM) for 10 min at 37711C. After that, all of the cells harvested with the concentration of 1.0� 106 cells/mL by centrifugation in the cold were washed twice with 0.9% NaCl solution. Finally, these cells were resuspended again in a volume of HEPES (20 mM, pH 7.4) equal to that in DMEM, and were then disrupted for 10 min in a VC 130 PB ultrasonic disintegrator Electrophoresis 2009, 30, 1077–10831078 X. Liu et al. & 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com (Sonics & Materials). During sonic disruption, the temperature was maintained below 41C with circulating ice water. The broken cell suspensions were centrifuged at 12 000g for 5 min and the suspensions were immediately analyzed or kept at �201C for up to 2 days. 2.4 Microchip and LIF detector A schematic diagram of the microfluidic chip channels design is shown in Fig. 1. The glass microchip was provided by Dalian Institute of Chemical Physics, Chinese Academy of Sciences (China). The double-T channels were 65 mm wide and 15 mm deep, and the detection occurred 12 mm downstream from the injection cross in the separation channels. The reservoir positions are depicted in Fig. 1. A home-made intelligent eight-path-high-voltage elec- tric device [27] and MCE-LIF detector were employed. Optics collection system was confocal optics mode structure [28]. A 473 nm semiconductor double-pumped solid-state laser (20 mW) was used as the excitation source. The confocal detection module employed an objective to focus the beam to the center of the microchannel. The emitted fluorescence was filtered by a 520710 nm narrow band filter, and detected by a PMT. The data were collected at 20 Hz using a CT-22 data acquisition card. 2.5 MCE Prior to electrophoresis, the channels of microchip were rinsed with 1.0 M NaOH and doubly distilled water, respectively, for 15 min and equilibrated with running buffer for 15 min. Then, 10 mL of running buffer solutions was filled into the reservoirs of B, BW, and SW respectively, and 10 mL sample solution was filled into the sample reservoir, S. After that, the chip was placed on the LIF- detector worktable. By regulating the three dimensions manipulator, the laser beam was focused at the detection point. Sample injection was carried out using a pinch injection mode [29]. Four electrodes, randomly chosen from the intelligent eight-path-high-voltage electric device, were inserted into the reservoirs to apply voltages for electro- phoresis. Two sets of voltages were applied for sample loading and electrophoresis separation according to Table 1. 3 Results and discussion 3.1 Fluorescent derivatization reagent and back- ground experiments In this experiment, DBZTC was chosen as O��2 labeling reagent. Upon reaction with O��2 , nonfluorescent DBZTC was oxidized to yield strongly fluorescent DBZTC oxide, which owns better rigidity and a larger conjugated system. The reaction of DBZTC with O��2 is shown in Scheme 1. Derivatization conditions were optimized in the literature [26], and the highest derivation efficiency was achieved in 20 mM HEPES (pH 7.4) for 10 min at 37711C with 10 mM DBZTC. DBZTC reacts with O��2 in a 1:1 molar ratio and the derivation efficiency is almost up to 100% according to our previous experiments [26]; hence, the level of O��2 could be expressed by the level of DBZTC oxide in biological system. DBZTC oxide was used as the standard analyte for further electrophoresis parameters optimization. XA/XO reaction [30], which is a standard procedure for generating O��2 , was introduced for background experi- ments. Figure 2A is the emission spectra of DBZTC and DBZTC with XA/XO. It showed that the maximum emis- sion was 530 nm for DBZTC with XA/XO, no emission fluorescence was observed for DBZTC, when excited at 473 nm. As can be seen in Fig. 2B(a), no fluorescence peak was observed for DBZTC using the MCE with LIF detection. Figure 2B(b) and (c) represent the electropherograms of the mixtures of 10 mM DBZTC with 10 mM XA/10mU XO, and 10 mM DBZTC with 20 mM XA/20mU XO, respectively. As expected, fluorescence intensity increased with the incre- ment of O��2 concentrations. Together with the above results, no background fluorescence was produced and the feasibility of the proposed method for O��2 detection can be ensured. Figure 1. Schematic diagram of the microchip channels design. S, sample reservoir; B, buffer reservoir; SW, sample waste reservoir; BW, buffer waste reservoir. Table 1. Typical output voltage program for injection and separation Time section Run time/s Applied voltage/V V01 (S) V02 (SW) V03 (B) V04 (BW) 1st (injection) 20 400 0 200 240 2nd (separation) 60 1000 1000 1800 0 Electrophoresis 2009, 30, 1077–1083 Microfluidics and Miniaturization 1079 & 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com 3.2 Optimization of the microchip CE conditions For the optimized determination of DBZTC oxide, three kinds of buffers (HEPES, phosphate, and borate buffer) were tested. The DBZTC oxide was determined using one of the three buffers at pH 7.2–8.6, when other electrophoresis conditions were the same. Derivatization in HEPES buffer had higher signal responses and smoother baseline. There- fore, HEPES buffer was selected as the running buffer solution. The buffer pH, concentration, and separation electric field, which are the main factors affecting the peak shape, have been optimized and the results were as follows: Buffer pH mainly affects surface characteristics of the chip channels, thereby affecting the EOF and mobilities of derivatives. pH also has a large effect on fluorescence intensity of analyte. According to the previous report [26], when pH was in range of 7.0–8.6, DBZTC oxide exhibits relatively strong fluorescence, and thus the effects of pH on peak height were tested in the range of 7.0–8.6. As shown in Fig. 3A, peak height reached the maximum at pH 7.4 and at the same pH, an excellent peak shape and smoother base- line were also observed. Therefore, pH 7.4 was adopted for further testing. The effect of the buffer concentration on the peak height was studied at pH 7.4, in the range of 10–70mM. As shown in Fig. 3B, the peak height increased with buffer concentration ranging from 10 to 50 mM. As buffer concentration was above 50 mM, the peak height remained unchangeable. The optimal concentration was selected as 50 mM, since it renders an excellent peak shape and low current (12 mA). The effects of separation electric field on the peak height and the separation column efficiency over the range of 240–440 V/cm were investigated. In Fig. 3C and D, when the electric field was 360 V/cm, the peak height and theo- retical plates were both at the maximum. As a result, the optimal electric field was selected as 360 V/cm. Under the above optimized separation conditions, MCE analysis of 10 mM DBZTC oxide was performed, the repre- sentative microchip electropherogram for five repetitive injections is shown in Fig. 4. 3.3 Reproducibilities, linearity, and detection limit A standard solution of 10 mM DBZTC oxide was used to investigate the reproducibilities. The electropherogram of five repetitive injections of DBZTC oxide is shown in Fig. 4, the RSDs of migration time and peak areas are 2.6 and 3.8%, respectively. Figure 2. (A) Emission spectra of DBZTC and the mixture of DBZTC with XA/XO in 20 mM HEPES (pH 7.4); lex5 473 nm. (B) Electropherograms of (a) 10 mM DBZTC; (b) 10 mM DBZTC, 10 mM XA, and 10 mU XO; (c) 10 mM DBZTC, 20 mM XA, and 20 mU XO. XA/XO reaction is a standard procedure for generat- ing O��2 . Experimental conditions: injection time, 20 s; injection electric field, 400 V/cm; separation electric field, 360 V/cm; effective separation distance, 12 mm; running buffer, 40 mM HEPES (pH 7.4). Scheme 1. The reaction of DBZTC with O��2 Electrophoresis 2009, 30, 1077–10831080 X. Liu et al. & 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com The calibration curve for the standard sample was determined covering the concentration range of 1.0� 10�7–2.0� 10�5M. Nine concentration levels, namely 0.10, 0.40, 1.0, 2.0, 3.0, 4.0, 6.0, 10.0, 20.0 mM, were exam- ined. For each level, three measurements were taken and the average signal was used to make the calibration curve. Figure 5 shows the typical electropherogram for seven different concentration levels of DBZTC oxide. Figure 3. Influence of (A) buffer pH, (B) buffer concentration, (C) separation electric field on peak height, and (D) Effect of separation electric field on theoretical plates. Each point represents an average of experiments repeated three times. Experimental conditions: (A) 40 mM HEPES buffer; separa- tion electric field, 360 V/cm; injec- tion time, 20 s; injection electric field, 400 V/cm; effective separa- tion distance, 12 mm. (B) HEPES buffer, pH 7.4; other conditions were the same as in (A) except for buffer concentration. (C) An aliquot of 50 mM HEPES buffer, pH 7.4; other conditions were the same as in (B) except for separa- tion electric field. (D) The condi- tions were the same as in (C). Figure 4. Electropherogram for five repetitive injections of 10 mM DBZTC oxide. Experimental conditions: running buffer, 50 mM HEPES (pH 7.4); injection time, 20 s; injection electric field, 400 V/cm; separation electric field, 360 V/cm; effective separation distance, 12 mm. Figure 5. Typical electropherogram for seven different concen- tration levels of DBZTC oxide in the range of 0.40–10 mM. Other conditions were the same as in Fig. 4. The inset shows the peak height signals against DBZTC oxide concentration in the 0.40–10 mM range. Electrophoresis 2009, 30, 1077–1083 Microfluidics and Miniaturization 1081 & 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com The standard curve was linear in the range of 4.0� 10�7–1.0� 10�5M. The calibration equation and regression coefficient were: y5 0.7970x13.4440 and R5 0.9988 (n5 7) in terms of peak height enhancement as a function of DBZTC oxide concentration. The concentration LOD was calculated on the basis of an S/N of 3 and was 1.5� 10�7M. Considering a repeatable injection volume of 200 pL with double-T design, the calculated mass LOD was 0.03 fmol (n5 3). 3.4 Selectivity of the method In biological samples, O��2 is present in low concentrations, and other RS and biological compounds may interfere with the determination of O��2 . To assess the selectivity of the method toward determination of 10.0 mM O��2 , a mixture of 10 mM DBZTC with 10.0 mM H2O2 was analyzed under the optimized electrophoresis conditions, and no H2O2 peak was observed. Other biological compounds such as Vc and glutathione were also investigated and did not make interferences with the determination of O��2 . These observations revealed that the present derivatization and separation method provided relatively high selectivity toward O��2 , especially without interference from a 1000-fold molar excess of H2O2, and allowed its quantitative determination in biological samples. 3.5 Determination of O��2 in HepG2 cells and RAW264.7 macrophage extracts The proposed method was applied to the analysis of HepG2 cells and RAW264.7 macrophages. Before electrophoresis experiments, to exclude the native interference of these samples, the fluorescence of cell samples without DBZTC labeling were examined with Edinburgh FLS 920 spectro- fluorimeter, and no native fluorescence was detected at lex/ lem5 473/525 nm. The electropherogram of DBZTC-labeled HepG2 cells sample is shown in Fig. 6A-a. A peak was clearly detected and was identified as O��2 peak by comparison of migration time with standard solution (Fig. 4). To further confirm the O��2 peak, Tiron, a cell-permeable O �� 2 scavenger [31], was added to the HepG2 cells suspension. The corre- sponding electropherogram is given in Fig. 6A-b, where O��2 peak completely disappeared. The electrophoresis experi- ments of DBZTC-labeled HepG2 cells extract have been repeated three times, and the RSD of peak areas was about 4.4%. The O��2 concentration for HepG2 cells extract was quantified by the standard curve and the regression equation and was calculated as 0.7170.03 mM (mean7SD). The proposed method was then applied to the analysis of RAW264.7 macrophages stimulated by 2.0 ng/L PMA. PMA is a stimulator of cell respiratory burst to give rise to ROS. Approximately, the O��2 peak was also observed by the migration time of 30 s and disappeared after 100 mM Tiron added to suspension (Fig. 6B). The experiments of DBZTC- labeled RAW264.7 macrophages have also been repeated three times, and the RSD of peak areas was about 4.6%. The O��2 level of RAW264.7 macrophages was calculated as 0.8870.04 mM. The result is consistent with that detected by fluorescence spectrum of our group previously (0.9270.02 mM) [26]. To validate the method, recovery experiments were determined by adding known amounts of DBZTC oxide standard solution to the cell samples. These samples were performed under the optimal conditions and the results are shown in Table 2. The results indicate that the method is reproducible and satisfactory for determining the level of O��2 in cells. Figure 6. (A) Electropherogram of HepG2 cells extracts without (a) and with (b) the Tiron treatment. (B) Electropherogram of RAW264.7 macrophages without (a) and with (b) Tiron treat- ment. Cell concentration is 1.0� 106 cells/mL. Experimental conditions were the same as in Fig. 4. Electrophoresis 2009, 30, 1077–10831082 X. Liu et al. & 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com 4 Concluding remarks In this study, an MCE with LIF method was developed for determination of O��2 using DBZTC as the fluorescent reagent. With the optimized derivatization conditions and electrophoresis parameters, rapid, sensitive detection and quantification of O��2 was obtained within 30 s. A mass detection limit of 0.03 fmol was achieved owing to the minute sample volume. The method was applied for the determination of O��2 in HepG2 cells and PMA-stimulated RAW264.7 macrophages. Experimental results showed that the developed method was simple, rapid, and sensitive, and could be applied for determination of O��2 without interference from a 1000-fold molar excess of H2O2 in various biological systems. MCE offers favorable potentials for facilitating investigation of the cellular homeostasis and the pathogenesis of various diseases associated with O��2 at the molecular level. Moreover, the method also provides a new strategy for determination of other ROS in complex biological matrix. This work was supported by National Basic Research Program of China (973 Program, 2007CB936000), National Natural Science Funds for Distinguished Young Scholar (No.20725518), Major Program of National Natural Science Foundation of China (No.90713019), National Natural Science Foundation of China (No.20875058), The Science and Technology Develop- ment Programs of Shandong Province of China (No. 2008 GG30003012), and Natural Science Foundation of Shandong Province in China (No.Y2008B15). The authors have declared no conflict of interest. 5 References [1] Aguirre, J., Rios, M. M., Hewitt, D., Hansberg, W., Trends Microbiol. 2005, 13, 111–118. [2] Kaori, I., Keizo, T., Miho, A., Nobuko, K. et al., Science 2008, 320, 661–664. [3] Halliwell, B., Gutteridge, J. M., Biochem. J. 1984, 219, 1–14. [4] Huang, J., Li, D. J., Diao, J. C., Hou, J., Zou, G. L., Talanta 2007, 72, 1283–1287. [5] Wang, H. M., Cai, R. X., Lin, Z. X., Talanta 2006, 69, 509–514. [6] Nilsson, U. A., Haraldsson, G., Bratell, S., Sorensen, V., Akerlund, S., Acta Physiol. Scand. 1993, 147, 263–270. [7] Zielonka, J., Vasquez, V. J., Kalyanaraman, B., Free Radic. Biol. Med. 2006, 34, 1050–1057. [8] Munzel, T., Afanasev, I. B., Klescchyov, A. L., Harrison, D. G., Arterioscler. Thromb. Vasc. Biol. 2002, 22, 1761–1768. [9] Zhao, H. T., Joseph, J., Fales, H. M., Sokoloski, E. A., Levine, R. L., Proc. Natl. Acad. Sci. USA 2005, 102, 5727–5732. [10] Parmentier, C., Wellman, M., Siest, G., Leroy, P., Elec- trophoresis 1999, 20, 2938–2944. [11] Shihabi, Z. K., Electrophoresis 2006, 27, 4215–4218. [12] Meany, D. L., Thompson, L. D., Arriaga, E. A., Anal. Chem. 2007, 79, 4588–4594. [13] Dittrich, P. S., Tachikawa, K., Manz, A., Anal. Chem. 2006, 78, 3887–3908. [14] Reyes, D. R., Iossifidis, D., Auroux, P. A., Manz, A., Anal. Chem. 2002, 74, 2623–2636. [15] Auroux, P. A., Iossifidis, D., Reyes, D. R., Manz, A., Anal. Chem. 2002, 74, 2637–2652. [16] Vilkner, T., Janasek, D., Manz, A., Anal. Chem. 2004, 76, 3373–3386. [17] Go¨tz, S., Karst, U., Anal. Bioanal. Chem. 2007, 387, 183–192. [18] Uchiyama, K., Nakajima, H., Hobo, T., Anal. Bioanal. Chem. 2004, 379, 375–382. [19] Ling, Y. Y., Yin, X. F., Fang, Z. L., Electrophoresis 2005, 26, 4759–4766. [20] Sun, Y., Yin, X. F., J. Chromatogr. A 2006, 1117, 228–233. [21] Sun, Y., Yin, X. F., Ling, Y. Y., Fang, Z. L., Anal. Bioanal. Chem. 2005, 382, 1472–1476. [22] Qin, J. H., Ye, N. N., Lin, B. C., Electrophoresis 2005, 26, 1155–1162. [23] Wulf, D., Physiol. Rev. 2002, 82, 47–95. [24] Setsukinai, K., Urano, Y., Kakinuma, K., Majima, H. J., Nagano, T., J. Biol. Chem. 2003, 278, 3170–3175. [25] Zhu, L. L., Lu, M., Yin, X. F., Talanta 2008, 75, 1227–1233. [26] Gao, J. J., Xu, K. H., Tang, B., FEBS J. 2007, 274, 1725–1733. [27] Li, Q. L., Tang, B., Tian, H. X., Chinese Patent No.200510104343.3. [28] Zhou, X. M., Liu, D. Y., Zhong, R. T., Dai, Z. P. et al., Electrophoresis 2004, 25, 3032–3039. [29] Fu, L. M., Yang, R. J., Lee, G. B., Anal. Chem. 2003, 75, 1905–1910. [30] Benov, L., Sztejnberg, L., Fridovich, I., Free Radic. Biol. Med. 1998, 25, 826–831. [31] Reynolds, G. A., Drexhage, K. H., J. Org. Chem. 1977, 42, 885–888. Table 2. Recoveries of O��2 in HepG2 cells and RAW264.7 macrophages samples (n5 3) Sample O��2 concentration (ı`M) Added (mM) Found (mM) Mean (mM) Average recovery (%) RSD (%) HepG2 0.7170.03 1.00 1.68, 1.62, 1.75 1.6870.07 97.3 4.8 Macrophages 0.8870.04 1.00 1.80, 1.96, 1.84 1.8670.08 98.6 4.5 Electrophoresis 2009, 30, 1077–1083 Microfluidics and Miniaturization 1083 & 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com