Anal. Chem. 1999, 71, 1485-1490
Accelerated Articles
An Integrated Microfabricated Device for Dual Microdialysis and On-Line ESI-Ion Trap Mass Spectrometry for Analysis of Complex Biological Samples Fan Xiang, Yuehe Lin, Jenny Wen, Dean W. Matson, and Richard D. Smith*
Environmental Molecular Sciences Laboratory, Pacific Northwest National Laboratory, P.O. Box 999, Richland, Washington 99352
A microfabricated dual-microdialysis device in a single integrated microfabricated platform was constructed using laser micromachining techniques for the rapid fractionation and cleanup of complex biological samples. On-line dual microdialysis and ESI-MS of biological samples was demonstrated using an ion trap mass spectrometer. The mass spectra obtained demonstrated the efficiency of dual microdialysis for removing both high-molecular-weight and low-molecular-weight species that interfere with effective ESI-MS analysis of target biopolymers. Signal-tonoise ratios were also greatly improved compared to direct sample infusion. In addition to its compactness, negligible dead volume, and robustness, the device can be used at a flow rate of only 200 nL/min, an order of magnitude lower than that obtained previously. This reduced sample consumption and improved sensitivity with ESI-MS. The results suggest the potential for integration of such microfabricated devices with other sample manipulations for the rapid ESI-MS analysis of complex biological samples.
The ever-increasing demand for the capability to handle small samples with improved sensitivity and higher throughput has stimulated research into the miniaturization of analytical instrumentation. Microfabricated devices (“chips”) are increasingly recognized as a convenient platform on which to execute “nanoscale” separations, chemical procedures, and other sample manipulations. Separations on chips have been accomplished based * Corresponding author: (tel) 509-376-0723; (fax) 509 376-7722; (e-mail)
[email protected]. 10.1021/ac981400w CCC: $18.00 Published on Web 03/19/1999
© 1999 American Chemical Society
on zone electrophoresis,1-6 open-channel chromatography,7 micellar electrokinetic chromatography,8 free-flow electrophoresis,9 and gel electrophoresis.10-11 Integrated devices, such as a hybrid device that performs both PCR amplification and separation, have been reported.12 Recently, cell lysis was demonstrated on a microchip with subsequent multiplex PCR amplification and electrophoretic separation.13 Most of the devices developed to date have been fabricated from glass using photolithography, wet chemistry, and chemical-etching processes adapted from established microelectronics manufacturing procedures.14 Since some microchemical applications will be limited by the material properties of glass or quartz, the use of polymeric materials for microfabricated (1) Harrison, D. J.; Manz, A.; Fan, Z.; Lu ¨ di, H.; Widmer, H. M. Anal. Chem. 1992, 64, 1926-1932. (2) Manz, A.; Harrison, D. J.; Verpoorte, E.; Fettinger, J. C.; Paulus, A.; Lu ¨ di, H.; Widmer, H. M. J. Chromatogr. 1992, 593, 253-258. (3) Seiler, K.; Harrison, D. J.; Manz, A. Anal. Chem. 1993, 65, 1481-1488. (4) Harrison, D. J.; Glavina, P. G. Sens. Actuators B 1993, 10, 107-116. (5) Jacobson, S. C.; Hergenro¨der, R.; Koutny, L. B.; Warmack, R. J.; Ramsey, J. M. Anal. Chem. 1994, 66, 1107-1113. (6) Jacobson, S. C.; Hergenro ¨der, R.; Koutny, L. B.; Ramsey, J. M. Anal. Chem. 1994, 66, 1114-1118. (7) Jacobson, S. C.; Hergenro ¨der, R.; Koutny, L. B.; Ramsey, J. M. Anal. Chem. 1994, 66, 2369-2373. (8) Moore, A. W., Jr.; Jacobson, S. C.; Ramsey, J. M. Anal. Chem. 1995, 67, 4184-4189. (9) Raymond, D. E.; Manz, A.; Widmer, H. M. Anal. Chem. 1994, 66, 28582865. (10) Effenhauser, C. S.; Paulus, A.; Manz, A.; Widmer, H. M. Anal. Chem. 1994, 66, 2949-2953. (11) Woolley, A. T.; Mathies, R. A. Anal. Chem. 1995, 67, 3676-3680. (12) Woolley, A. T.; Hadley, D.; Landre, P.; deMello, A. J.; Mathies, R. A.; Northrup, M. A. Anal. Chem. 1996, 68, 4081-4086. (13) Waters, L. C.; Jacobson, S. C.; Kroutchinina, N.; Khandurina, J.; Foote, R. S.; Ramsey, J. M. Anal. Chem. 1998, 70, 158-162. (14) Ko, W. H.; Suminto, J. T. In Sensors; Gopel, W., Hasse, J., Zemmel, J. N., Eds.; VCH: Weinhein, Germany, 1989; Vol. 1, pp 107-168.
Analytical Chemistry, Vol. 71, No. 8, April 15, 1999 1485
devices has also recently attracted increasing attention.15 Electrospray ionization mass spectrometry (ESI-MS) is known to be useful for the sensitive analysis of biological molecules.16 The combination of ESI-MS with “nanoscale” sample manipulations in small microfabricated devices is particularly attractive due to the speed and sensitivity than can be achieved with MS, as well as the amenability of ESI-MS to the low flow rates and small sample consumption characteristic of such devices. For example, it is well recognized that low flow rates when using CE-MS and smaller capillary diameters with ESI-MS provide enhanced sensitivity.17-19 Recently, the potential for use of microchips in combination with ESI-MS has been reported by several groups.20-24 A microfabricated device was demonstrated for automating the continuous injection approach to protein identification using microelectrospray ion trap mass spectrometry.22 Multiple-channel microchips have been directly interfaced to ESI-MS, demonstrating the potential for high-throughput applications.23,24 The analysis of cellular extracts and other complex biological samples is of great interest for many applications in, for example, biotechnology, pathogen detection, medical research, bioremediation, and clinical applications. The potential exists for using ESI-MS to detect low-level biological marker molecules (“biomarkers”) from a complex mixture and to provide detailed structural information or more confident identification of microorganisms using multistage MS.25 Such applications will, however, require that complex samples be readily separated, fractionated, or “cleaned up” to the point where useful mass spectra can be obtained. Interferences from both the high-molecular-weight and low-molecular-weight components in complex mixtures impose limitations for the direct use of ESI-MS for analysis of target biomolecules in such systems. A variety of techniques have been explored for reducing sample matrix contributions and for improving spectrum quality in ESI-MS analyses.26-28 Previously, we reported a novel dual-microdialysis configuration for fast and efficient fractionation and cleanup of complex biological samples for ESI-MS analysis,25 which extends an approach for processing biological samples prior to ESI-MS being developed in our (15) Harrison, D. J., Berg, A. van den, Proceedings of the µTAS’98 Workshop, 1998. (16) Cole, R. B., Ed. Electrospray Ionization Mass Spectrometry; John Wiley & Sons: New York, 1997. (17) Smith, R. D.; Wahl, J. H.; Goodlett, D. R.; Hofstadler, S. A. Anal. Chem. 1993, 65, A574-A584. (18) Wahl, J. H.; Goodlett, D. R.; Udseth, H. R.; Smith, R. D. Anal. Chem. 1992, 64, 3194-3196. (19) Wahl, J. H.; Goodlett, D. R.; Udseth, H. R.; Smith, R. D. Electrophoresis 1993, 14, 448-457. (20) Henry, C. Anal. Chem. 1997, 69, 359A-361A. (21) Ramsey, R. S.; Ramsey, J. M. Anal. Chem. 1997, 69, 1174-1178. (22) Figeys, D.; Ning, Y.; Aebersold, R. Anal. Chem. 1997, 69, 3153-3160. (23) Xue, Q. F.; Foret, F.; Dunayevskiy, Y. M.; Zavracky, P. M.; Mcgruer, N. E.; Karger, B. L. Anal. Chem. 1997, 69, 426-430. (24) Xue, Q. F.; Dunayevskiy, Y. M.; Foret, F.; Karger, B. L. Rapid Commun. Mass Spectrom. 1997, 11, 1253-1256. (25) Liu, C. L.; Hofstadler, S. A.; Bresson, J. A.; Udseth, H. R.; Tsukuda, T.; Smith, R. D. Anal. Chem. 1998, 70, 1797-1801. (26) Little, D. P.; Chorush, R. A.; Speir, J. P.; Senko, M. W.; Kelleher, N. L.; McLafferty, F. W. J. Am. Chem. Soc. 1994, 116, 4893-4897. (27) Stults, J. T.; Marsters, J. C. Rapid Commun. Mass Spectrom. 1991, 5, 359363. (28) Emmett, M. R.; Caprioli, R. M. J. Am. Soc. Mass Spectrom. 1994, 5, 605613.
1486 Analytical Chemistry, Vol. 71, No. 8, April 15, 1999
laboratory.29-32 The initial design for the dual-dialysis sample cleanup device had a dead volume of ∼80 µL, required ∼20 min, and consumed 150 µL of sample (total) during the analysis period. Recently we have developed a laser microfabricated dialysis device for faster and more efficient desalting.33 Initial studies indicated a significant reduction of sample consumption due to the lower flow rates, providing enhanced sensitivity for ESI-MS. In the present work, we have evaluated the feasibility of a new microfabricated dual-microdialysis device in conjunction with on-line ESI-MS. EXPRIMENTAL SECTION Construction of the Microfabricated Device. Figure 1 shows a photograph of the microfabricated dual-microdialysis device and a three-dimensional view illustrating the design and assembly of the device. Two microdialysis membranes were sandwiched between three polymer layers micromachined with serpentine flow channels. The serpentine channels were fabricated on 30 × 30 × 6 mm polycarbonate chips using a Potomac LMT-4000 excimer laser direct-write micromachining system (Potomac Photonics, Lanham, MD). Channel 3 (60 µm deep, 150 µm wide) was machined directly into the polycarbonate chip by using multiple parallel laser passes, and the depth of the channel was controlled by selecting the appropriate laser repetition rate (1 kHz) and machining rate (5 µm/s). One end of channel 3 was connected to the electrospray tip (i.e., emitter) by a “zero dead volume” chromatographic fitting machined into the edge of the center chip, and the other end was connected to channel 2 through a hole of 300-µm diameter drilled through the center chip. Channel 2, of 300-µm width, was formed from a sheet of 125-µm-thick polyimide and subsequently attached to the polycarbonate chip by silicone adhesive. The channel was produced by laser cutting around the channel perimeter followed by stripping the remaining polyimide and adhesive from the center. Channels 1 and 4 were fabricated in the same manner as channel 2 and have 300- and 500-µm widths, respectively. All of fused-silica capillaries for fluid access were connected to the channel headers using fingertight microfittings (Upchurch Scientific, Oak Harbor, WA), which were mounted directly to the planar exposed faces of the chips. Compared to the previous dual microdialyzer, this design not only made the dual-stage microdialysis device highly compact and rugged but also eliminated the dead volumes from the tubing connections between the two stages. A flow diagram for the dual-microdialysis device is shown in Figure 2. Samples were injected through the sample inlet. During operation, a portion of the sample diffuses through the high MWCO membrane and then flows through the hole drilled through the chip into the second microdialysis stage. The remaining sample in the first stage then flows from the end of the channel (e.g., to “waste”). This partial flow appears to effectively (29) Wu, Q.; Liu, C. L.; Smith, R. D. Rapid Commun. Mass Spectrom. 1996, 10, 835-838. (30) Liu, C. L.; Wu, Q. Y.; Harms, A. C.; Smith, R. D. Anal. Chem. 1996, 68, 3295-3299. (31) Muddiman, D. C.; Wunschel, D. S.; Liu, C. L.; Pasa Tolic, L.; Fox, K. F.; Fox, A.; Anderson, G. A.; Smith, R. D. Anal. Chem. 1996, 68, 3705-3712. (32) Liu, C. L.; Muddiman, D. C.; Tang, K. Q.; Smith, R. D. J. Mass Spectrom. 1997, 32, 425-431. (33) Xu, Naxing; Lin, Yuehe; Hofstadler, S. A.; Matson, Dean; Call, C. J.; Smith, R. D. Anal. Chem. 1998, 70, 3553-3556.
Figure 1. (Left) Photographs of a microfabricated dual-microdialysis device. (Right) Exploded view showing construction of the device.
Figure 2. Flow diagram of a microfabricated dual-microdialysis device showing sample inlet, buffer inlet, and microelectrospray tip.
sweep out cellular residue remaining in the sample injection channel and yields a reduced potential for first-stage membrane clogging. The sample entering the second stage is exposed to a countercurrent flow of buffer solution across the low MWCO membrane. The low-molecular-weight matrix and salt present in the sample in the second-stage channel diffuses through the membrane and is removed by the buffer solution, as described previously.33 The cleaned sample then flows directly to the electrospray tip. For an on-line ESI-MS experiment, the high voltage required for ESI was applied to the metal syringe needle used for the buffer delivery. Details of the membranes and flow rates used in specific experiments are discussed in Results and Discussion. Materials and Samples. All of membranes were obtained from Sialomed, Inc. (Columbia, MD) and were made of cellulose ester. All proteins were purchased from Sigma Chemical Co. (St. Louis, MO) and were used without further purification. Ammonium acetate (Fluka Chemika-Biochemika, Ronkonkoma, NY)
was dissolved in deionized water to a concentration of 10 mM and was used as the dialysis counterflow buffer. A 10 mM phosphate-buffered saline solution (PBS; Sigma Chemical Co.) in deionized water, which contained 0.137 M NaCl and 0.0027 M KCl, was used to prepare protein solutions. Escherichia coli strain K12 was cultured to stationary phase in Lenox broth34 in our laboratory. A cellular lysate of E. coli was prepared as follows: 10 mg of pelleted cells was suspended in 1 mL of 10 mM PBS. The suspension was then lysed in a Mini-Bead Beater (Biospec Products, Bartiesville, OK) with 0.1-mm zirconia silica beads at 5000 rpm for 60 s. The lysate was centrifuged at 8160 RCF for 2 min and the supernatant injected into the dual-microdialysis device. Mass Spectrometry. A Finnigan (San Jose, CA) LCQ quadrupole ion trap mass spectrometer was used for all experiments. The heated inlet capillary was maintained at 200 °C. On-line ESIMS was performed in the positive ion mode, and a typical ESI voltage was ∼2 kV. For MS experiments, the maximum sample injection time was 60 ms, and 2 microscans were summed for each scan; 10 scans were averaged for each spectrum shown in Figures 3-5, except those spectra from nondialyzed samples. RESULTS AND DISCUSSION Performance. The motivation behind the dual-microdialysis approach, consisting of two separation stages, as well as the concomitant shortcomings in terms of speed and dead volume of the initial design, have been described previously.25 To evaluate (34) Ausubel, F. M., et al., Eds. Current Protocols in Molecular Biology; John Wiley & Sons: New York, 1998.
Analytical Chemistry, Vol. 71, No. 8, April 15, 1999
1487
Figure 4. Dual-microdialysis on-line ESI-MS spectra of a protein mixture consisting of 30 µM bovine serum albumin, 8 µM cytochrome c, and 2.4 µM ubiquitin in 0.01 M PBS. (A) 100 000, (MWCO) membrane used on the first stage of the dual-microdialysis device; (B) 50 000 (MWCO) membrane used on the first stage of the dual-microdialysis device. Peak assignments: u, ubiquitin; c, cytochrome c; a, bovine serum albumin.
the performance of the dual-microdialysis device described in this report, a protein mixture consisting of 30 µM BSA, 8 µM
cytochrome c, and 2.4 µM ubiquitin in 10 mM PBS was prepared. A 50 000 MWCO membrane and a 8000 MWCO membrane were
Figure 3. ESI-MS spectra of a protein mixture consisting of 30 µM bovine serum albumin, 8 µM cytochrome c, and 2.4 µM ubiquitin in 0.01 M PBS. (A) sample directly infused without passing through the dual-microdialysis device; (B) sample passed by the dual-microdialysis device. Identities of the peaks: u, ubiquitin; c, cytochrome c. 1488 Analytical Chemistry, Vol. 71, No. 8, April 15, 1999
Figure 5. ESI-MS spectra for a E. coli cell lysate. (A) spectrum obtained from the sample directly infused without passing through the dualmicrodialysis device; (B) spectrum obtained using on-line dual-microdialysis sample processing; two expanded views show detail of peaks at m/z 969 and 1124; (C) CID spectrum of m/z 969 at a relative collision energy of 35%; (D) CID spectrum of m/z 1124 at a relative collision energy of 35%.
used in the first- and the second-stage microdialysis, respectively. The sample was injected at 0.5 µL/min and the cleanup buffer at 10 µL/min. As shown in Figure 3A, direct infusion of the protein mixture produced no useful ESI-mass spectrum. After the on-line dual microdialysis in the microfabricated device, the spectrum clearly showed peaks from cytochrome c, and ubiquitin, as well as removal of the unresolved envelope of peaks at higher m/z (Figure 3B). The improvement in spectral quality resulting from the use of the dual-microdialysis device enabled effective peak assignments and accurate molecular weight determinations. For analyzing complex biological samples, initial sample fractionation and cleanup can simplify the analytical procedure. On-line dual-microdialysis using the microfabricated device provides a fast and efficient means to do this. The molecular weight range selected for analysis can be controlled by selection of the appropriate MWCO membranes. Figure 4 shows a comparison between mass spectra of the same protein mixture used above that were obtained using membranes having MWCO values of 100 000 (A) and 50 000 (B) in the first stage of the dualmicrodialysis device. As expected, some portion of the BSA diffused through the 100 000 (MWCO) membrane, but that fraction was effectively blocked by use of the 50 000 (MWCO) membrane. It is believed that the transfer efficiency of proteins passed through the first stage of the dual-microdialysis device depends on the molecular weight and conformation of the proteins, hydrophobicity of individual proteins, solution flow rate, and pressure difference across the membrane. A quantitative study of the efficiency and separation power for the fractionation of the
microfabriacted microdialysis devices is currently being conducted at our laboratory and will be the subject of a future report. Application to Complex Cellular Samples. There is an enormous need for increasingly rapid and sensitive identification of microorganisms across a spectrum of environmental, food safety test, biological, and medical research, as well as pathogen detection. One potentially feasible approach is based upon the mass spectrometric analysis of cellular constituents. However, the capability for applying this technology will depend crucially upon the “front-end” processing required to deliver a tractable sample to the mass analyzer. The use of a microfabricated dual-microdialysis device for sample processing is attractive due to the device ruggedness, the potential for extension to multichannel devices, and the greatly increased speed and ease of automation. To evaluate the front-end processing efficiency of the microfabricated dialysis device, an E. coli lysate with a total protein concentration of ∼1 mg/mL was analyzed by ESI-MS before and after dual microdialysis. Direct infusion of the crude cell lysate produced the spectrum shown in Figure 5A. The spectrum was not useful due to the sample complexity, the high concentration of NaCl, and the resulting low signal-to-noise ratio. After passing the sample through the dual-microdialysis device with 50 000 (MWCO) membrane in the first stage and 8 000 (MWCO) membrane in the second stage, a large improvement in spectral quality was observed (Figure 5B). An ∼20-fold increase of signal-to-noise ratio was obtained. The characteristic peaks from major cellular components were readily discernible in the spectrum and could provide the basis for further MS/MS experiments.25 Lower flow Analytical Chemistry, Vol. 71, No. 8, April 15, 1999
1489
rates inherent to the use of the microfabricated dual-microdialysis unit provided greater sensitivity with the ESI-MS analysis. With the current device, on-line ESI-MS could be conducted at a few hundred nanoliters per minutes, more than 1 order of magnitude reduction compared to the previous dual-microdialysis system.25 Another feature of the current device that should be noted is the reduced dead volume as a result of the elimination of tubing connections between the two stages. Those features allow faster analysis of smaller samples. MS/MS analysis of distinctive ions provides a potential basis for MS peak identification in complex spectra (i.e., serve as “biomarkers”). Panels C and D of Figure 5 show two MS/MS spectra of m/z 969 and 1124 ions in the expanded views of Figure 5B. The total spectrum acquisition time was set to 2 min with maximum 1-s ion injection time for each precursor ion. The actual ion injection time was ion intensity dependent and was determined by the automatic gain control (AGC) feature of LCQ software. The spectra in Figure 5 C,D are an average of 38 scans and provide an excellent signal-to-noise ratio. In the present studies, using a flow rate of 200 nL/min, only 16 µL of sample was consumed during the MS/MS data acquisition from 40 cellular components. We are presently using this approach at our laboratory for the identification of microorganisms based upon the use of MS/MS for automatically selected precursor species over a wide m/z range, to generate a “global” MS/MS spectral display. Our aim is to determine whether distinctive MS/MS spectral patterns can be used to identify biomarkers for specific microorganisms. The detailed description of these MS/MS studies is beyond the scope of this paper and will be presented separately. MS/MS results shown in Figure 5 were chosen to demonstrate the device’s ability to obtain mass spectra from complex biological samples with online sample processing. The combination of low flow rate that can be obtained for the device with ESI and the precursor ion accumulation capability of the ion trap MS enables high sensitivity detection. Even a peak with relatively low intensity in the spectrum
1490 Analytical Chemistry, Vol. 71, No. 8, April 15, 1999
(Figure 5B; m/z 1124) can result in a high-quality MS/MS spectrum, as shown in Figure 5D. CONCLUSIONS We have developed a microfabricated dual-microdialysis device for fast and efficient fractionation and cleanup for on-line ESI-MS analysis of complex biological samples. The device was constructed using laser micromachining technology and is both compact and robust in operation. Mass spectra obtained for model protein mixtures demonstrated excellent performance for removing both high-molecular-weight and low-molecular-weight contaminants. Direct analysis of crude cell lysates was also demonstrated using the dual-microdialysis device, and ESI-MS/MS analysis on selected ions illustrated the potential for the characterization of microorganisms if suitable biomaker species can be identified. Further improvements to such microfabricated devices to enable the use of lower flow rates with on-line ESI-MS would further decrease sample utilization and enhance sensitivity. Straightforward improvements in the size reduction and speed increase can be made using the current approach and methods. Integration of these microfabricated devices with other sample manipulations is presently being explored. ACKNOWLEDGMENT This research was sponsored by the Human Genome Program, Office of Biological and Environmental Research, U.S. Department of Energy, and by Laboratory Directed R&D at Pacific Northwest National Laboratory (PNNL). PNNL is operated for the U.S. Department of Energy by Battelle Memorial Institute under Contract DE-AC06-76RLO 1830. Helpful discussions with Drs. Naxing Xu and Julie Horner are gratefully acknowledged. Received for review December 17, 1998. Accepted February 24, 1999. AC981400W
