Applied Microbiology and Biotechnology

, Volume 64, Issue 3, pp 289–299

Micro total analysis system (μ-TAS) in biotechnology

Authors

  • S. J. Lee
    • Metabolic and Biomolecular Engineering National Research Laboratory, Department of Chemical and Biomolecular Engineering, BioProcess Engineering Research Center and Center for Ultramicrochemical Process SystemsKorea Advanced Institute of Science and Technology
    • Metabolic and Biomolecular Engineering National Research Laboratory, Department of Chemical and Biomolecular Engineering, BioProcess Engineering Research Center and Center for Ultramicrochemical Process SystemsKorea Advanced Institute of Science and Technology
    • Department of Biosystems and Bioinformatics Research CenterKorea Advanced Institute of Science and Technology
Mini-Review

DOI: 10.1007/s00253-003-1515-0

Cite this article as:
Lee, S.J. & Lee, S.Y. Appl Microbiol Biotechnol (2004) 64: 289. doi:10.1007/s00253-003-1515-0

Abstract

Nanobiotechnology raises fascinating possibilities for new analytical assays in various fields such as bioelectronic assembly, biomechanics and sampling techniques, as well as in chips or micromachined devices. Recently, nanotechnology has greatly impacted biotechnological research with its potential applications in smart devices that can operate at the level of molecular manipulation. Micro total analysis system (μ-TAS) offers the potential for highly efficient, simultaneous analysis of a large number of biologically important molecules in genomic, proteomic and metabolic studies. This review aims to describe the present state-of-the-art of microsystems for use in biotechnological research, medicine and diagnostics.

Introduction

Recent developments in nanotechnology have opened up possibilities for new, revolutionary approaches to the analysis of individual molecules on the atomic scale, and the potential for the assembly of molecular structures and devices that are a few to tens of nanometers in size. This approach makes use of various biological research areas such as molecular self-assembly, spatial positioning, microconstruction, biocomposite fabrication and nanomachines. Therefore, nanobiotechnology, a new discipline that employs biomolecules as building blocks and biomolecular self-assembly as one of the construction methods, offers exciting possibilities for the miniaturization of analytical instruments.

Microfabrication technologies (see terminology in Table 1) can be widely used in many fields, which have all contributed to the rapidly growing miniaturization of biotechnological and chemical processes (Fig. 1). Micro total analysis system (μ-TAS), sometimes called ‘lab-on-a-chip’, must perform the functions of large analytical devices in small units (Kopf-Sill 2002). When fully developed, μ-TAS will contain elements for the acquisition, pretreatment, separation, post-treatment and detection of a variety of samples. In this context, many review papers for a wide range of its applications in genomics, proteomics, clinical diagnostics, drug discovery and biosensors have appeared (Table 2).
Table 1

Terminology used in this paper

Terminology

Micro total analysis systems (μ-TAS)

Miniaturization of an overall analytical process from sample preparation through reaction and separation to detection

Microfabricated devices

Microstructures fabricated by methods basically applied in microelectronics, such as micromachining, microlithography, injection modeling and embossing

Microfluidics

Transporting and manipulating minute amounts of fluid through microchannels on the chips

Lab-on-a-chip

A microfabricated device, integrating several laboratory processes on a single chip

Fig. 1

The micro total analysis system (μ-TAS) linking of biotechnology and nanotechnology

Table 2

Reviews in the field of μ-TAS or microchip analysis

Topic

References

Current topics in μ-TAS/microchip

Merkle 1999; Laval et al. 2000; Auroux et al. 2002; Battersby and Trau 2002; Reyes et al. 2002; Haruyama 2003; Huikko et al 2003; Hong and Quake 2003; Nakamura and Karube 2003

Proteome analysis

Sanders and Manz 2000; Figeys and Pinto 2001

DNA analysis

Khandurina et al. 2000; Sanders and Manz 2000; Medintz et al. 2001; Galvin 2002

Bio/clinical/forensic assay

Guetens et al. 2000; Sundberg et al. 2000; Judos et al. 2001; Chovan and Guttman 2002; Verpoorte 2002; Bilitewski et al. 2003

Electrokinetic microchip analysis

Bruin 2000

Microchip-based separation

Regnier et al. 1999; Dolnik et al. 2000

Mass spectrometry on microchip

Oleschuk and Harrison 2000; Wang et al. 2000a

Film-based microfluidic device

Weigel et al. 2001, 2003

In this paper, we review various uses of μ-TAS devices, which include analytical technology for genomics and proteomics, integration of mass spectrometer data, and other biological applications. The most important advantage of μ-TAS is the possibility to create complete analytical microsystems by integrating various functional modules into a single device. Such functional modules may include those capable of sample preparation, integrated optics for excitation and detection of fluorescently labeled cells, cultivation of sorted cells, DNA amplification by the polymerase chain reaction (PCR), or single cell enzymatic analysis (Fig. 2). In particular, it is believed that this kind of highly efficient μ-TAS will be particularly valuable to the pharmaceutical industry for the screening of combinatorial libraries, as well as in clinical analyses, DNA-based diagnostics, and genotyping. The latest advances made through cross-fertilization of nanotechnology and biotechnology to study biological macromolecules such as DNA, proteins and other important biomacromolecules are discussed (see Table 3). Selected bioanalytical applications of micro-analytical systems, including biological reactions and assays, and analytical separation of biomolecules in micro-scale, are also described.
Fig. 2

Key technologies and units that need to be incorporated into μ-TAS

Table 3

Representative references for microchip-based analysis of DNA, protein and other biomolecules. CE capillary electrophoresis, CEC capillary electrochromatographic, MEKC micellar electrokinetic chromatography, CE-MS. capillary electrophoresis-mass spectrometry, FIA flow injection analysis

Method

Analyte

DNA

Protein

Other biomolecules

Enzyme

Peptide

Antibody

Other proteins

CE

Woolley and Mathes 1994

Cohen et al. 1999

Rocklin et al. 2000

Rodriguez et al. 1997

Colyer et al. 1997

Guttman 1996

Woolley et al. 1996, 1997

Burke and Regnier 2001

Gottschlich et al. 2001

Shimura and Kasai 1997

Liu et al. 2000

Wang et al. 2000b

Jacobson and Ramsey 1996

Xue et al. 2001

Jiang et al. 2000

Yao et al. 1999

Munro et al. 2000

Hadd et al. 1999

Bromberg and Mathies 2003

Duffy et al. 1998

 

Starkey et al. 2001

Gottschlich et al. 2000

Zugel et al. 2000

Gao et al. 2001

Burns et al. 1996, 1998

Chabinyc et al. 2001

Munro et al. 2001

Badal et al. 2002

Pumera et al. 2001

Chen et al. 2002

CEC

He et al. 1999

Slentz et al. 2001

MEKC

Heeren et al. 1996

Hutt et al. 1999

Rodriguez et al. 2000

Culbertson et al. 2000

CE-MS

Deng et al. 2001a, 2001b

Li et al. 1999

Kameoka et al. 2001

Diffusion-based

(T-sensor)

Kamholz et al. 1999

Weigel et al. 2001

μ-FIA

Hadd et al. 1997

Duffy et al. 1999

Kerby and Chien 2001

Xiong and Regnier 2001

Real time PCR

Belgrader et al. 1999

 

Lagally et al. 2000, 2001

μ-TAS for DNA analysis and genomics

Nucleic acid-based analysis in μ-TAS has advanced rapidly since the completion of the human genome project. So far, the labor- and material-intensive nature of genetic typing and other molecular analysis has restrained their practical application. A variety of genotyping platforms of μ-TAS are currently being developed to overcome these latter problems (Wilding et al.1998; Fan et al. 1999; Lowe 2000; Huang et al. 2002). Some of the enabling technologies made possible by microelectromechanical systems (MEMS) include PCR microsystems for DNA amplification, and the identification and analysis of single nucleotide polymorphisms (SNPs) to facilitate improved understanding and diagnosis of various genetic diseases. μ-TAS devices facilitate nucleic acid-based analysis through very rapid fragment separation using capillary electrophoresis (CE) and high-performance liquid chromatography (HPLC), together with mixing and transport of reagents and biomolecules in an integrated system.

Microfabricated electrophoresis chips for high-speed DNA genotyping

The development of DNA analysis systems and methods with enhanced speed, sensitivity, and high-throughput is vital to the efficient utilization of the sequence information generated by the human genome project and other genome sequencing efforts. Electrophoresis is a ubiquitous technique in molecular biological and medical analyses. In recent years, its implementation on a smaller scale to reduce the amount of sample required for the analysis has been the focus of research. Functional models of integrating miniaturized separation units into compact devices using high-speed CE have been developed, and their feasibility has been demonstrated by several groups (Jacobson et al. 1994; Woolley and Mathes 1994; Woolley et al. 1997; Munro et al. 2001). One of the first groups to realize on-chip microfluidic CE used a glass device with electroosmotic flow (EOF)—the movement of bulk solution in a capillary due to an applied electric field (Manz et al. 1992; Harrison et al. 1993). Higher efficiency, faster response time and reduced reagent consumption were successfully demonstrated for these devices. After the early demonstrations of the usefulness of electrophoresis microchips (and their further development) by Manz’s group, CE channels on planar glass substrates were fabricated by Woolley and Mathes (1994) using photolithography and chemical etching; these CE chips allowed sequencing of 150 bp DNA with 97% accuracy in less than 10 min. Other glass microchips were subsequently constructed with cross-column geometry, nondiscriminatory injection procedures and fluorescence detection schemes. These prototypes showed that microchip devices provide very fast response time for sensor applications (Jacobson et al. 1994). Other groups applied microfabricated device technology to the analysis of DNA restriction fragments, as first demonstrated by Jacobson and Ramsey (1996), who created a promising device capable of restriction digesting a DNA sample in a 0.7 nl reaction chamber for 10 s and resolving the DNA fragments on a 1% (w/v) hydroxyethyl cellulose gel within 3 min. Chip performance and DNA separation were monitored by laser-induced fluorescence (LIF) and, therefore, the whole assay was completed in less than 4 min. Burns et al. (1996, 1998) reported the development of a set of micromachined components for DNA analysis in which the integration of pumps and other components into a self-contained miniaturized device provided significant improvement in the speed, portability and cost of DNA analysis. Munro et al. (2001) investigated a number of silica coating methods to find those that improved DNA analysis in chip-based CE. Several silica-coatings were evaluated for their ability to provide an environment for optimal separation conditions resulting in neat PCR product separations, and verified for a variety of clinical PCR samples. More details of μ-TAS developed for DNA sequencing and separation can be found in recent reviews (Auroux et al. 2002; Khandurina and Guttman 2002).

Microfluidic chips for genomics

DNA analysis coupled with in situ PCR in microfluidic chip involves complex processes such as handling of liquid solutions, measurement of solution volumes, mixing of reagents, DNA template and primers, controlled thermal reactions, loading of the sample to an electrophoresis gel, and final detection of DNA products. The high surface-to-volume ratio of microfluidic devices facilitates rapid heat transfer, and smaller sample volumes possess less heat capacity. Therefore, rapid temperature changes and quick thermal cycling are possible.

There are a number of microsystems currently being developed to facilitate genomic analysis. A number of groups have investigated chip-based PCR systems, in which micromachined heating chambers of various configurations have been incorporated (Burns et al. 1996; Chen et al.1996; Shoffner et al. 1996; Taylor et al. 1997). The most notable advance towards fulfilling the concept of μ-TAS in this area was made by the integration of the PCR modules into a single functional unit capable of amplifying and CE-sizing the β-globin PCR product within 20 min (Woolley et al. 1996). Additionally, this system allowed identification of Salmonella genomic DNA in less than 45 min, showing its potential for future clinical applications. The micromachined chemical amplifier was further developed to perform high-speed PCR through a process analogous to an electronic amplifier, in which the sample moves continuously through the regions of various temperatures, resulting in successful amplification of DNA in 1.5 to 20 min (Kopp et al. 1998). Next, an integrated system for PCR analysis was developed that combined dual Peltier thermoelectric elements with electrophoretic sizing and detection on a microchip (Khandurina et al. 2000). This method enabled reduction of the PCR cycle number and increase of the speed, resulting in a total analysis time of less than 20 min.

These systems were further improved by the development of an integrated microfluidic device that combined stochastic PCR amplification of a single DNA template molecule with CE analysis of the products (Lagally et al. 2000, 2001). This integrated monolithic system incorporated several 280 nl PCR chambers etched on a glass structure, connected to microfluidic valves and hydrophobic vents for sample introduction and immobilization during thermal cycling. The low volume reactor and the use of thin film heaters permitted thermal cycle times as short as 30 s. Importantly, a histogram of the normalized peak areas from repetitive PCR analyses enabled the quantitation of single viable template molecule copies in the reactor. Following this work, a fully automated and integrated microfluidic system was developed that performed nanoliter PCR analysis on direct cheek cell samples. This system included all the steps needed for DNA analysis: injection, mixing, lysis, PCR, separation, sizing, and detection (He et al. 2001).

In a continued effort to shorten reaction times, Landers and co-workers developed chips based on IR radiation-mediated heating and compressed air cooling for DNA amplification in the capillaries of on-chip microchambers (Oda et al. 1998; Hühmer and Landers 2000; Giordano et al. 2001a). This resulted in thermal cycling times of as short as 17 s per cycle. A tungsten lamp, thermocouple feedback and computer interface were used to accurately control the temperature through a combination of light intensity and airflow.

Following this, a polymeric microchip with a 1.7 μl chamber containing a thermocoupler was used to successfully amplify a 500-bp fragment of lambda phage DNA (Giordano et al. 2001b) in 15 cycles, in a total amplification time of 240 s. This system provides a unique approach for PCR amplification in microfluidic chips.

DNA hybridization assays in microfluidic chips

Almost all DNA-based assays involve hybridization procedures, which typically take 3–18 h, and thus are a rate-limiting step in the assays. Therefore, reducing the time required to carry out hybridization has been the objective of several studies. Fan et al. (1999) described dynamic hybridization of DNA in a microfluidic chip using paramagnetic beads for the analysis of gene expression. In this work, different DNA fragments were attached to poly(T) chain (oligo-dT25) modified paramagnetic beads, and assayed for their reactivity with a fluorescent DNA probe. Hybridization to dense arrays of immobilized oligomers took place very rapidly, in just a few seconds. It was thought that the rapid delivery of probe to the beads facilitated the interaction between probe and sample, leading to the increased hybridization rate. The small sample volume also means that a higher concentration of probe can be used without raising concerns over cost or supply, and thus hybridization can be driven much more rapidly to completion. After hybridization, denaturation by on-chip heaters is carried out to remove the bound DNA samples to be interrogated again by another probe having a different sequence of interest. Therefore, this strategy seems to be promising for both qualitative and quantitative screening of DNA samples.

Anderson et al. (2000) presented another sophisticated microfluidic device, a polycarbonate-based system developed for automated multistep genetic assays. In this system, more than 10 reagents could be manipulated in over 60 sequential reactions. This system was used for the detection of mutations in a 1.6 kb region of the HIV genome from serum samples. Extraction and concentration of nucleic acids from milliliter aqueous samples, microliter chemical amplification, serial enzymatic reactions, metering, mixing and final nucleic acid hybridization could all be performed in the device. This device was fabricated using plastic rather than glass or silicon for reasons of cost, since a disposable device is required for most clinical applications. This example is truly impressive, and the principles underlying its development are sure to be useful in developing devices for many different applications in the future. In a recent study by Shamansky et al. (2001), a microchannel flow-based sensor was developed using a DNA photopatterning method for use as a DNA sensor. Different DNA probes could be injected through each channel for capture by incorporating multi channels. This system could be applied as a prototype microfluidic sensor device to separate and identify synthetic fluorescently labeled DNA targets.

Following this work, Lee et al. (2001) fabricated one dimensional (1D) single-stranded DNA (ssDNA) line arrays and two dimensional (2D) DNA hybridization arrays, in which polydimethylsiloxane (PDMS) microchannels were placed perpendicular to a 1D line array to transport target oligonucleotide solutions. The oligonucleotide hybridization of DNA and RNA was visualized using a surface plasmon resonance (SPR) imaging method.

μ-TAS for protein analysis and proteomics

The field of proteomics, which aims to understand cellular functions and regulation of proteins, is growing rapidly. However, implementation of proteomics towards high-throughput functional genomics platforms is limited by its current macro-technical capabilities, prompting researchers to investigate newer microsystems for proteomic research. As with protein microarray systems (Zhu and Snyder 2003), the technology for chip-based proteomic analysis is still in its infancy. In the near future, however, microfluidic systems will likely be used more intensively than microarray-based systems in proteomic analysis. Proteomics will benefit more than genomics by miniaturization. This is mainly because there is no method to amplify proteins, and thus proteins are more difficult to obtain in suitable amounts compared with DNA, which can be amplified by PCR. As the amount of protein samples available is often very limited, it is evident that the new technology employed must be sensitive and allow handling of small sample volumes without any loss of analyte in the sample processing steps prior to analysis. Microfabrication technology has the potential to improve the development, automation, cost-effectiveness, and throughput of proteome analysis. Most of the microfluidic applications in proteomics have so far been focused on chip-based separation of proteins in conjunction with mass spectrometry (MS) (Ekstrom et al. 2000; Miliotis et al. 2000). Conventional proteomics processes include the extraction of proteins from cells, separation and visualization of proteins by 1D- or 2D-gel electrophoresis, excision of bands or in-gel digestion of the proteins, followed by MS analysis of the resulting peptide mixtures. This procedure is rather slow and labor intensive, creating a great demand for the development of fast, high-throughput integrated multisample analysis systems. Microfluidic chips have the potential to control and automate a series of sample processing steps, including protein purification, separation, digestion, and identification by MS.

Protein separation in microchips

Chromatography and electrophoresis represent the two major separation techniques for proteins. The latter has been by far more popular in microchip applications, due to its easier realization in miniaturized formats. Numerous examples of microchip electrophoretic separations of amino acids, peptides and proteins have been demonstrated and reviewed (Li et al. 1999; Bruin 2000; Dolnik et al. 2000; Munro et al. 2000; Badal et al. 2002).

Capillary sodium dodecyl sulfate (SDS) electrophoretic separation of six proteins, ranging from 9 to 116 kDa, was successfully performed on a microchip in 35 s, which is much faster than the fastest separation in CE reported so far (Yao et al. 1999). Following this, a protein-sizing assay that incorporated SDS-PAGE gels with high-speed microchip separation was presented by Bousse et al. (2001). In this device, all protein separation, staining, destaining and detection steps were performed within 30 min. Thanks to the integration of all the required assay steps, this represents a much greater increase in speed than that obtained in other assay methods such as electrophoretic separations.

Microchip-based CE of proteins often employs UV absorbance as a detection method. However, in the microchip environment, extremely short path lengths result in low absorbance values and, therefore, the more sensitive laser induced fluorescence (LIF) detection is commonly used. Postcolumn labeling of proteins is a desirable strategy for protein detection in CE to avoid problems with labeling of multiple sites. The applicability of this technique to microscale protein separation has been investigated by several research groups (Colyer et al. 1997; Csapo et al. 2000; Liu et al. 2000), who concluded that fast and accurate microseparation, post-column labeling, mixing, and destaining were clearly beneficial in combination with the non-covalent protein labeling method. Using this method, a high separation efficiency and a good detection limit (as little as a picogram) were demonstrated (Csapo et al. 2000).

The 2D separation system is of great interest because it shows a multiplicative increase in the peak capacity over the 1D method. Microfluidic devices are potentially useful for multidimensional separation because high-efficiency separation can be achieved and small sample volume can be manipulated with minimal dead volume within the interconnecting channels. An efficient 2D separation was demonstrated on a microfabricated device using open channel electrochromatography (OCEC) as the first dimension and CE as the second dimension (Gottschlich et al. 2001). A 25 cm separation channel was modified with octadecyl groups and coupled to a 1.2 cm straight separation channel for CE. Higher sampling rates of the effluent from the first dimension into the second dimension further allowed efficient CE analysis of the OCEC separation products. Using this system, fluorescently labeled products from tryptic digests of β-casein were successfully analyzed in 13 min.

Recently, other 2D microfabricated devices have been introduced (Rocklin et al. 2000; Chen et al. 2002). In the device fabricated by Rocklin et al. (2000), combined micellar electrokinetic chromatography (MEKC) and CE were incorporated for the analysis of various peptide mixtures, in which 2D analysis was carried out every few seconds, with a total analysis time of less than 10 min for trypsin-digested peptides. The resolution achieved with 2D separation has been demonstrated to be much better than each dimension alone. This system seems to be useful for rapid automated fingerprinting of proteins and peptides with possible coupling to mass spectrometry (MS) detection.

Microfluidic system as an interface to MS

MS is a vital tool in the analysis of proteins and peptides and is regularly employed to analyze proteins that have been separated by various methods. MS performs further separation, detection, and identification of a broad range of compounds. Widespread application of microfabricated devices coupled to MS requires the incorporation of a microfabricated electrospray ionization (ESI) source, and indeed considerable success has recently been achieved in this area (Bruin 2000; Dolnik et al. 2000). These devices are also equipped with the process of sample introduction for protein identification, as they can transport the sample fluid electrokinetically or by pressure, and generate an electrospray via an attached capillary of more complex emitter couplings. Proteins have been identified real time by peptide mass fingerprinting with high detection sensitivity (0.1–2 pmol/μl sample loaded) using a hybrid microchip nanoelectrospray device (Mazar et al. 2001). Automated MS analysis in μ-TAS was demonstrated using a porous microfabricated digestion chip integrated with a sample pretreatment robot and microdispenser for transferring the reaction products to a matrix-assisted laser desorption/ionization (MALDI) target plate (Ekstrom et al. 2000). This type of system shows potential in the enrichment of peptides and proteins from 2D electrophoresis or in the identification of proteins through peptide mass fingerprinting. A new design for high-throughput microfabricated CE-ESI-MS analysis with automated sampling from a microwell plate was reported by Zhang and co-workers (1999, 2001), in which a sample loading port, separation channel, and a liquid junction for coupling the device to the MS system were combined with a miniaturized sub-atmospheric electrospray interface. A computer-activated electro-pneumatic distributor was used for sample loading and channel washing. Fully automated CE-ESI-MS analysis of protein digests was successfully demonstrated using this system.

Microfluidic systems for immunoassay and other applications

Immunoassays are important tools in clinical diagnostics, biomedical science and industry, representing one of the simplest methods for the qualitative and quantitative analysis of proteins. Various miniaturized immunoassays, mostly based on competitive antigen-antibody interactions, have been successfully performed with electrophoresis microchips, which are fast and efficient, and are likely to be feasible for use in portal-based automated analysis (Sanders and Manz 2000; Figeys and Pinto 2001). In one such system, immunoreagents were immobilized directly onto the channel walls of a microchip, and the feasibility of the system was successfully tested using protein A as the immobilized protein and fluorescently labeled rabbit immunoglobulin G as a sample (Dodge et al. 2001). This approach showed great potential for many high-throughput screening applications. In addition, a new class of microfluidic immunoassay system was developed based on solid supported lipid bilayers (Yang et al. 2001). The bilayers, created on the polydimethylsiloxane (PDMS) surfaces of an array of parallel microchannels, contained dinitrophenyl (DNP)-conjugated lipid for binding with fluorescently labeled anti-DNP antibodies present at different concentrations in each channel. This methodology was used for rapid and accurate heterologous assays in a single experiment, and the amount of proteins required was significantly reduced compared to that needed in conventional methods.

Nanovials are another example of the use of microfabricated devices for protein analysis (Litborn et al. 2000). These very small wells featuring high surface-to-volume ratio are fabricated in different substrates, and showed increased reaction efficiency compared to conventional microcentrifuge vials. Cohen et al. (1999) reported a microchip-based enzyme assay for protein kinase based on the fact that the substrates and products had substantially different electrophoretic mobilities, which allowed their efficient separation by microchip CE. Microchip electrophoresis has also been adopted for use in other immunoassay protocols.

A microchip immunoassay system for serum cortisol was developed in which the separation and measurement on the microchip were completed in 30 s (Koutny et al. 1996). The major advantages of microchip-based immunoassays are lower analysis times, lower amounts of sample, and good portability.

μ-TAS for other biologically important molecules

Compared to the large amount of work focused on the miniaturization of nucleic acid and protein assays, the application of microfluidic devices to the analysis of other biologically important macromolecules such as carbohydrates, lipids and fatty acids has so far been sparse, probably due to the limited development of CE techniques for these analytes. Analysis of lipids and fatty acids is complicated due to their poor solubility in aqueous buffers and the requirement of micellar or non-aqueous organic solvent systems. On the other hand, carbohydrate analysis by CE is complicated because of the wide diversity of possible oligosaccharide structures (Guttman 1996), and the necessity of derivatization to introduce charges and UV or fluorescent labels into the neutral sugar molecules. Application of μ-TAS for metabolite analysis has also been rare, but some progress has been made recently. Wang et al. (2000b) performed successful simultaneous microchip-based bioassay of a mixture of metabolites including glucose, uric acid, ascorbic acid and acetaminophen. Enzymatic oxidation of glucose was carried out in a microchannel, followed by the separation and electrochemical detection of the neutral peroxides and the anionic uric and ascorbic acids. The chip-based protocol allowed integration of miniaturized bioassays, separation and amperometric detection in a single device, and was comparable to bench-top analyzers and conventional biosensors in terms of performance, speed, sample volume and size. Despite the lack of more examples of microchip applications in the analysis of these biomolecules, it is reasonable to anticipate rapid growth of this field in the near future.

Micro flow cytometry

The integration of various analysis systems within a chip is one of the greatest potentials of microfabricated devices. Flow cytometry techniques, such as those used in fluorescence-activated cell sorting (FACS), are based on focusing cells into a single line for individual analysis. This method is widely used in clinical medicine, basic biological research and even in material science. Microfabricated FACS (μ-FACS) devices offer a number of advantages over conventional FACS systems. A conventional FACS system normally relies on the sorting of droplets in an open system. In contrast, μ-FACS devices operate in a closed system, thus reducing the risk when working with infectious cells and biohazardous specimens. Several groups have developed microfabricated devices for cell analysis by flow cytometry (Schrum et al. 1999) or for cell sorting (Blankenstein and Larsen 1998; Fiedler et al. 1998; Fu et al. 1999a, 1999b). More recently, Wolff et al. (2003) demonstrated μ-FACS with several novel integrated functional structures. In this system, they integrated a chamber on the chip for holding and culturing the sorted cells, and optics for cell detection. The risk of losing cells during cell handling processes could be excluded. This approach will contribute to the ongoing challenges for developing a fully integrated micro cell-sorting and analyzing system.

Integrated microdevices and other novel applications

Miniaturization of conventional bioanalytical methods and integration of all processes in microchips have been one of the central trends driving the development of the μ-TAS concept. Researchers have therefore focused on exploring ways of building integrated devices that combine separations with pre-and post-processing steps such as PCR-based DNA amplification, restriction digestion, other enzymatic assays, analyte pre-concentration, filtering, dialysis, as well as various post-derivatization and detection methods. Although, as described above, some of these complex devices have recently been developed, the ultimate integration of all of these steps into a single microchip has not been realized to date.

Toward this goal, two integrated microchips are worth mentioning. The first was developed for clinical applications by integrating two key steps, cell isolation and PCR, in a basic clinical analytical procedure. White blood cells were isolated from whole blood on 3.5 μm feature-sized wier-type filters, followed by direct PCR amplification of isolated genomic DNA (Wilding et al. 1998). The second is an integrated DNA analysis device based on a silicon chip containing a sample loading stage with a drop-volume-metering feature, a mixer, a thermal reactor for amplification, a gel loading stage, and a gel electrophoresis channel with an integrated optical detection system (Burns et al. 1996, 1998). Fluid handling was achieved by a set of hydrophobic vents and external air pressure, which drove single droplets precisely through the channel system. Although they have not yet been used for actual clinical applications, this type of integration provides an idea of what the future holds for clinical diagnostics using μ-TAS.

Conclusions and future directions

μ-TAS is advancing rapidly via the fusion of nanotechnology with biotechnology. We believe that μ-TAS devices are beginning to make significant contributions to biotechnology, biomedicine and drug discovery. The first generation microfluidic-based analytical devices were already on the market in 1999, with many more in active development. However, there are a number of issues requiring careful consideration when developing microsystems: the difficulty of surface engineering and exact control of fluids in microchannels, detection limits, sample preparation, increased integration, and the question of whether chips are to be reused. These issues need to be properly addressed for μ-TAS devices to be used in a wide range of applications. With all these efforts, the promises of μ-TAS—integration of all laboratory equipment functions on a single chip, the commercialization of truly hand-held, easy-to-use, disposable, analytical μ-TAS instruments—are expected to be realized in the near future.

Acknowledgements

This review was made possible by the support from the Center for Ultramicrochemical Process Systems sponsored by KOSEF, National Research Laboratory Program (2000-N-NL-01-C-237) of the Ministry of Science and Technology, the Advanced Backbone IT Technology Development Project (IMT2000-B3–2) of the Ministry of Information and Communication (MIC), and BK21 project.

Copyright information

© Springer-Verlag 2004