Microfluidic Cell Culture and Metabolism Detection with Electrospray

Jun 11, 2010 - Simultaneous determination of niacin and pyridoxine at trace levels by using diode array high-performance liquid chromatography and liq...
0 downloads 4 Views 2MB Size
Download PDF
Anal. Chem. 2010, 82, 5679–5685

Microfluidic Cell Culture and Metabolism Detection with Electrospray Ionization Quadrupole Time-of-Flight Mass Spectrometer Dan Gao,†,‡ Huibin Wei,† Guang-Sheng Guo,‡ and Jin-Ming Lin*,† The Key Laboratory of Bioorganic Phosphorus Chemistry and Chemical Biology, Department of Chemistry, Tsinghua University, Beijing 100084, China, and State Key Laboratory of Chemical Resource Engineering, Beijing University of Chemical Technology, Beijing 100029, China A novel method for the characterization of drug metabolites was developed by integrating chip-based solid-phase extraction (SPE) with an online electrospray ionization quadrupole time-of-fight mass spectrometer (ESI-Q-TOFMS). The integrated microfluidic device was composed of circular chambers for cell culture and straight microchannels with shrink ends to pack the solid-phase material for sample cleanup and concentration prior to mass analysis. By connecting the two separated microchannels with polyethylene tubes, drug metabolism studies related to functional units, including cell culture, metabolism generation, sample pretreatment, and detection, were all integrated into the microfluidic device. To verify the feasibility of a drug metabolism study on the microfluidic device, the metabolism of vitamin E in human lung epithelial A549 cells was studied. The metabolites were successfully detected by online ESI-Q-TOF-MS with high sensitivity and short analysis time (8 min). By integrating several parallel channels, the desalting and concentration process could be simultaneously achieved. The total sample pretreatment time only needed about 15 min, and solvent consumption could be reduced to less than 100 µL. All this demonstrated that the developed microfluidic device could be a potential useful tool for cellular drug metabolism research. Metabolomics studies are very useful in medical and scientific research. It is an important platform to understand all metabolites in a biological organism and its dynamic multiparametric metabolic response to pathophysiological stimuli.1-4 Recently, a variety of analytical technologies have been established to detect cellular metabolites, such as capillary * To whom correspondence should be addressed. E-mail: jmlin@ mail.tsinghua.edu.cn. Fax/Tel: +86 10 62792343. † Tsinghua University. ‡ Beijing University of Chemical Technology. (1) Ippolito, J. E.; Xu, J.; Jain, S. J.; Moulder, K.; Mennerick, S.; Crowley, J. R.; Townsend, R. R.; Gordon, J. I. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 9901–9906. (2) Kell, D. B. Drug Discovery Today 2006, 11, 1085–1092. (3) Nicholson, J. K.; Lindon, J. C. Nature 2008, 455, 1054–1056. (4) Yan, J.; Sun, Y. H.; Zhu, H.; Marcu, L.; Revzin, A. Biosens. Bioelectron. 2009, 24, 2604–2610. 10.1021/ac101370p  2010 American Chemical Society Published on Web 06/11/2010

electrophoresis-mass spectrometry (CE-MS),5-7 gas chromatography/mass spectrometry (GC/MS),8 and liquid chromatography-mass spectrometry (LC-MS).1,9,10 Traditional methods for the analysis of cellular metabolism often need to cultivate cells in Petri dishes or microtiter plates and then have a periodic collection of cells or culture medium for following detection. However, these conventional methods require a large number of cells, reagents, and complicated sample pretreatment. The micrototal analysis systems (µTAS) have been a growing interest for analytical chemistry since microfluidic technology provides several advantages such as low reagent consumption, rapid analysis, and miniaturization. Many functional components including cell culture11-16 and sample pretreatment17-19 can be integrated into microfluidic devices. Moreover, microfluidic devices provide a miniaturized environment for working with mammalian cells. Electrospray ionization-mass spectrometry (ESI-MS), a high sensitive and selective detection system, is compatible with low flow rates on chip. It can also provide structural information that is essential for a metabolism study. Furthermore, ESI coupled (5) Lapainis, T.; Rubakhin, S. S.; Sweedler, J. V. Anal. Chem. 2009, 81, 5858– 5864. (6) Soga, T.; Igarashi, K.; Ito, C.; Mizobuchi, K.; Zimmermann, H. P.; Tomita, M. Anal. Chem. 2009, 81, 6165–6174. (7) Edwards, J. L.; Chisolm, C. N.; Shackman, J. G.; Kennedy, R. T. J. Chromatogr., A 2006, 1106, 80–88. (8) Fiehn, O.; Kopka, J.; Dormann, P.; Altmann, T.; Trethewey, R. N.; Willmitzer, L. Nat. Biotechnol. 2000, 18, 1157–1161. (9) Tolstikov, V. V.; Lommen, A.; Nakanishi, K.; Tanaka, N.; Fiehn, O. Anal. Chem. 2003, 75, 6737–6740. (10) Yoshida, H.; Mizukoshi, T.; Hirayama, K.; Miyano, H. J. Agric. Food Chem. 2007, 55, 551–560. (11) Hung, P. J.; Lee, P. J.; Sabounchi, P.; Aghdam, N.; Lin, R.; Lee, L. P. Lab Chip 2005, 5, 44–48. (12) Gomez-Sjoberg, R.; Leyrat, A. A.; Pirone, D. M.; Chen, C. S.; Quake, S. R. Anal. Chem. 2007, 79, 8557–8563. (13) Nevill, J. T.; Cooper, R.; Dueck, M.; Breslauer, D. N.; Lee, L. P. Lab Chip 2007, 7, 1689–1695. (14) Toh, Y. C.; Zhang, C.; Zhang, J.; Khong, Y. M.; Chang, S.; Samper, V. D.; van Noort, D.; Hutmacher, D. W.; Yu, H. R. Lab Chip 2007, 7, 302–309. (15) Hufnagel, H.; Huebner, A.; Gulch, C.; Guse, K.; Abell, C.; Hollfelder, F. Lab Chip 2009, 9, 1576–1582. (16) Liu, K.; Pitchimani, R.; Dang, D.; Bayer, K.; Harrington, T.; Pappas, D. Langmuir 2008, 24, 5955–5960. (17) Xu, N. X.; Lin, Y. H.; Hofstadler, S. A.; Matson, D.; Call, C. J.; Smith, R. D. Anal. Chem. 1998, 70, 3553–3556. (18) Lion, N.; Gobry, V.; Jensen, H.; Rossier, J. S.; Girault, H. Electrophoresis 2002, 23, 3583–3588. (19) Lion, N.; Gellon, J. O.; Jensen, H.; Girault, H. H. J. Chromatogr., A 2003, 1003, 11–19.

Analytical Chemistry, Vol. 82, No. 13, July 1, 2010

5679

with tandem MS (MS/MS) can be used to confirm unknown compounds through their mass-to-charge ratio (m/z) and MS/ MS fragmentations with high resolution. One challenge of coupling microfluidic devices to MS through electrospray ionization is the development of a stable and effective interface, and the interface between microfluidic systems and ESI-MS has been successfully resolved in early years.20,21 Tan et al.22 have described a multiple-channel microchip directly coupled to ESI-MS for highthroughput analysis. To monitor on chip cellular metabolites by ESI-MS, the pretreatment of sample is the most important procedure to remove undesired compounds such as salts, buffers that interfere with MS. In addition, preconcentration of metabolites is necessary to concentrate the samples leading to high sensitivity for the assay. Solid-phase extraction (SPE) is an important and widely used sample preparation method for the extraction and preconcentration of samples, as well as cleanup interferences from analytical samples.23-27 In microfluidic systems, the small amount of sample is very suitable for SPE handling within a chip without sample-transfer steps between devices. Besides, microfluidic devices have the advantage of fabricating identical multicomponents on a chip for high-throughput parallel processing. Therefore, there is an attractive interest in coupling SPE to chip-based systems. Recently, many different methods have been reported to load stationary phases into microfluidic devices, such as coating,28 packing,29 or in situ polymerized monolith.30,31 Vitamin E consists of eight structurally related tocopherols and tocotrienols. Due to their ability to function as fat-soluble antioxidants, these compounds have been regarded as important nutrients for the maintenance of human health.32 All vitamin E forms have a chromanol ring and a 13-carbon-long hydrophobic side chain. They are metabolized via a similar cytochrome P450 (CYP450) dependent mechanism, including ω-hydroxylation/ oxidation and β-oxidation resulting in shortening the side chain, the terminal metabolites excreted in the urine.33-37 This mechanism was demonstrated by the identification of a series of (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31) (32) (33) (34) (35) (36) (37)

Zhang, B. L.; Foret, F.; Karger, B. L. Anal. Chem. 2000, 72, 1015–1022. Le Gac, S.; Arscott, S.; Rolando, C. Electrophoresis 2003, 24, 3640–3647. Tan, A. M.; Benetton, S.; Henion, J. D. Anal. Chem. 2003, 75, 5504–5511. Gilar, M.; Belenky, A.; Wang, B. H. J. Chromatogr., A 2001, 921, 3–13. Benetton, S.; Kameoka, J.; Tan, A. M.; Wachs, T.; Craighead, H.; Henion, J. D. Anal. Chem. 2003, 75, 6430–6436. Miyazaki, S.; Morisato, K.; Ishizuka, N.; Minakuchi, H.; Shintani, Y.; Furuno, M.; Nakanishi, K. J. Chromatogr., A 2004, 1043, 19–25. Yang, Y. N.; Li, C.; Lee, K. H.; Craighead, H. G. Electrophoresis 2005, 26, 3622–3630. Hagan, K. A.; Meier, W. L.; Ferrance, J. P.; Landers, J. P. Anal. Chem. 2009, 81, 5249–5256. Kutter, J. P.; Jacobson, S. C.; Ramsey, J. M. J. Microcolumn Sep. 2000, 12, 93–97. Oleschuk, R. D.; Shultz-Lockyear, L. L.; Ning, Y. B.; Harrison, D. J. Anal. Chem. 2000, 72, 585–590. Yang, Y. N.; Li, C.; Kameoka, J.; Lee, K. H.; Craighead, H. G. Lab Chip 2005, 5, 869–876. Jemere, A. B.; Oleschuk, R. D.; Ouchen, F.; Fajuyigbe, F.; Harrison, D. J. Electrophoresis 2002, 23, 3537–3544. Brigelius-Flohe, R.; Kelly, F. J.; Salonen, J. T.; Neuzil, J.; Zingg, J. M.; Azzi, A. Am. J. Clin. Nutr. 2002, 76, 703–716. Swanson, J. E.; Ben, R. N.; Burton, G. W.; Parker, R. S. J. Lipid Res. 1999, 40, 665–671. Parker, R. S.; Swanson, J. E. Biochem. Biophys. Res. Commun. 2000, 269, 580–583. Birringer, M.; Pfluger, P.; Kluth, D.; Landes, N.; Brigelius-Flohe, R. J. Nutr. 2002, 132, 3113–3118. Sontag, T. J.; Parker, R. S. J. Biol. Chem. 2002, 277, 25290–25296. Wu, J. H. Y.; Hodgson, J. M.; Ward, N. C.; Clarke, M. W.; Puddey, I. B.; Croft, K. D. Free Radical Biol. Med. 2005, 39, 483–494.

5680

Analytical Chemistry, Vol. 82, No. 13, July 1, 2010

carboxychromanol intermediates in human hepatoma HepG2 cells and human lung epithelial A549 cells incubated with vitamin E forms.38-41 In recent years, it was reported that some sulfated long-chain carboxychromanols were generated by the sulfation of the intermediate metabolites that occurred parallel to β-oxidation when A549 cells were incubated with tocopherols. In this paper, we presented an integrated microfluidic device with two functional parts for a drug metabolism study: one part with a cell culture chamber to seed, cultivate, and stimulate A549 cells, and the cells could keep viability for at least 5 days; another part with a narrow section was designed to load solid-phase material to perform sample cleanup and the concentration process necessary for ESI-Q-TOF-MS detection. These two parts were combined together with polyethylene tubes and directly coupled to ESI-Q-TOF-MS for online metabolism detection. Vitamin E, which is known to undergo metabolism to phytyl chain-shortened metabolites, was selected as a model drug to validate the feasibility of a drug metabolism study on the developed microfluidic device. After vitamin E incubated with about 200 A549 cells for 24-72 h, the trace level of its metabolites were successfully detected by MS with high sensitivity and high throughput. Therefore, this established platform provided a microscale alternative to a currently used in vitro method of drug metabolism research. EXPERIMENTAL SECTION Microchip Fabrication. Microfluidic devices were fabricated from PDMS (Sylgard 184, Dow corning) using standard soft lithography and replica molding techniques as reported elsewhere.42 Briefly, a SU-8 2050 negative photoresist (Microchem) was spin-coated on a cleaned silicon wafer. After spinning, the wafer was prebaked (6 min at 65 °C and then 3 min at 95 °C) and then exposed to UV light through the photomask. After postbaking and development, the master was hard-baked for 5 min at 65 °C. A 10:1 weight mixture of PDMS prepolymer and curing agent was poured onto the silicon master and baked in an oven at 80 °C for 2 h. After curing, the PDMS replicas were peeled off from the master and the connection holes were punched before sealing. The PDMS replicas were irreversibly sealed with glass slides by oxygen plasma (PDC-32G, Harrick Plasma, Ithaca, NY) for 90 s. The devices were cured at 60 °C for 2 h to reinforce the bonding. Cell Culture. A549 cells were donated by Dr. Lei (Department of Biochemistry, Tsinghua University). Cells were cultured in Dulbecco’s modified Eagle’s minimal essential medium (DMEM, Gibco, Grand Island, NY) containing 10% fetal bovine serum, 100 µg/mL penicillin, and 100 µg/mL streptomycin. Cells were maintained at 37 °C in a 5% CO2-humidified air atmosphere and passaged every 2 or 3 days. Prior to experiments, cells were released from tissue culture flasks by trypsinization with 0.25% Trypsin EDTA (Gibco). The viability of cells was determined by a Live/Dead assay kit (Invitrogen, CA, USA). Cultures were maintained between 1.0 × 105 and 1.6 × 107 cells/mL. Serial dilutions were performed with supplemented DMEM. (38) You, C. S.; Sontag, T. J.; Swanson, J. E.; Parker, R. S. J. Nutr. 2005, 135, 227–232. (39) Jiang, Q.; Freiser, H.; Wood, K. V.; Yin, X. M. J. Lipid Res. 2007, 48, 1221– 1230. (40) Freiser, H.; Jiang, Q. J. Nutr. 2009, 139, 884–889. (41) Freiser, H.; Jiang, Q. Anal. Biochem. 2009, 388, 260–265. (42) Unger, M. A.; Chou, H. P.; Thorsen, T.; Scherer, A.; Quake, S. R. Science 2000, 288, 113–116.

For the cell culture in microfluidic devices, the devices were initially rinsed with DI water, 70% ethanol, and PBS and then exposed to UV irradiation for 5 min to ensure sterility. The microfluidic devices were first modified with 0.1% poly-L-lysine. Cell suspension with a concentration of 5.0 × 104 cells/mL was seeded into the microfluidic device using a Hamilton gastight syringe which was connected via polyethylene tubing. It was very important to remove all the air in the channels before cell injection. Devices were then incubated for 24 h at 37 °C in a 5% CO2 humidified air atmosphere before injection of fresh medium. The cell culture medium was refreshed every day. If necessary, cells could be trypsinized under sterile conditions, and then, cells were flushed out by applying a syringe pump at a flow rate of 5 µL/min and stored in a sterile tube for further analysis. The on-chip cells were passaged every 4 to 5 days. Metabolism of Tocopherols in A549 Cells on Microfluidic Chips. To determine the metabolism of R-tocopherol and γ-tocopherol (R-T, purity >96%, and γ-T, purity >98%, Sigma, St. Louis, MO) in A549 cells cultured on microfluidic devices, cells were first maintained in DMEM supplemented with 10% fetal bovine serum without antibiotics under standard culture conditions. After the cell attachment on the surfaces of the microchannels, the medium was replaced with tocopherol enriched DMEM prepared according to the method of Swanson et al.43 Briefly, tocopherol enriched fetal bovine serum (FBS) was first prepared by adding corresponding tocopherols to the serum from ethanol stock solution. The tocopherol enriched FBS was incubated overnight at 4 °C before being added to the DMEM medium. The final concentration of FBS was 1% and tocopherol concentration ranged from 50 to 250 µM, and ethanol concentrations were less than 0.5%. Cells were incubated with tocopherol enriched medium in the microchannels at 37 °C for 24-72 h before analysis. For all experiments, the cell viability was tested by the Live/Dead assay kit. Extraction of Tocopherol Metabolites from Culture Medium and Analysis by ESI-Q-TOF-MS. After incubation of cells with tocopherol enriched medium, an integrated on-chip SPE column was used to desalt and concentrate the tocopherol metabolites prior to analyze by ESI-Q-TOF-MS. Microdevices were filled with unique polymeric SPE beads (Bond Elut Plexa, 45 µm) which are an alternative for the extraction of acidic analytes. The beads were injected into the microchannels with a syringe at the flow rate of 5 µL/min and then conditioned with methanol and water, respectively, prior to extraction. The extraction procedure consisted of pressure-driven load, wash, and elution steps. One cell culture channel was only connected with one micro-SPE column with a polyethylene tube during sample pretreatment. The medium was first transferred and loaded onto the polymer bed at a flow rate of 10 µL/min using a 500 µL Hamilton gastight syringe (Hamilton, Las Vegas, NV). A 100 µL wash solution of 5% (v/v) methanol-water was flowed over the polymer bed to remove any unbound materials, proteins, or salts. After that, the microSPE column was connected directly to an ESI ion source by a fused silica capillary with an inner diameter of 50 µm and outer diameter of 365 µm. Finally, the metabolites were eluted from a polymer bed with 5% ammonia in methanol and analyzed by ESI-

Q-TOF online. Ammonia (5%) was added to methanol to enhance the polarity of acidic metabolites that benefited from the elution of metabolites from the immobile phase. Mass Spectrometry. The experiments were carried out on a quadrupole ion trap mass spectrometer equipped with a modified micro-ESI source and operated in the ion spray mode at 3000 V in the negative ion mode. The heated inlet capillary was maintained at 200 °C. The flow rate of sample for direct infusion was 100 µL/h. A coaxial nebulizer N2 gas flow around the ESI emitter was used to assist the generation of ion. All mass spectra were externally calibrated by tunemix (Agilent, USA) at the negative ion mode with the mass range of m/z 50-1500. MS/MS analysis was carried out using argon as a collision gas to fragment precursor ions via collision-induced dissociation prior to mass analysis. Suitable collision energy for small-molecule fragmentation was found by maximizing the intensity of the m/z 149 fragment generated from the dissociation of γ-T.

(43) Parker, R. S.; Sontag, T. J.; Swanson, J. E. Biochem. Biophys. Res. Commun. 2000, 277, 531–534.

(44) Bedair, M. F.; Oleschuk, R. D. Anal. Chem. 2006, 78, 1130–1138. (45) Freire, S. L. S.; Wheeler, A. R. Lab Chip 2006, 6, 1415–1423.

RESULTS AND DISCUSSION Microdevice Design and Combination with Mass Spectrometry. For the cell metabolism assay, two different functional parts of the microfluidic devices were connected together by polyethylene tubes (Figure 1A). For cell culture, a microchannel with a diameter of 4 mm, circular shape, was designed in a straight channel to ensure uniform distribution of cell suspension in all directions (Figure 1B). Another separated microchannel with a shrink end was used to immobilize SPE beads for desalting and concentrating metabolites of tocopherols before ESI-Q-TOF-MS detection. The dimensions of the microchannel used to load SPE beads were 15 mm long and 1 mm wide, following with a shrink end which was narrower than the diameter of SPE beads, so that SPE beads can be compact immobilized in the microchannels, as shown in Figure 1C. Figure 1D shows several parallel channels were connected together for high-throughput analysis. Recently, a lot of research has been focused on the combination of a microfluidic device with a mass spectrometer, mainly by fusedsilica capillaries or nanospray needles.5,20,44,45 In this work, we combined the microchip with an ESI ion source by a fused-silica capillary with a polytetrafluoroethylene (PTFE) cannula on the end. We could match the diameter of the joint for connecting an ESI ion source and the injector as well as the inlets and outlets of the microchannels on the PDMS chip. Microfluidic Cell Culture. A549 cells were seeded into the cell culture chamber, and then, the microfluidic device was placed in a Petri dish to keep it sterile and incubated at 37 °C in 5% CO2 and 95% air. Because of the permeability of PDMS, the cell culture medium inside the microchannels can easily evaporate after a few hours. To diminish the evaporation, a layer of cell culture medium was covered on the surface of the microchip. After this treatment, no evaporation was observed in the microchannels within a few days. In order to know whether the microenvironment in the microchannels had any negative effect on the cell viability, a Live/Dead assay kit composed of calcein AM and ethidium homodimer-1 (EthD-1) was used to investigate the cell viability. As shown in Figure 2, the cells on chip were fully spread and attached on the glass surface and

Analytical Chemistry, Vol. 82, No. 13, July 1, 2010

5681

Figure 1. Microfluidic device for cell culture and metabolism detection. (A) The schematics of one unit for cell culture and sample pretreatment prior to ESI-Q-TOF-MS detection. (B) The dimensions of the cell culture channel. (C) The dimensions of the micro-SPE column. (D) An image of the microfluidic device filled with a red dye in the cell culture section, including six units.

Figure 2. Images of the adherent A549 cells were cultivated on the microfluidic chip for 5 days and maintained their healthy morphology and propagates. On the fifth day, the cells were stained with a Live/ Dead assay kit.

showed healthy morphology in the chambers for at least 5 days. On the fifth day, approximately 100% of cells kept viability with bright green fluorescence, which could be used for further studies. Optimization Conditions of On-Chip SPE for ESI-Q-TOFMS Detection. The unique polymer beads were selected as the packing material to desalt and absorb the metabolites of tocopherols. The microporous center prevents the macromolecular proteins from entering the beads and the hydrophobic center of 5682

Analytical Chemistry, Vol. 82, No. 13, July 1, 2010

the beads prevents salt absorption, while keeping small analytes including neutralized acids from entering and being retained there. γ-T and γ-CEHC, which have similar structures and properties with the metabolites of tocopherols, were selected as models to evaluate the desalting and absorption abilities of the beads. Tocopherols and γ-CEHC stock solution (5 mM) were made in ethanol prior to use and diluted 1:1000 with DMEM to achieve a final concentration of 5 µM. They were then injected into the micro-SPE columns as detailed above and detected by ESI-Q-TOFMS. The mass spectral results were obtained under the mass range m/z 50-1500. As shown in Figure 3A,C, γ-T with [M - H]- ) 415.38 and γ-CEHC with [M - H]- ) 263.13 can be detected, and their structures were further identified by MS/ MS spectra. An extra energy of 25 eV was applied on the extracted ions m/z 415 and m/z 263 to get the corresponding fragments. As shown in Figure 3B,D, the peak of m/z 149 was the common fragment generated from γ-chroman compounds, while the peaks of m/z 204 and 218 indicated the existence of γ-CEHC. The above results showed that the polymer beads were appropriate for extraction of the metabolites from tocopherols. The packing amount of polymer beads and the sampling volume have great effects on the absorption ability. The sample volume in the microchannel was about 3 µL. With 3 µL of γ-CEHC and the concentrations increasing from 1 to 10 µM, the intensities of monitor ion peaks were increased as well (data not shown). In addition, about 200 cells in the microchannel were used to metabolize tocopherols which determined the metabolites were at trace levels. Consequently, the sample capacity of a micro-SPE column has the ability to analyze metabolites. It was also demonstrated by the following experimental results. The flow rates were important factors influencing the response of the system, including the relative mass spectrometry signal intensity, reproducibility, and lifetime of on-chip SPE and the ESIQ-TOF-MS combination system. The flow rates must match the whole detection procedure. We investigated sample injection

Figure 3. Mass spectra of standard γ-T (A) and γ-CEHC (C) solutions and corresponding tandem mass spectra (B, D). Spectra were obtained in the negative ion mode. Table 1. Effect of Sample Injection Flow Rate on the Ion Intensity of Monitored γ-Tocopherol Extracted by the Micro-SPE Column sample injection flow rate (µL/min) ion intensity

2

4

6

8

10

3570.3 ± 393.9

3495.7 ± 210.4

3644.0 ± 162.3

3883.0 ± 223.1

3967.7 ± 40.5

velocity, washing flow rate, and elution flow rate. To ensure the SPE column worked well on the microfluidic chip, the velocity of sample injection should not be too high. The velocity ranged from 2 to 10 µL/min, and the standard γ-tocopherol was loaded onto

Figure 4. Relationship between the ion intensity of the peak m/z 263 and the concentration of γ-CEHC. The stock solution was serial diluted with cell culture medium. Triplicate standard solution was analyzed three times (n ) 9).

Table 2. Monitor Ions and Structures of Tocopherols and Related Metabolites

the micro-SPE column in triplicate at each velocity. The results showed that it had no obvious influence on extraction efficiency (Table 1). Therefore, we selected 10 µL/min as the sample injection velocity in the following experiments, and it took only 18 s for sample loading. The washing flow rate in the range of Analytical Chemistry, Vol. 82, No. 13, July 1, 2010

5683

Figure 5. (A) Mass spectra of extracts from A549 cells incubated with 100 µM γ-T for 72 h. γ-CEHC (B) and 9-carboxychromanol (D) metabolites were detected, and their structures were identified with corresponding tandem mass spectra (C, E).

10-40 µL/min was investigated, and 10 µL/min was selected, because the higher flow rates caused the loss of sample at different degrees. The elution flow rate could influence the intensity of ESIQ-TOF-MS, and the flow rate of 100 µL/h was found to be an appropriate rate, because the micro-SPE column was in good condition and the signal-to-noise ratio was good under this elution flow rate. Moreover, repeated extraction experiments showed that the integrated on-chip SPE column could be used at least four times without reducing the extraction efficiency. Under the optimum conditions, γ-CEHC (one metabolite of γ-T) was used to evaluate the sensitivity of online metabolite detection on the microfluidic devices. The serial diluted γ-CEHC solution was concentrated and detected according to the processes detailed above. Figure 4 shows the relationship between the intensity of the peak of m/z 263 and the concentration of γ-CEHC with the lower concentration enlarged (inset). The intensity of 5684

Analytical Chemistry, Vol. 82, No. 13, July 1, 2010

the monitor ion increased with the increasing concentration of γ-CEHC. The calibration curve obtained by plotting the peak areas was linear in the range of 0.01-5 µM, and the fitting formula was Y ) 5.66 × 102 + 1.2 × 104X, while the R2 ) 0.993. The results showed that the limit of detection was 2.6 ppb. Identification of Tocopherols Metabolites in A549 Cells. A chip-based SPE column combined with a ESI-Q-TOF mass spectrometer was applied for cellular metabolite analysis, offering several advantages that the traditional methods could not realize, such as high integration, simplified sample pretreatment, fast and easy analysis, high sensitivity, and high throughput. In addition, MS/MS could be used to characterize and identify metabolites with sufficient sensitivity. In light of a recent finding that vitamin E forms were metabolized to long-chain carboxychromanols in A549 cells, the metabolites of tocopherols in a cellular system were further characterized using ESI-Q-TOF-MS with a fully integrated

on-chip sample pretreatment procedure. A different concentration of tocopherols could be used to simultaneously stimulate A549 cells inside different channels by integrating several parallel channels on a microfluidic device which provided the potential for high throughput analysis. Because the metabolites of tocopherols are prone to oxidation, it is important to examine whether these compounds are stable over incubation time. γ-CEHC, as a model, was diluted in DMEM and then incubated at 37 °C in a CO2 incubator for 3 days. After detection by ESI-Q-TOF-MS, it was found that about 90% γ-CEHC was recovered. The result indicated that metabolites of tocopherols were stable, so that they could be further analyzed. When tocopherols were incubated with A549 cells for 72 h, long-chain carboxychromanol, 9-carboxychromanol, and a trace level of γ-CEHC were metabolized from γ-T, but no metabolites from R-T could be detected. Monitor ions and structures of tocopherols and metabolites from A549 cells were shown in Table 2. The peak ion of m/z 375.25 and the fragment of m/z 149.05 indicated the existence of 9-carboxychromanol, while monitor ion of m/z 263.13 and the fragment of m/z 149.06 showed the existence of γ-CEHC (Figure 5). The entire online detection time only needed about 8 min. 9-Carboxychromanol appeared to be efficiently secreted into the medium and was the predominant metabolite of γ-T in A549 cultures, which was consistent with previous observations by You et al.38 As shown in Figure 6A, when A549 cells were incubated with the increasing concentration of γ-T for 72 h, the accumulation of 9-carboxychromanol metabolite from γ-T in culture medium increased, but the metabolite of γ-CEHC had indistinct variation. However, the amount of total metabolites increased proportionally to the concentration of γ-T. When A549 cells were incubated with 100 µM γ-T with the incubation time prolonged from 24 to 72 h, the concentration of the 9′-COOH metabolite from γ-T increased, as shown in Figure 6B. These results demonstrated the possibility of performing cellular metabolism with a microvolume sample and detecting the metabolic products of A549 cells using ESI-Q-TOF-MS with a fully integrated on-chip sample pretreatment procedure. CONCLUSIONS An integrated microfluidic device was developed for drug metabolite characterization with online ESI-Q-TOF-MS. The microfluidic device offers several clear advantages over conventional drug metabolism methods, including small reagent consumption, simplified sample pretreatment, and rapid detection (8 min). The utilization of ESI-Q-TOF-MS provides a sensitive, selective detection strategy, which can also provide structural information. Furthermore, the throughput could be increased by integration of several parallel cell culture chambers with the same number of on-chip SPE columns. For complex metabolism mixtures,

Figure 6. (A) Dependence of the accumulation of 9-carboxychromanol and γ-CEHC metabolites in the culture medium of A549 cells incubated with varied concentrations of γ-T from 50 to 250 µM for 72 h on-chip. (B) Time-dependent accumulation of 9-carboxychromanol metabolite in cultured A549 cells on-chip. γ-T (100 µM) was incubated with A549 cells for 24, 48, and 72 h. The metabolites were detected by ESI-Q-TOF-MS on line with a fully integrated on-chip sample pretreatment procedure. The standard error bars mean the variation of three individual experiments.

gradient elution and longer SPE columns can be designed to achieve separation and concentration prior to ESI-Q-TOF-MS detection. All these could be easily achieved and demonstrated that the developed microfluidic device could be a potential useful tool for the drug metabolism assay. ACKNOWLEDGMENT This work was supported by National Natural Science Foundation of China (Nos. 20935002, 90813015). Received for review March 13, 2010. Accepted May 26, 2010. AC101370P

Analytical Chemistry, Vol. 82, No. 13, July 1, 2010

5685