Development of a Microchip-Based Bioassay System Using Cultured

Feb 24, 2005 - Development of a Microfluidic Platform for Single-cell Secretion Analysis Using a Direct Photoactive Cell-attaching Method. Kihoon JANG...
4 downloads 14 Views 700KB Size
Download PDF
Anal. Chem. 2005, 77, 2125-2131

Development of a Microchip-Based Bioassay System Using Cultured Cells Makiko Goto,† Kiichi Sato,‡,§ Atsushi Murakami,| Manabu Tokeshi,§ and Takehiko Kitamori*,†,§,⊥

Department of Applied Chemistry, School of Engineering, The University of Tokyo, Hongo, Bunkyo-ku, Tokyo 113-8656, Japan, Department of Applied Biological Chemistry, School of Agriculture and Life Sciences, The University of Tokyo, Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan, Kanagawa Academy of Science and Technology (KAST), Sakado, Takatsu-ku, Kawasaki, Kanagawa 213-0021, Japan, Citizen Watch Company, Ltd., Shimotomi, Tokorozawa, Saitama 359-8511, Japan, and Core Research for Evolutional Science and Technology (CREST), Japan Science and Technology Corporation (JST), Honcho, Kawaguchi, Saitama 332-0012, Japan

We developed a novel bioassay system using a glass microchip and cultured cells. A microchamber for cell culture and microchannels for reactions and detection were fabricated on a Pyrex glass substrate by photolithography and wet etching techniques. Cell culture, chemical and enzymatic reactions, and detection were integrated into the microchip. To keep different temperatures locally in three areas of the microchip, we designed and fabricated a temperature control device. Nitric oxide released from macrophage-like cells stimulated by lipopolysaccharide was successfully monitored with the microchip, the temperature control device, and a thermal lens microscope. The total assay time was reduced from 24 to 4 h, and detection limit of NO was improved from 1 × 10-6 to 7 × 10-8 M compared with conventional methods. Moreover, the system could monitor a time course of the release, which is difficult to measure by conventional batch methods. We conclude that this system is promising for a rapid bioassay system with very small consumption of cells. Applications of systems, widely known as micro total analysis systems (µ-TAS)1 or lab-on-a-chip,2 have been spreading rapidly. Most of them are based on electrophoretic separation of DNA to realize rapid genetic analysis. By contrast, on-chip integrations of analytical processes, except for CE, have received less attention, though other integrated systems also have desirable characteristics, such as reductions in reagent and sample consumption, required space, and analysis time. We have utilized these advantages to demonstrate a number of applications, including flow injection analyses,3,4 solvent extractions,5-7 and microreactor * Corresponding author: (fax) +81-3-5841-6039; (e-mail) kitamori@ icl.t.u-tokyo.ac.jp. † Department of Applied Chemistry, The University of Tokyo. ‡ Department of Applied Biological Chemistry, The University of Tokyo. § Kanagawa Academy of Science and Technology. | Citizen Watch Co., Ltd. ⊥ Japan Science and Technology Corp. (1) Harrison, D. J.; von den Berg, A. Micro Total Analysis Systems ‘98; Kluwer Academic Publishers: Dordrecht, 1998. (2) Graves, D. J. Trends Biotechnol. 1999, 17, 127-134. (3) Sato, K.; Tokeshi, M.; Kitamori, T.; Sawada, T. Anal. Sci. 1999, 15, 641644. 10.1021/ac040165g CCC: $30.25 Published on Web 02/24/2005

© 2005 American Chemical Society

processes.8,9 In these systems, scale merits of the microspace, i.e., a short diffusion distance, a large specific interface area, and a rapid and efficient reaction, were used to full advantage. Moreover, by combination of the microunit operations connected with pressure-driven flow, a complex chemical system, known as continuous flow chemical processing (CFCP), can be constructed.5 We have reported that CFCP has a great potential to realize various chemical systems consisting of reaction, extraction, and analysis easily and efficiently without troublesome mechanical operations. The major advantages of these microchip-based systems accrue especially to biochemical applications. We have demonstrated rapid and sensitive immunoassay systems for protein analyses10-12 and enzyme reactor use.13 Microchip techniques also appear to provide some advantages for cellular biochemical systems, because the scale of the liquid microspace inside the microchip is fitted to the size of the cells. For instance, in a microchamber fabricated on a microchip, rapid and secure exchange of media or reagents will be achieved by simple operations under continuous medium flows. Moreover, if both a microchamber for cell culture and microchannels for analyses of the conditioned medium are fabricated on a microchip, analyses of the medium become possible without transference into another microchip. Therefore, mechanical operations or handling procedures of the assay are simplified by integration into a microchip. (4) Sorouraddin, H. M.; Hibara, A.; Proskrunin, M. A.; Kitamori, T. Anal. Sci. 2000, 16, 1033-1137. (5) Tokeshi, M.; Minagawa, T.; Kitamori, T. Anal. Chem. 2000, 72, 17111714. (6) Tokeshi, M.; Minagawa, T.; Kitamori, T. J. Chromatogr., A 2000, 894, 1923. (7) Hisamoto, H.; Horiuchi, T.; Tokeshi, M.; Kitamori, T. Anal. Chem. 2001, 73, 1382-1386. (8) Hisamoto, H.; Saito, T.; Tokeshi, M.; Hibara, A.; Kitamori, T. Chem. Commun. 2001, 2662-2663. (9) Ueno, M.; Hisamoto, H.; Kitamori, T.; Kobayashi, S. Chem. Commun. 2003, 936-937. (10) Sato, K.; Tokeshi, M.; Odake, T.; Kimura, H.; Ooi, T.; Nakao, M.; Kitamori, T. Anal. Chem. 2000, 72, 1144-1147. (11) Sato, K.; Tokeshi, M.; Kimura, H.; Kitamori, T. Anal. Chem. 2001, 73, 12131218. (12) Sato, K.; Yamanaka, M.; Takahashi, H.; Tokeshi, M.; Kimura, H.; Kitamori, T. Electrophoresis 2002, 23, 734-739. (13) Slyadnev, M. N.; Tanaka, Y.; Tokeshi, M.; Kitamori, T. Anal. Chem. 2001, 73, 4037-4044.

Analytical Chemistry, Vol. 77, No. 7, April 1, 2005 2125

Although several papers about a microchip-based cell sorter14 or microchip CE analysis of cellular components15 have been published, microsystems utilizing specific functions of cultured cells have not appeared yet. To realize microchip-based cellular systems, stable long-term cell culture in a chip is necessary and this has not been reported except in a limited way. Recently, culture of yeast cells was achieved by a three-dimensional flow control in a microchip,16 and we successfully cultured a human hepatoma cell line in a glass microchip.17 A bioassay using cultured cells is one of the most important analytical methods in a search for new drugs, a safety evaluation of foods, drugs, and chemical compounds, pharmacokinetics, and basic biochemical studies. In the assay system, specific biological effects can be monitored by means of specific cellular responses induced by very low concentration of chemical substances, which is not achieved by ordinary chemical analyses.18 However, because conventional bioassay methods are rather time-consuming and troublesome and require many precious cells and reagents, a novel system resolving these problems is desired. In particular, there is a greater demand for rapid bioassay with small resource consumption as research in the life sciences progresses. To overcome the drawbacks of conventional assays, system integration into a microchip seems to be effective, because microchipbased chemical systems enhance reaction efficiency and some different unit operations can be easily integrated by CFCP. To realize microchip-based bioassay systems, there are some important technical problems to solve. The most important one is longterm cell culture in a restricted part of the microchip while retaining functions of the cells. For this, medium flow cell culture with a highly organized microfluidic system and a temperature control device that can control the temperature locally in one part of a microchip are necessary. In this study, a microchip-based bioassay system that realizes all processes required for a bioassay, i.e., from cell culture to analysis on a microchip, was developed. We designed a monitoring system for nitric oxide released from stimulated mouse peritoneal macrophages as a model case; this process is widely used in screening of immunostimulation drugs.19,20 The chip has microfabricated structures for cell culturing, carrying out the enzymatic reaction (nitrate reduction) and the colorimetric diazo coupling reaction, and detecting with a thermal lens microscope (TLM). Moreover, to realize a more efficient assay, we designed and fabricated a local temperature control device, which can keep different optimum temperatures in three areas on the microchip. EXPERIMENTAL SECTION Microchip Fabrication. The concept of a microchip-based bioassay system is illustrated in Figure 1. To realize an assay on a chip, all processes required for the assay, i.e., cell culture, (14) Fu, A. Y.; Chou, H. P.; Spence, C.; Arnold, F. H.; Quake, S. R. Anal. Chem. 2002, 74, 2451-2457. (15) McClain, M. A.; Culbertson, C. T.; Jacobson, S. C.; Allbritton, N. L.; Sims, C. E.; Ramsey, J. M. Anal. Chem. 2003, 75, 5646-5655. (16) Peng, X. Y.; Li, P. C. H. Anal. Chem. 2004, 76, 5273-5281. (17) Tanaka, Y.; Sato, K.; Yamato, M.; Okano, T.; Kitamori, T. Anal. Sci. 2004, 20, 411-413. (18) Umezawa, Y.; Ozawa, T.; Sato, M. Anal. Sci. 2002, 18, 503-516. (19) Amano, F.; Akamatsu, Y. Infect. Immun. 1991, 59, 2166-2174. (20) Chi, D. S.; Qui, M.; Krishnaswamy, G.; Li, C.; Stone, W. Nitric Oxide 2003, 8, 127-132.

2126 Analytical Chemistry, Vol. 77, No. 7, April 1, 2005

Figure 1. Concept of the microchip-based bioassay system.

Figure 2. (A) Illustration and (B) photograph of the bioassay microchip. (C) An enlarged illustration of the microchamber for cell culture.

chemical stimulation, chemical and enzymatic reactions, and detection, were integrated into the microchip. The chip design is shown in Figure 2A. The chip had a microchamber for cell culture, microchannels for chemical and enzymatic reactions, and a detection area with six holes: one medium inlet, three reagent inlets, one cell introduction hole, and one outlet. The chip was composed of two Pyrex glass plates (30 mm × 70 mm), i.e., cover and bottom plates with thickness of

200 and 700 µm, respectively. The cover plate included the six holes: five were for the medium and reagent inlets and the outlet, each with a diameter of 500 µm; and one was for the cell introduction with a diameter of 1 mm. The bottom plate had microchannels and the microchamber for cell culture, which were designed by CAD software and fabricated by photolithography and wet etching. Details of the microchip fabrication were described previously.21 Figure 2B shows a photo of the bioassay microchip. The channels were 200 µm in width and 100 µm in depth. The microchamber had the shape shown in Figure 2C and dimensions of 100 µm in depth, 5 mm in length, and maximal width of 1 mm. Its volume was ∼0.3 µL. To retain cells in the microchamber, the microchip had a shallow channel 140 µm in width, 10 µm in depth, and 130 µm in length, which was designed to use as a dam. At first the microchamber and microchannels were etched with a hydrofluoric acid solution, and then the dam was etched. Etching time for the dam was much shorter than the first etching step, and therefore, the dam was shallower than the microchamber and microchannels. Finally, the two plates were thermally laminated. Materials. Male mice of inbred stains C3H/HeN (6-10 weeks old) were purchased from Nippon Bio-Supply Center (Tokyo, Japan). Mouse macrophage-like cell line J774.1 was obtained from the Cell Research Center for Biomedical Research, Tohoku University (Sendai, Japan). Lipopolysaccharide (LPS), penicillin, and streptomycin were purchased from Sigma-Aldrich Co. (St. Louis, MO). The spontaneous NO donor reagent, 3-(2-hydroxy1-methyl-2-nitrosohydrazino)-N-methyl-1-propanamine (NOC7), was from Dojindo (Kumamoto, Japan), and Griess reagents (sulfanilamide and N-1-naphthylethylendiamine) and nitrate reductase were obtained from the NO2/NO3 Assay Kit (Dojindo). Collagen for coating microchambers was Cellmatrix Type 1-P (Nitta Gelatin, Japan). D-MEM/F-12 medium, fetal bovine serum (FBS), and phosphate-buffered saline (PBS) were purchased from Invitrogen Corp. (Carlsbad, CA). Apparatus. We have developed a laser-induced TLM, which is especially useful for ultrasensitive determination in a microspace.22,23 The TLM was comprised of a microscope with two laser-oscillation apparatuses and other optical devices. The excitation beam was the 532-nm emission line of a YAG laser (CrystaLaser, Reno, NV) with output power of 50 mW, and its intensity was modulated by a mechanical chopper with a modulation frequency at 1.03 kHz. A He-Ne laser (Melles Griot, Carlsbad, CA; output power of 15 mW) with an emission line of 632.8 nm was used for a probe beam. The two beams were made coaxial by a dichroic mirror and tightly focused by an objective lens (Nikon Corp., Tokyo, Japan; 10×, NA 0.25). The transmitted beams were collected by a condenser lens and filtered. Only the probe beam intensity was monitored by a photodiode. The preamplified signal from the photodiode was synchronously amplified with a lock-in amplifier LI-575 (NF Corp., Yokohama, Japan). (21) Hibara, A.; Tokeshi, M.; Uchiyama, K.; Hisamoto, H.; Kitamori, T. Anal. Sci. 2001, 17, 89-93. (22) Uchiyama, K.; Hibara, A.; Kimura, H.; Sawada, T.; Kitamori, T. Jpn. J. Appl. Phys. 2000, 39, 5316-5322. (23) Tokeshi, M.; Uchida, M.; Hibara, A.; Sawada, T.; Kitamori, T. Anal. Chem. 2001, 73, 2112-2116.

Temperature Control Device. To keep the cell culture and reaction areas on a microchip at different temperatures, we developed a Peltier element-based temperature control device. The device was designed and fabricated to keep in contact with the bottom plate of the microchip and control the temperatures of these areas individually. Small Peltier elements (1 mm × 3 mm) were arranged on a plastic plate with heat sinks. Then copper plates were put on the Peltier elements (Figure 3A, B). Thermistors were embedded in the copper plates. Peltier elements and thermistors were connected to temperature controllers (Figure 3C). Cell Culture in a Microchip. The microchip was washed with 0.1 M NaOH, 0.1 M HCl, and pure water, successively, and then it was autoclaved at 120 °C for 15 min. The inner walls of the microchamber were coated with 0.005% collagen in 0.12 M HCl solution. Next, the microchip was placed in an incubator for 1 h, it was removed, and then the inner walls were rinsed with PBS. Peritoneal macrophages were obtained from a mouse. The cells were washed with PBS and suspended in 200 µL of DMEM/F-12 medium containing 50 units/mL of penicillin, 50 µg/mL of streptomycin, and 5% heat-inactivated FBS. Cell concentration of this suspension was 1 × 103 cells/µL. The suspension was dropped into the cell inlet hole (Figure 4), and then the inlet hole was covered with fresh medium. On the other hand, J774.1 cells were suspended in the medium (1 × 104 cells/µL) and the suspension was introduced into the microchamber in the same way. After introducing the cells into the microchamber, the microchip were incubated at 37 °C in a humidified atmosphere with 5% CO2. Assay for NO. Assay of NO was carried out using nitrate reductase and Griess reagents.24 The NO released from cells to the medium was oxidized by dissolved O2, forming NO2- and NO3-. Then, for the NO assay in the microchannels, NO3- was reduced to NO2- by nitrate reductase. Next, all the NO2- was reacted with Griess reagents and the resulting colored product was detected by TLM with a YAG laser (533 nm) as the excitation laser and a He-Ne laser (632 nm) as the probe. All reactions in the microchip are shown in Figure 5. Before the actual assay using cells, NO donor reagent, NOC7, was used as a standard sample of NO solution in order to optimize the system and calibrate the concentration of NO. This assay system consisted of a microchip clamped by a holder, syringes, and fluorocarbon capillary tubes. The connection between the microchip and a syringe via capillary tubing has been described previously.21 Every reagent was introduced into the microchip from each syringe through fluorocarbon capillaries by microsyringe pumps. By using the system, precise control of very small amount of solutions could be realized with very simple and easy operations. The standard sample was introduced into the microchip from the inlet hole located upstream from the microchamber. Nitrate reductase and Griess reagents (sulfanilamide and N-1naphthylethylendiamine) were introduced from reagent inlets, respectively, to react with the sample in this order. The liquid flow was controlled with microsyringe pumps (KD Scientific, Boston, MA) and Hamilton gastight syringes. Capillaries were made of fluorocarbon polymer (IWASE, Yokohama, Japan) and connected to the inlet and outlet holes of the microchip. The (24) Green, L. C.; Wangner, D. A.; Glogowski, J.; Skipper, P. L.; Wishnok, J. S.; Tannenbaum, S. R. Anal. Biochem. 1982, 126, 131-138.

Analytical Chemistry, Vol. 77, No. 7, April 1, 2005

2127

Figure 3. (A) Illustration and (B) photo showing the arrangement of Peltier elements. (C) Photo of the temperature control device.

Figure 4. Cross section of the microchamber. Cell suspension was dropped from the cell inlet hole into the microchamber. The suspension was held back because of the dam structure.

capillary for reagent introduction was connected to the syringes, and the outlet capillary was connected to a waste reservoir. The sample inlet capillary was inserted into the sample reservoir. Details of the fluidic control system were described previously.21 Standard samples of NO solutions were also assayed on a bulk scale using the NO2/NO3 Assay Kit and the standard protocol. Samples were in a 96-well microplate, and nitrate reductase, sulfanilamide, and N-1-naphthylethylendiamine were introduced using micropipets. Total reaction time of the nitrate reductase reaction and diazo coupling reaction was ∼2 h. Then the resulting products were measured at 492 nm with a microplate reader Multiskan Jx (Thermo Labsystems, Vantaa, Finland). Monitoring of NO Released from Macrophages. After the microchip was washed, autoclaved, and then coated with collagen solution as described above, the cell suspension was introduced into the microchamber from the cell inlet hole. This was followed by another incubation at 37 °C with 5% CO2 from 2 h to overnight. After incubation, the microchip was clamped with a holder and the cell inlet hole was covered with a poly(dimethylsiloxane) sheet (500 µm in thickness). This microchip was put on the temperature 2128 Analytical Chemistry, Vol. 77, No. 7, April 1, 2005

Figure 5. Chemical processes carried out in the microchip for bioassay of macrophage-stimulating agent.

control device to keep the optimum temperature for three areas, i.e., cell culture, enzymatic reaction, and chemical reaction areas. Then fresh medium was introduced through the sample inlet hole located upstream from the microchamber, and nitrate reductase and Griess reagents were introduced from the reagent inlets, respectively. For the assay, the medium was switched by a switching valve to the sample solution, which contained macrophage stimulating chemicals. As a model agent, LPS, one of the strongest activators of macrophages, was used; the LPS was

areas on a microchip at different temperatures, something that was very difficult by using conventional temperature control devices. Cell Culture in a Microchip. We have previously reported a microchip-based cell culture.25 This microchip had a square microchamber for the cell culture. However, with this shape, air bubbles frequently remained in the microchamber corners and caused serious damage to the cultured cells. Moreover, even if the corners were filled with the medium, they became dead space for the medium flow in case of continuous flow cell culture and bioassay. Therefore, a shape with corners should be avoided. We designed a streamlined microchamber without corners (Figure 2C) and fabricated it. Figure 8 shows micrographs of mouse Figure 6. Complete system for the microchip-based bioassay.

dissolved in the medium. The TLM detection was started at the same time as LPS was introduced into the microchip. All reagents were introduced at the same flow rate (0.2 µL/min) by microsyringe pumps (Figure 6). The chip could be reused after washing it using acidic and alkaline solutions, successively. RESULTS AND DISCUSSION Temperature Control Device. Figure 7A shows the temperature control device, the microchip, and the microchip holder. The areas of copper plates, i.e., the temperature-controlled areas of cell culture, enzymatic reaction, and diazonium-forming reaction, were 27, 140, and 247 mm2, respectively. After optimization, we decided to control the temperatures at 37, 50, or 20 °C for cell culture, enzyme reaction, or diazonium-forming reaction areas, respectively. The differences of temperatures between the surface of the copper plates and the upper surface of the microchip were about 1.6, 1.3, and 0.3 °C for 37, 50, and 20 °C areas, respectively, and the temperatures were kept stable. Therefore, the inside temperature of the microchannels and the microchamber were thought to be kept intermediate temperatures. Figure 7B shows a thermograph of the microchip surface. Although distances between copper plates were less than 3 mm, temperatures on the microchip surface at the edge of each area differed less than (4 °C from the preset temperatures and could be kept stable. This temperature control device could successfully keep very small

Figure 8. Micrographs of (A) mouse peritoneal macrophages and (B, C) J774.1 cells in the microchip. The dam successfully kept cells in the microchamber.

peritoneal macrophages and J774.1 cells in the microchamber. We successfully cultured from several hundred to several thousand cells; they grew well all over the microchamber and no air remained. When the cell suspension was dropped into the cell

Figure 7. (A) Photo of the temperature control device with the microchip. (B) Thermograph of the microchip surface.

Analytical Chemistry, Vol. 77, No. 7, April 1, 2005

2129

Figure 9. Micrographs of J774.1 cells cultured in the microchip for 2 days (A) under continuous medium flow conditions and (B) static conditions. The cells stained in blue were dead.

inlet hole, the cell suspension spread throughout the entire microchamber, but cells did not go over the dam and flow downstream from it (Figure 8C). To avoid cell leakage into the downstream regions, static culturing without medium flow was required for the first 2 h. After the static culturing, cells were well attached on the under surface of the microchamber and did not flow downstream from the dam even if medium flow was then used. Our method restricted cell culture to the culture chamber. Most of the cells, however, did not grow well or showed weak activity after long-term static cultivation in the microchamber because of lack of oxygen and nutrients, accumulation of cell wastes, or change of pH. Figure 9 shows micrographs of trypan blue-stained J774.1 cells, which were cultured for 2 days with or without medium flow in the microchamber. The cell survival rate was much higher, and clear cell proliferation was observed under the continuous medium flow condition. In the microchannel and the microchamber, flow rate was so slow (12.7 mm/min, in the microchannel) that medium was diffused enough in the entire microchamber that cells lived well. We observed no difference in cell growth between in the center of chamber and near the wall. These results indicated that the cells cultured in the microchamber require a continuous fresh medium flow and the system is effective for a long-term cultivation of cells having high metabolic rate and oxygen consumption. In the case of NO monitoring, the minimum precultivation time was from 2 h to overnight, and the continuous flow culture was not always necessary. Assay for NO. Nitric oxide dissolved in the medium was determined by a three-step reaction with TLM detection in the microchip. When all reagents required for the determination of NO were introduced at the same rate of 0.2 µL/min, it took ∼90 s for the NO solution to flow through the microchamber. Reaction times of the nitrate reductase reaction, diazonium-forming reaction, and diazo coupling reaction were 70, 100, and 60 s, respectively. Therefore, the total time from introduction of reagents to TLM detection was ∼5.5 min, whereas ∼2 h is required in the conventional assay using a commercial assay kit. Figure 10 shows the calibration curves of NO assayed on the bulk scale (A) and in a microchip (B). The lower limit of NO detection in the standard sample was 7 × 10-8 M (S/N ) 2), which was better than on the bulk scale (1 × 10-6 M). Reduction of the reaction time was thought to be realized mainly by optimization of the reaction temperature of each reaction and mixing rate of the reagents. Highly sensitive determination was a benefit of the high reaction efficiency and the ultrasensitive detector, TLM. 2130 Analytical Chemistry, Vol. 77, No. 7, April 1, 2005

Figure 10. Calibration curves of NO dissolved in the medium (A) determined on the bulk scale and (B) in a microchip.

Figure 11. Typical results of NO released from J774.1 cells. Signal intensity was converted to the NO concentration by the calibration curve of Figure 10B.

The fabricated microchip has three reaction areas. Therefore, the chip is applicable for other reaction systems that consist of three reactions or less. Each reaction can be optimized in various ways. The reaction time can be controlled by flow rate of each reagent and length and cross section of the microchannel. The reaction rate can be regulated by the mixing rate of the reagents and reaction temperatures. Therefore, we conclude that the microchip is very useful for a variety of bioassays that require various reactions. Moreover, by integration of other analytical methods, i.e., extraction, separation by chromatography, or immunoassay, very sophisticated systems can be realized. Monitoring of NO Released from Macrophages. Figure 11 shows typical monitoring results for NO released from J774.1 cells stimulated by a macrophage-stimulating agent, LPS. Rather broad peaks continued to appear for 1 h with weak and long tailing. Reference to the calibration curve showed that 3 pmol of NO was released. Since there were ∼3000 cells in the microchamber, we calculated that 1 fmol of NO/cell was released, which agreed with the widely accepted value.26-28 From these results, we conclude that the system is useful for the bioassay of the macrophageactivating activity of sample solutions. With the microchip-based system, the time course of the cell response at the early stage of activation of the macrophage-like (25) Tamaki, E.; Sato, K.; Tokeshi, M.; Sato, K.; Aihara, M.; Kitamori, T. Anal. Chem. 2002, 74, 1560-1564. (26) Saito, S.; Matsuura, M.; Tominaga, K.; Kiriike, T.; Nakano, M. Eur. J. Biochem. 2000, 267, 37-45. (27) Deakin, A. M.; Payne, A. N.; Whittle, B. J. R.; Moncada, S. Cytokine 1995, 7, 408-416. (28) Amano, F.; Noda, T. FEBS Lett. 1995, 368, 425-428.

cells could be monitored; conventional methods have not been able to monitor this. Such results will be interesting for basic biochemical and immunological studies. We successfully obtained results from as few as 3000 cells in a microchamber, whereas 105-106 cells/well are required in conventional batch methods. Fewer numbers of required cells is preferable especially in the case of precious primary cells. Moreover, the consumption of the analyte and reagents was also reduced remarkably, which made the analysis cost-effective. The total assay time including the cultivation of the cells was reduced from 24 to 4 h, and real-time monitoring of NO released from cells was also realized. In conventional batch methods, it is very difficult to monitor cellular responses in real time because it takes a very long time to assay the conditioned medium after they are obtained. Therefore, we conclude that this microchip-based bioassay system has great potential for new assay schemes. CONCLUSIONS In this study, we realized a novel bioassay system using a microchip and cultured cells. All processes of the bioassay, i.e., cell culture, chemical stimulation of cells, chemical and enzymatic reactions, and detection, were successfully integrated on a microchip. We developed the temperature control device that could keep different temperatures for individual areas on the microchip. To realize highly integrated systems on a microchip, it is important that each process is performed under the best temperature condition, but with conventional temperature control devices, local control is not easy because the area of each process is very small and areas are close to each other. However, by using our temperature control device, three areas with separation distances of less than 3 mm were successfully kept at different temperatures. Mouse peritoneal macrophages and mouse macrophage-like cell line could be cultured successfully in the entire microculture chamber. Cell culture under medium flow condition was effective for long-term cultivation of cells. The long-term culture is very useful when cells are used for biosensors or bioreactors. Nitric oxide dissolved in the medium was successfully determined by the microchip-based flow analysis, which consisted of three reactions, and determination limit of NO was also improved by using the microchip, temperature control device, and TLM. Reaction times in microchannels were much shorter than on the bulk scale.

By combining these techniques, we have succeeded in monitoring NO released from cells cultured under the continuous flow condition. With the developed microsystem, immunostimulation activity of the sample could be assayed by using 3000 cells within 4 h, while 105-106 cells and 24 h are required for an assay in conventional batch methods. In addition, the system featured simplicity of operation; many laborious pipetting operations are necessary for conventional methods, but the integrated system required only switching of the microsyringe pumps and a valve. Moreover, the time course of NO release from macrophage-like cells was successfully monitored; this is very difficult by conventional batch methods. This system is a novel experimental system that allows easy monitoring of real-time cellular responses with high sensitivity. The developed system could evaluate various effects of a sample solution on cells by monitoring cell responses. It is possible to apply this system to many other bioassays using cultured cells when microchannels and reagent inlets are designed according to needs. We concluded the system has a great potential as a rapid bioassay system for drug screening and risk evaluation of chemical compounds. We expect it will become a practicable system by improving the system instruments, making a detailed optimization and multiarraying microchips. ACKNOWLEDGMENT We are grateful to Prof. Akira Okubo, Ms. Namiho Yasuda (University of Tokyo), and Mr. Yoshikuni Kikutani (KAST) for useful discussions and technical support. This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports and Culture of Japan and the Nanotechnology Project of the Ministry of Agriculture, Forestry and Fisheries of Japan. We gratefully acknowledge financial support from the New Energy and Industrial Technology Development Organization (NEDO) of the Ministry of Economy, Trade and Industry, Japan.

Received for review September 27, 2004. Accepted December 21, 2004. AC040165G

Analytical Chemistry, Vol. 77, No. 7, April 1, 2005

2131